JP4369809B2 - Image processing apparatus, image processing method, and computer program - Google Patents

Description

  The present invention relates to an image processing apparatus, an image processing method, a computer program, and a storage medium for storing a computer program that execute processing for an output prohibition determination pattern included in image data of a document image.

  In recent years, due to improvements in image processing technology and image forming technology, when copying banknotes and securities using a digital color copying machine, it is possible to make faithful copies that cannot be easily distinguished from the original copy. It has become. For this reason, it is necessary to take measures so that special manuscripts such as banknotes and securities cannot be copied at all or cannot be copied correctly.

  For example, in a company, even in the case of general documents other than special manuscripts such as banknotes and securities, there are many confidential documents that are prohibited from being output such as copying from the viewpoint of maintaining confidentiality of the document contents. It is necessary to take measures to prevent such confidential documents from being copied at all, or from being copied correctly.

  For this reason, various inventions have been made in the past that aim to exert restrictions on copying special manuscripts and confidential documents. Hereinafter, specific examples of such invention will be introduced.

  First, as a method for discriminating special manuscripts such as banknotes and securities, the input image data is compared with a specific mark (pattern data) registered in advance by a pattern matching method, and a specific mark exists. A method for discriminating that a document is a special document has been proposed. The inventions described in Patent Document 1 and Patent Document 2 are typical examples. Thus, when it is determined that the document is a special document, it is easy to prohibit the copying, that is, not to copy.

  Next, as an invention for discriminating a confidential document prohibited from being copied, for example, Patent Document 3 and Patent Document 4 describe an invention in which a mark indicating a confidential document attached to a confidential document is detected. ing. This utilizes the fact that a copy-prohibited confidential document is generally stamped with a mark indicating that it is a confidential document consisting of a confidential seal, a copy-prohibited mark, and the like. Thus, when it is determined that the original is a confidential document, it is easy to prohibit copying, that is, not to copy.

  Further, Patent Documents 5 and 6 propose a technique for suppressing copying by embedding a background pattern in an original image that is desired to be prohibited from copying. This is to use a sheet on which a background pattern having a base area and a message area is created on the background of the sheet used for the original image. The copy-forgery-inhibited pattern does not stand out so much in the original image, and does not hinder the interpretation of information included in the original image. However, when a manuscript image in which such a tint block pattern is embedded is copied, for example, a message area pattern emerges. Therefore, by giving characters such as “copy prohibited” as the pattern of the message area, it becomes obvious at a glance that the copy was a confidential document prohibited from copying. Can be generated.

Japanese Patent Laid-Open No. 6-125259 JP 2001-86330 A JP-A-7-36317 JP 7-87309 A Japanese Patent Laid-Open No. 9-164739 JP 2001-197297 A

[First issue]
In order to determine a copy-prohibited document such as a confidential document, Patent Documents 1 and 2 compare input image data with a specific mark (pattern data) registered in advance by a pattern matching method. 4 detects a mark indicating a confidential document such as a secret mark. In this case, image data obtained by an image input means capable of selecting a plurality of types of reading operations, for example, image data of a document image read by a scanner that performs a document movement type reading operation and a scanner that performs a sensor movement type reading operation When the image data of the original image read in step 1 is compared, a difference in detection accuracy such as mark detection is expected.

[Second problem]
As described above, the input image data is compared with a specific mark (pattern data) registered in advance by the pattern matching method, and if it matches, special copying such as banknotes and securities is prohibited. Can be prevented (Patent Documents 1 and 2). However, if copying is prohibited by such a method, all pattern data of the original to be prohibited from copying must be registered in advance. Therefore, even if it is effective for special manuscripts such as banknotes and securities, it is extremely difficult to apply it to an unspecified number of general manuscripts such as confidential documents.

  Further, as described above, copying of a confidential document can be prevented by prohibiting copying when a mark indicating a confidential document such as a confidential mark is detected (Patent Documents 3 and 4). ). However, in this case, when an image reading operation is executed with a mark portion indicating a confidential document such as a confidential mark covered with paper or the like, it is a confidential document as a matter of course. Therefore, there is a problem that it is not possible to determine the presence of a mark indicating that the copying cannot be prevented even if the confidential document should be prohibited from being copied.

  Further, as described above, for example, a confidential document is generated as a manuscript image in which a background pattern is embedded in the background, and when such a confidential document is copied, a part of the background pattern emerges. Psychologically, it is possible to generate a regulatory power against copying (Patent Documents 5 and 6). However, such a method using a tint block pattern only provides a copy suppression effect, and does not restrict the copying act itself. For this reason, there is a problem that a confidential document can be copied for a person who does not mind that the tint block pattern is raised.

  An object of the present invention is to make it possible to extract information in an image with high accuracy even in image data obtained by an image input means capable of selecting a plurality of types of reading operations.

  An object of the present invention is to make it possible to exert a stronger deterrent against the act of reproducing a copy-inhibited document such as a confidential document.

  The image processing apparatus according to claim 1 is an image input unit capable of selecting a plurality of types of image reading operations, and one or more pre-processing is selectively performed on image data obtained by the image input unit. Image pre-processing means to be executed, information extraction means for extracting specific information contained in the image data from the image data obtained by the image input means, and image reading operation selected by the image input means Preprocessing selection means for selecting a preprocessing according to the image processing and causing the image preprocessing means to execute the selected preprocessing.

  According to a second aspect of the present invention, in the image processing apparatus according to the first aspect, the information extracting means extracts a characteristic included in the image data as specific information included in the image data.

  According to a third aspect of the present invention, in the image processing apparatus according to the second aspect, the feature extracted by the information extracting means is indicated by an array of dots.

  According to a fourth aspect of the present invention, in the image processing apparatus according to the third aspect, the dots constituting the feature extracted by the information extracting means are expressed by a single color material.

  According to a fifth aspect of the present invention, in the image processing apparatus according to the third aspect, the dots constituting the feature extracted by the information extracting means are superimposed on the background of the original image from which the image data is based.

  According to a sixth aspect of the present invention, in the image processing apparatus according to the third aspect, the dots constituting the feature extracted by the information extracting means are superimposed on the front surface of the original image from which the image data is based.

  According to a seventh aspect of the present invention, in the image processing apparatus according to the third aspect, the arrangement of dots constituting the feature extracted by the information extracting means is a plurality of types superimposed on the original image from which the image data is based. It is one kind of dot pattern among the dot patterns.

  According to an eighth aspect of the present invention, in the image processing apparatus according to the seventh aspect, as a plurality of types of dot patterns, a dot pattern composed of dots of a size that is easy to reproduce and reproduce and a dot of a size that is difficult to reproduce and reproduce. Dot pattern.

  According to a ninth aspect of the present invention, in the image processing apparatus according to any one of the first to eighth aspects, the plurality of types of image reading operations by the image input means are configured to obtain image data by moving the sensor relative to the document. And an operation of obtaining image data by moving the document relative to the sensor.

  According to a tenth aspect of the present invention, in the image processing apparatus according to any one of the first to ninth aspects, the preprocessing selecting unit performs a noise removal process when a certain type of image reading operation is selected by the image input unit. I do.

  According to an eleventh aspect of the present invention, in the image processing apparatus according to any one of the first to ninth aspects, the preprocessing selecting means is the first when the image reading operation of a certain type is selected by the image input means. Density correction is performed, and when another type of image reading operation is selected by the image input unit, second density correction different from the first density correction is performed.

  The invention of the image processing apparatus according to claim 12 is an image input means capable of selecting a plurality of types of image reading operations, and specific information contained in the image data for the image data obtained by the image input means. Information extracting means for extracting, and the information extracting means executes an information extracting process corresponding to the image reading operation selected by the image input means.

  According to a thirteenth aspect of the present invention, in the image processing apparatus according to the twelfth aspect, the information extraction unit extracts a characteristic included in the image data as specific information included in the image data.

  According to a fourteenth aspect of the present invention, in the image processing apparatus according to the thirteenth aspect, the feature extracted by the information extracting means is indicated by an array of dots.

  According to a fifteenth aspect of the present invention, in the image processing apparatus according to the fourteenth aspect, the dots constituting the feature extracted by the information extracting means are expressed by a single color material.

  According to a sixteenth aspect of the present invention, in the image processing apparatus according to the fourteenth aspect, the dots constituting the feature extracted by the information extracting means are superimposed on the background of the original image from which the image data is based.

  According to a seventeenth aspect of the present invention, in the image processing apparatus according to the fourteenth aspect, the dots constituting the feature extracted by the information extracting means are superimposed on the front surface of the original image from which the image data is based.

  According to an eighteenth aspect of the present invention, in the image processing apparatus according to the fourteenth aspect, the arrangement of dots constituting the feature extracted by the information extraction means is a plurality of types superimposed on the original image from which the image data is based. It is one kind of dot pattern among the dot patterns.

  According to a nineteenth aspect of the present invention, in the image processing apparatus according to the eighteenth aspect, as a plurality of types of dot patterns, a dot pattern composed of dots of a size that is easy to reproduce and reproduce and a dot of a size that is difficult to reproduce and reproduce. Dot pattern.

  According to a twentieth aspect of the present invention, in the image processing apparatus according to any one of the twelfth to nineteenth aspects, the image reading operation by the image input means is performed by moving the sensor with respect to the document. And an operation of obtaining image data by moving the document relative to the sensor.

  According to a twenty-first aspect of the present invention, in the image processing apparatus according to any one of the twelfth to twentieth aspects, the information extracting unit is included in the image data in accordance with an image reading operation selected by the image input unit. Change the threshold for extracting specific information.

  According to a twenty-second aspect of the present invention, in the image processing apparatus according to the twenty-first aspect, the threshold for extracting the specific information is a threshold for determining the presence or absence of a dot detection signal.

  The invention according to claim 23 is the image processing apparatus according to any one of claims 1 to 22, wherein image output means for performing output processing necessary for image data obtained by the image input means and outputting the image data; Output mode changing means for changing the mode of image output by the image output means in accordance with specific information obtained by the image extracting means.

  According to a twenty-fourth aspect of the present invention, in the image processing apparatus according to the twenty-third aspect, the output mode changing means expresses the prohibition of faithful reproduction of the specific information obtained by the image extracting means. The output image is processed into an image that cannot be used as a copy.

  According to a twenty-fifth aspect of the present invention, in the image processing apparatus according to the twenty-third aspect, in the case where the output mode changing means expresses the prohibition of faithful reproduction of the specific information obtained by the image extracting means. Process the output image into an unreadable image.

  According to a twenty-sixth aspect of the present invention, in the image processing apparatus according to the twenty-third aspect, the output mode changing unit is configured to output the image when the specific information obtained by the image extracting unit expresses prohibition of copying. The paper output process by the output means is stopped.

  According to a twenty-seventh aspect of the present invention, in the image processing apparatus according to the twenty-third aspect, the output mode changing means performs copying when the specific information obtained by the image extracting means expresses prohibition of copying. Notify the outside that the image of the prohibited document has been scanned.

  The invention of an image processing method according to claim 28, wherein an image input process capable of selecting a plurality of types of image reading operations and one or more pre-processes are selected for the image data obtained in the image input process. Image pre-processing step to be executed, information extraction step for extracting specific information contained in the image data from the image data obtained in the image input step, and image reading selected by the image input step A preprocessing selection step of selecting a preprocessing according to the operation and causing the selected preprocessing to be executed in the image preprocessing step.

  According to a twenty-ninth aspect of the present invention, in the image processing method according to the twenty-eighth aspect, in the information extracting step, a characteristic included in the image data is extracted as specific information included in the image data.

  According to a thirty-third aspect of the present invention, in the image processing method according to the twenty-ninth aspect, the feature extracted in the information extracting step is indicated by an array of dots.

  According to a thirty-first aspect of the invention, in the image processing method according to the thirty-third aspect, the dots constituting the feature extracted in the information extracting step are expressed by a single color material.

  According to a thirty-second aspect of the present invention, in the image processing method according to the thirty-third aspect, the dots constituting the feature extracted in the information extracting step are superimposed on the background of the original image that is the source of the image data.

  According to a thirty-third aspect of the present invention, in the image processing method according to the thirty-third aspect, the dots constituting the feature extracted in the information extracting step are superimposed on the front surface of the original image from which the image data is based.

  According to a thirty-fourth aspect of the present invention, in the image processing method according to the thirty-third aspect, the arrangement of dots constituting the feature extracted in the information extracting step is a plurality of types superimposed on the original image from which the image data is based. It is one kind of dot pattern among the dot patterns.

  According to a thirty-fifth aspect of the present invention, in the image processing method according to the thirty-fourth aspect, as a plurality of types of dot patterns, a dot pattern composed of dots of a size that is easy to reproduce and reproduce and a dot of a size that is difficult to reproduce and reproduce. Dot pattern.

  A thirty-sixth aspect of the invention is the image processing method according to any one of the twenty-eighth to thirty-fifth aspects, wherein a plurality of types of image reading operations by the image input step are performed by moving the sensor with respect to the document. And an operation of obtaining image data by moving the document relative to the sensor.

  A thirty-seventh aspect of the present invention is the image processing method according to any one of the twenty-eighth to thirty-sixth aspects, wherein in the preprocessing selection step, a noise removal process is performed when a certain type of image reading operation is selected in the image input step. I do.

  According to a thirty-eighth aspect of the invention, in the image processing method according to any one of the twenty-eighth to thirty-sixth aspects, in the preprocessing selection step, a first type of image reading operation is selected when the image input operation is selected in the image input step. Density correction is performed, and when another type of image reading operation is selected in the image input process, second density correction different from the first density correction is performed.

  The image processing method according to claim 39 is an image input step capable of selecting a plurality of types of image reading operations, and specific information included in the image data for the image data obtained in the image input step. And an information extraction process for extracting the information in accordance with the image reading operation selected in the image input process.

  According to a 40th aspect of the present invention, in the image processing method according to the 39th aspect, in the information extracting step, as a specific information included in the image data, a characteristic included in the image data is extracted.

  According to a forty-first aspect of the present invention, in the image processing method according to the forty-fourth aspect, the feature extracted in the information extracting step is indicated by an array of dots.

  According to a forty-second aspect of the present invention, in the image processing method according to the forty-first aspect of the present invention, dots constituting the feature extracted in the information extracting step are expressed by a single color material.

  The invention according to claim 43 is the image processing method according to claim 41, wherein the dots constituting the feature extracted in the information extracting step are superimposed on the background of the original image from which the image data is based.

  According to a 44th aspect of the present invention, in the image processing method according to the 41st aspect, the dots constituting the feature extracted in the information extracting step are superimposed on the front surface of the original image from which the image data is based.

  According to a 45th aspect of the present invention, in the image processing method according to the 41st aspect, the arrangement of dots constituting the feature extracted in the information extracting step is a plurality of types superimposed on the original image from which the image data is based. It is one kind of dot pattern among the dot patterns.

  The invention according to claim 46 is the image processing method according to claim 45, wherein a plurality of types of dot patterns are composed of a dot pattern composed of dots of a size that is easy to reproduce and reproduce and a dot of a size that is difficult to reproduce and reproduce. Dot pattern.

  According to a 47th aspect of the present invention, in the image processing method according to any one of the 39th to 46th aspects, the plurality of types of image reading operations in the image input step are configured to obtain image data by moving a sensor relative to a document. And an operation of obtaining image data by moving the document relative to the sensor.

  According to a 48th aspect of the present invention, in the image processing method according to any one of the 39th to 46th aspects, the information extraction step is included in the image data in accordance with the image reading operation selected in the image input step. Change the threshold for extracting specific information.

  According to a 49th aspect of the present invention, in the image processing method according to the 47th aspect, the threshold for extracting the specific information is a threshold for determining the presence or absence of a dot detection signal.

  The invention according to claim 50 is the image processing method according to any one of claims 28 to 49, wherein an image output process for outputting the image data obtained by the image input means by performing necessary image processing; An output mode changing step of changing the mode of image output by the image output step according to the specific information obtained by the image extraction step.

  The invention according to claim 51 is the image processing method according to claim 50, wherein, in the output mode change step, the specific information obtained by the image extraction step expresses the prohibition of faithful reproduction. The output image is processed into an image that cannot be used as a copy.

  The invention according to claim 52 is the image processing method according to claim 50, wherein, in the output mode change step, the specific information obtained by the image extraction step represents a prohibition of faithful reproduction. Process the output image into an unreadable image.

  The invention according to claim 53 is the image processing method according to claim 50, wherein, in the output mode changing step, the specific information obtained in the image extracting step expresses prohibition of copying. The paper output process by the output process is stopped.

  According to a 54th aspect of the present invention, in the image processing method according to the 50th aspect, in the output mode changing step, copying is performed when the specific information obtained in the image extracting step expresses prohibition of copying. Notify the outside that the image of the prohibited document has been scanned.

  The invention according to claim 55 is installed in a computer provided in an image processing apparatus having an image input function capable of selecting a plurality of types of image reading operations, and the image data obtained by the image input function is installed in the computer. An image preprocessing function for selectively executing one or more preprocessing, an information extraction function for extracting specific information contained in the image data for the image data obtained by the image input function, A computer-readable computer program that selects a pre-process corresponding to an image reading operation selected by an image input function and executes a pre-process selection function that causes the image pre-process function to execute the selected pre-process.

  According to a 56th aspect of the present invention, in the computer program according to the 55th aspect, the information extracting function extracts a characteristic included in the image data as specific information included in the image data.

  According to a 57th aspect of the present invention, in the computer program according to the 56th aspect, the feature extracted by the information extraction function is indicated by an array of dots.

  According to a 58th aspect of the present invention, in the computer program according to the 57th aspect, dots constituting the feature extracted by the information extraction function are expressed by a single color material.

  According to a 59th aspect of the present invention, in the computer program according to the 57th aspect, the dots constituting the feature extracted by the information extraction function are superimposed on the background of the original image from which the image data is based.

  According to a sixty-sixth aspect of the present invention, in the computer program according to the twenty-seventh aspect, the dots constituting the feature extracted by the information extracting function are superimposed on the front surface of the original image from which the image data is based.

  The invention according to claim 61 is the computer program according to claim 57, wherein the arrangement of dots constituting the feature extracted by the information extraction function is a plurality of types of dots superimposed on the original image from which the image data is based. It is one kind of dot pattern among the patterns.

  The invention according to claim 62 is the computer program according to claim 61, wherein, as the plurality of types of dot patterns, a dot pattern composed of dots of a size that is easy to reproduce and reproduce and a dot of a size that is difficult to reproduce and reproduce Dot pattern.

  According to a 63rd aspect of the present invention, in the computer program according to any one of the 55th to 62nd aspects, the image reading operation by the image input function obtains image data by moving the sensor relative to the document. And an operation for obtaining image data by moving the document relative to the sensor.

  According to a 64th aspect of the present invention, in the computer program according to any one of the 55th to 63rd aspects, the preprocessing selection function performs a noise removal process when a certain type of image reading operation is selected by the image input function. Do.

  According to a 65th aspect of the present invention, in the computer program according to any one of the 55th to 63rd aspects, the preprocessing selection function has a first density when a certain type of image reading operation is selected by the image input function. When another type of image reading operation is selected by the image input function, a second density correction different from the first density correction is performed.

  The invention according to claim 66 is installed in a computer provided in an image processing apparatus having an image input function capable of selecting a plurality of types of image reading operations. With respect to image data obtained by the image input function, A computer-readable computer program for executing an information extraction function for extracting specific information contained in image data in accordance with an image reading operation selected by the image input function.

  According to a 67th aspect of the present invention, in the computer program according to the 66th aspect, the information extracting function extracts a characteristic included in the image data as specific information included in the image data.

  According to a 68th aspect of the present invention, in the computer program according to the 67th aspect, the feature extracted by the information extraction function is indicated by an array of dots.

  According to a 69th aspect of the present invention, in the computer program according to the 68th aspect, the dots constituting the feature extracted by the information extraction function are expressed by a single color material.

  According to a 70th aspect of the present invention, in the computer program according to the 68th aspect, the dots constituting the feature extracted by the information extraction function are superimposed on the background of the original image from which the image data is based.

  According to a 71st aspect of the present invention, in the computer program according to the 68th aspect, the dots constituting the feature extracted by the information extraction function are superimposed on the front surface of the original image from which the image data is based.

  According to a 72nd aspect of the present invention, in the computer program according to the 68th aspect, the dot arrangement constituting the feature extracted by the information extraction function is a plurality of types of dots superimposed on the original image from which the image data is based. It is one kind of dot pattern among the patterns.

  The invention according to claim 73 is the computer program according to claim 72, wherein the plurality of types of dot patterns are composed of a dot pattern composed of dots of a size that is easy to reproduce and reproduce and a dot of a size that is difficult to reproduce and reproduce. Dot pattern.

  According to a 74th aspect of the present invention, in the computer program according to any one of the 66th to 73rd aspects, in the plural types of image reading operations by the image input function, the image data is obtained by moving the sensor relative to the document. And an operation for obtaining image data by moving the document relative to the sensor.

  According to a 75th aspect of the present invention, in the computer program according to any one of the 66th to 74th aspects, the information extraction function is specified in the image data in accordance with an image reading operation selected by the image input function. The threshold for extracting the information is changed.

  According to a 76th aspect of the present invention, in the computer program according to the 75th aspect, the threshold for extracting the specific information is a threshold for determining the presence or absence of a dot detection signal.

  The invention according to claim 77 is the computer program according to any one of claims 55 to 76, wherein the image output function for performing the output image processing necessary for the image data obtained by the image input means and outputting the image data; An output mode changing function for changing a mode of image output by the image output function according to specific information obtained by the image extraction function is executed by a computer.

  The invention according to claim 78 is the computer program according to claim 77, wherein the output mode change function is output when the specific information obtained by the image extraction function expresses the prohibition of faithful reproduction. The image is processed into an image that cannot be used as a copy.

  The invention according to claim 79 is the computer program according to claim 77, wherein the output mode changing function is output when the specific information obtained by the image extracting function expresses the prohibition of faithful reproduction. Process the image into an unreadable image.

  The invention according to claim 80 is the computer program according to claim 73, wherein the output mode changing function is configured to output the image when the specific information obtained by the image extracting function expresses prohibition of copying. Stops paper output processing by function.

  The invention according to claim 81 is the computer program according to claim 73, wherein the output mode changing function prohibits copying when the specific information obtained by the image extracting function expresses prohibition of copying. Notify the outside that the document to be scanned has been scanned.

  The invention according to claim 82 is a storage medium for storing the computer program according to any one of claims 55 to 81.

  According to the invention described in claims 1, 12, 21, 22, 28, 39, 48, 49, 55, 66, 75, 76, in the image data obtained by the image input means capable of selecting a plurality of types of reading operations. Also, information in the image can be extracted with high accuracy.

  According to the invention described in claims 2, 13, 29, 40, 56, and 67, the image data obtained by the image input means capable of selecting a plurality of types of reading operations is also a specific manuscript with high accuracy. Can be extracted.

  According to the invention described in claims 3, 14, 30, 41, 57, and 68, even in the image data obtained by the image input means capable of selecting a plurality of types of reading operations, the specific data represented by dots is accurately obtained. It can be extracted that it is a manuscript.

  According to the invention described in claims 4, 15, 31, 42, 58, and 69, even image data obtained by an image input means capable of selecting a plurality of types of reading operations is accurately expressed by dots of a single color material. Can be extracted.

  According to the inventions of claims 5, 16, 32, 43, 59, and 70, even in the image data obtained by the image input means capable of selecting a plurality of types of reading operations, it is expressed with dots accurately on the background of the document. Can be extracted.

  According to the invention described in claims 6, 17, 33, 44, 60, 71, even image data obtained by an image input means capable of selecting a plurality of types of reading operations is expressed with dots on the front surface of the document with high accuracy. Can be extracted.

  According to the invention described in claims 7, 18, 34, 45, 61 and 72, even image data obtained by an image input means capable of selecting a plurality of types of reading operations is expressed with a plurality of dot patterns with high accuracy. It is possible to extract a specific manuscript.

  According to the invention described in claims 8, 19, 35, 46, 62, 73, even in the case of image data obtained by an image input means capable of selecting a plurality of types of reading operations, the concealed character string at the time of copying is revealed with high accuracy. It is possible to extract that the document is a specific document expressed by a dot pattern having a conversion effect.

  According to the invention described in claims 9, 20, 36, 47, 63, 74, in the image data obtained by the image input means capable of selecting a plurality of types of reading operations, ie, a document moving type reading operation and a sensor reading type reading operation. In addition, it is possible to accurately extract information in an image, in particular, a specific document expressed by dots.

  According to the tenth, thirty-seventh and sixty-fourth aspects of the present invention, the image data obtained by the image input means capable of selecting a plurality of types of reading operations can be accurately obtained by removing noise according to the reading operations. It is possible to extract the inside information, in particular, a specific manuscript represented by dots.

  According to the eleventh, thirty-eight and thirty-sixth aspects of the present invention, the image data obtained by the image input means capable of selecting a plurality of types of reading operations by performing density correction according to the reading operations can be obtained with high accuracy. It is possible to extract the inside information, in particular, a specific manuscript represented by dots.

  According to the inventions of claims 23, 50 and 77, information in an image can be extracted with high accuracy even in image data obtained by an image input means capable of selecting a plurality of types of reading operations. Can be used for image processing.

  According to the invention described in claims 24, 51, and 78, it is determined that the image data obtained by the image input means capable of selecting a plurality of types of reading operations is a document for which accurate and faithful reproduction is prohibited. Depending on the determination result, the output image can be processed into a copy that cannot be used.

  According to the inventions of claims 25, 52, and 79, it is determined that the image data obtained by the image input means capable of selecting a plurality of types of reading operations is an original that is accurately and faithfully prohibited from being reproduced. The output image can be processed so as not to be readable according to the determination result.

  According to the inventions of claims 26, 53, and 80, it is possible to accurately determine that a document is prohibited from being copied even in image data obtained by an image input means capable of selecting a plurality of types of reading operations. Image formation can be interrupted according to the determination result.

  According to the invention described in claims 27, 54 and 81, the image data obtained by the image input means capable of selecting a plurality of types of reading operations, ie, the document movement type reading operation and the sensor reading type reading operation, can be obtained with high accuracy. It is possible to extract the inside information, in particular, a specific manuscript represented by dots, and notify the outside.

  According to the invention of claim 82, it has the same operational effect as the computer program of any one of claims 55 to 81.

  An embodiment of the present invention will be described with reference to FIGS.

1. Document Image First, a document image used in the image processing apparatus, the image processing method, and the like according to the present embodiment will be described with reference to FIGS.

  FIG. 1 is a front view illustrating a document image. FIG. 2 is a schematic diagram illustrating an example of a copy of an original image in which a dot pattern embedded in the original image is raised. FIG. 3 is a schematic diagram showing another example of a copy of an original image in which a dot pattern embedded in the original image appears floating. FIG. 4 is an enlarged schematic diagram illustrating the dot pattern illustrated in FIG. FIG. 5 is an enlarged schematic diagram illustrating the dot pattern illustrated in FIG.

  In the example shown in FIG. 1, the original paper 102 for creating the original image 101 which is a contract is used as the original paper 102 in which a dot array for output prohibition determination, that is, a dot pattern 103 is embedded in the background. It has been. However, as another embodiment, the original paper 102 on which the dot pattern 103 is not formed may be used, and the dot pattern 103 may be simultaneously formed when the original image 101 that is a contract is created. That is, the dot pattern 103 may be preliminarily printed on the original paper 102 or may be simultaneously formed when characters, graphics, or the like are formed on the original paper 102.

  The dot pattern 103 described with reference to FIGS. 1 to 5 includes a base area 104 and a message area 105. The base area 104 is a ground area that occupies most of the original paper 102. The message area 105 is an area for expressing a message distributed in the base area 104. The base area 104 and the message area 105 are not different from each other in the basic configuration of the dot pattern 103 itself, and are divided into the base area 104 and the message area 105 as human value judgment for the dot pattern 103. It ’s just that. In the present embodiment, the message area 105 is composed of characters “copy prohibited”. As characters constituting such a message area 105, any characters, symbols, and the like can be used in addition to “copy prohibited”.

  In the original image 101 created using such original paper 102, when the original image 101 is copied, a part of the dot pattern 103 emerges. In this case, in the document image 101 illustrated in FIG. 2, the message area 105 expressed by the characters “copy prohibited” appears, and in the document image 101 illustrated in FIG. 3, the base area 104 appears.

  As described above, when the original image 101 in which the dot pattern 103 is embedded is copied, either the message area 105 or the base area 104 in the dot pattern 103 appears. The other area 104 or 105 is copied. This is because it is difficult to (reproduce). In other words, when the region 104 or 105 that does not rise up disappears, the remaining region 104 or 105 appears to rise up.

  The dot pattern 103 that causes such a phenomenon is constituted by a set of two types of dots 106 having different sizes as exemplified in FIGS. 4 and 5. That is, of these two types of dots 106, the small dots 106b are difficult to be copied (reproduced), and the large dots 106a are copied (reproduced). Therefore, in the dot pattern 103 illustrated in FIG. 2 in which the message area 105 rises after copying, as shown in FIG. 5, a small dot 106 b is used for the base area 104, and a larger dot is used for the message area 105. Dots 106a are used. On the other hand, in the dot pattern 103 illustrated in FIG. 3 in which the base area 104 is lifted after copying, as shown in FIG. 4, a small dot 106 b is used for the message area 105, and the base area 104 is formed. Large dots 106a are used.

  As another configuration example of the dot pattern 103, the message area 105 or the base area 104 that emerges is not limited to a dot, but can be configured by a thin line pattern, a specific pattern pattern, or the like.

  Here, in the present embodiment, the message area 105 or the base area 104 is treated as a feature amount. The feature amount referred to here is, for example, a certain feature included in image data obtained by reading the document image 101. Such a characteristic can function as predetermined information included in the image data. For example, if the message area 105 or the base area 104 that emerges is composed of dots 106, the size, density (number of dots per unit area), the distance between dots, and the dot arrangement are determined according to the message area 105 or base that emerges. If the area 104 is composed of a thin line pattern, the width of the line is assumed, and if the message area 105 or the base area 104 is formed with a specific pattern pattern, the characteristics of the pattern are characterized. Can be used as a quantity.

  In the above example, the feature quantity for the message area 105 or the base area 104 that emerges is limited. However, as such a feature quantity, the base area 104 or the message area 105 that does not emerge, or the message area 105 or the base area 104 that emerges. For each of the base region 104 or the message region 105 that does not appear, the respective feature amounts may be obtained. In short, if one or both of the base area 104 and the message area 105 can be read as data when the original image 101 on which the image data with the dot pattern 103 embedded is recorded is read, Can be treated as a feature quantity. Or, in the image data in which the dot pattern 103 existing in the form of data is embedded, if either or both of the base area 104 and the message area 105 constituting the dot pattern 103 can be read as data, Can be treated as a feature quantity.

  As another embodiment, as described above, the dot pattern may have a pattern other than the dot pattern 103 having the base region 104 and the message region 105 as illustrated in FIGS. 6 and 7. 6 is a front view illustrating a document image and its dot pattern as an embodiment different from that described with reference to FIGS. 1 to 5, and FIG. 7 is a document image different from FIG. It is a front view which illustrates a dot pattern. The dot pattern 103 illustrated in FIGS. 6 and 7 is composed of dots 106 having a single size. The dot pattern 103 illustrated in FIGS. 6 and 7 is the same as the pattern corresponding only to the base region 104 and the pattern corresponding to only the message region 105 in the dot pattern 103 illustrated in FIGS. It can be grasped as a thing. That is, even if the dot pattern 103 corresponds to the base area 104 or the dot pattern 103 corresponding to the message area 105, the dot pattern 103 is used as data when the document image 101 having such a dot pattern 103 is read. If it can be read, it can be treated as a feature quantity.

  Here, the dot pattern 103 illustrated in FIG. 6 is configured as a background image of the document image 101 drawn on the document sheet 102. In this sense, the background dot pattern is the same as the dot pattern 103 illustrated in FIGS. On the other hand, the dot pattern 103 illustrated in FIG. 7 is drawn on the original image 101 drawn on the original paper 102. Therefore, the dot pattern 103 illustrated in FIG. 7 is not a background dot pattern, but a so-called foreground dot pattern. This foreground dot pattern can also be applied to the dot pattern 103 composed of two types of areas, the message area 105 and the base area 104 illustrated in FIGS. That is, two different types of dot pattern configurations, that is, the background dot pattern or the foreground dot pattern, can be similarly applied to the dot pattern 103 illustrated in FIGS.

  Here, the dot patterns 103 shown in FIGS. 2 to 5, 6, and 7 are all dot patterns 103 in which the relationship between single dots 106 constituting the dot pattern 103 has a characteristic amount. That is, the feature amount of the dot pattern 103 shown in FIGS. 2 to 5, 6, and 7 can include the dot density (the number of dots per unit area) and the inter-dot distance. Are included in the relationship between the single dots 106 constituting the dot pattern 103. Therefore, a feature amount called dot density and a feature amount called inter-dot distance are compared and considered.

  First, the dot density, which is the density of the dots 106, is the number of dots per unit area, as described above. This feature amount can be determined by counting the number of dots in a certain unit area to be determined. When counting the actual number of dots, some detection omissions and false detections occur. Therefore, by setting a certain allowable value (threshold) for the number of dots counted, the feature amount can be detected using the density of the dots 106 as a feature amount. In this case, if the threshold value to be set with respect to the number of counted dots is a strict value, that is, a value with a small allowable range, a determination omission is likely to occur, and conversely, the threshold value set with respect to the number of counted dots. If the value is a loose value, that is, a value with a large allowable range, it is difficult to cause omission of determination.

  However, referring to the dot pattern 103 illustrated in FIG. 6, a certain dot 106 included in the dot pattern 103 is undetectable by the document image 101 drawn on the document sheet 102. On the other hand, in the dot pattern 103 illustrated in FIG. 7, since the dot pattern 103 is drawn on the original image 101 drawn on the original paper 102, each dot pattern 103 illustrated in FIG. In this state, the dots 106 are easily detected. However, the dots 106 that overlap the original image 101 drawn on the original paper 102 are not always correctly detected. For this reason, the degree to which the number of dots can be detected is greatly affected by the form of the original image 101 drawn on the original paper 102. Therefore, when the dot density is a feature amount, it can be said that even if a certain allowable value (threshold value) is set for some of the counted number of dots, a determination omission is likely to occur in the determination of the feature amount.

  Next, the case where the distance between dots, which is the distance d between adjacent dots 106, is used as a feature amount will be described with reference to FIGS. FIG. 8 is a schematic diagram illustrating a dot pattern 103 in which the distance between dots is d. FIG. 9 represents the distance distribution between adjacent dots 106 with the distance between dots on the horizontal axis and the appearance frequency on the vertical axis. It is a graph to do. FIG. 10 is a schematic view illustrating a dot pattern.

  As shown in FIG. 8, in a dot pattern 103 (see FIG. 10A) in which the distance d between two adjacent dots 106 is constant, two adjacent dots 106 constituting this dot pattern 103 are displayed. The inter-dot distance, which is the distance d between the two dots 106, is a feature amount. In this case, when the dot 106 is detected from the dot pattern 103 and a plurality of distances d between the adjacent dots 106 are determined, the distance d between the correct dots 106 as shown in the graph of FIG. Draws a sharp mountain distribution with a peak value of. Therefore, by setting a certain allowable value (threshold value) for the distance d, the distance d between the dots 106 can be detected as a feature amount. In this case, as is apparent from the graph of FIG. 9A, if the threshold value set for the distance d is a strict value, that is, if the allowable range is small, determination omission is likely to occur. On the other hand, if the threshold value to be set is a loose value, that is, a value with a large allowable range, it is difficult to cause omission of determination.

  However, when the distance d between the two dots 106 in the dot pattern 103 in which the distance d between two adjacent dots 106 is constant is used as the feature amount, the pattern illustrated in FIG. The pattern illustrated in b) cannot be distinguished. Here, FIG. 10A illustrates a dot pattern in which there are four other dots 106 separated by a distance d when viewed from the target dot 106, and FIG. 10B illustrates the dot pattern viewed from the target dot 106. A dot pattern having three other dots 106 separated by a distance d is illustrated. In this case, as an example, assuming that FIG. 10A is the dot pattern 103 to be detected, a pattern of dots 106 as illustrated in FIG. 10B on the original image 101 drawn on the original paper 102. Is included, or vice versa, the dot pattern 103 as a detection target can be detected when the dot density is a feature amount, whereas the inter-dot distance is a feature amount. In this case, it becomes impossible to detect the dot pattern 103 that is the detection target. That is, overdetection is likely to occur when the dot pattern 103 in which the distance d between two adjacent dots 106 is constant is to be detected.

  Therefore, FIG. 10C illustrates a dot pattern 103 that can correctly detect the dot pattern 103 that is a detection target using the inter-dot distance as a feature amount. In the dot pattern 103 illustrated in FIG. 10C, a plurality of types of distances d between dots are set, and the frequency of each type of inter-dot distance d is set to be different. For this reason, as shown in FIG. 9B, the peak value distribution of the distance between adjacent dots has a feature amount. Therefore, by detecting such a feature quantity as the peak value distribution of the distance between adjacent dots, the dot pattern 103 to be detected can be detected without overdetection.

  As described above, each of the dot patterns 103 shown in FIGS. 2 to 5, 6, and 7 has a characteristic amount in the relationship between the single dots 106 constituting the dot pattern 103. On the other hand, FIGS. 11 and 12 show an example in which the dot pattern 103 itself or the set of dot patterns 103 has a feature amount. That is, as illustrated in FIG. 11, the distance α-β between the dot 106α and the dot β, the distance β-γ between the dot β and the dot γ, and the distance between the dot α and the dot γ. α-γ is set to be different from each other, and a set of dot patterns 103 is constituted by these dots α, β, and γ. Such a dot pattern 103 can be easily determined by pattern matching as an example. Then, the dot pattern 103 itself as illustrated in FIG. 11 or a set of such dot patterns 103 has a feature amount. As a method for giving a feature amount to a set of dot patterns 103, for example, the density per unit area of the dot patterns 103 as illustrated in FIG. 11 can be adopted.

  Note that the dot pattern 103 in the present embodiment merely exemplifies specific information included in the image data, that is, an arrangement of dots that is an aspect of a certain characteristic included in the image data. The specific information included in the image data, that is, the characteristic included in the image data may be a representation of a line other than a dot or other shape, or may be a color other than the shape or other image feature. . Further, information on an image to be extracted, which will be described later, may be information such as the ID of a document included in the image data, information attached to the document, for example, information such as a creator. Further, the image feature may be a document type or a feature indicating a plurality of types of documents. Further, the type may have a hierarchical type configuration.

2. Outline of the Embodiment The image processing apparatus according to the present embodiment has a pattern for determining a confidential document or an output-prohibited document included in the image data of the original image 101, for example, a dot pattern 103 (for example, illustrated in FIG. 6). The feature amount (dot density, inter-dot distance, specific dot pattern, specific dot pattern density per unit area, etc.) of the background dot pattern 103 to be detected and the foreground dot pattern 103 illustrated in FIG. 7 is detected from the image data. A process of comparing the detected feature quantity with the reference feature quantity stored in the storage area as the feature quantity of the output prohibited dot pattern, and determining whether the detected feature quantity and the reference feature quantity are identical. Execute. By this processing, the function of the information extraction unit, the information extraction process, and the information extraction function for extracting specific information included in the image data is executed on the image data of the document image 101. In such information extraction processing, the identity determination is performed, for example, based on whether or not the difference between the detected feature value and the reference feature value is smaller than a certain threshold value. That is, if such a difference is smaller than a certain threshold value, it is determined that they are the same.

  The image processing apparatus according to the present embodiment performs a plurality of types of reading operations, for example, an image input unit capable of selecting a document moving type reading operation and a sensor moving type reading operation, for example, image data obtained by a scanner. Even if the image reading operation is obtained by any kind of reading operation, information in the image can be extracted with high accuracy.

  As a technique for this, in the present embodiment, for example, image preprocessing means for selectively executing one or more preprocessing on image data obtained by a scanner is provided, and image preprocessing step, image preprocessing is performed. Perform the function. As such pre-processing, for example, noise removal processing and automatic density correction processing are selected. When a document movement type reading operation is performed, if dust adheres to the document table (contact glass), vertical stripes may occur in the read image due to the dust. Therefore, in such a case, it is desirable to execute information extraction processing for determining whether or not specific information is included in the image data after performing preprocessing called noise removal processing. Further, even if the image data based on the same document image 101 is between the image data obtained by performing the document movement type reading operation and the image data obtained by performing the sensor movement type reading operation, A subtle contrast is produced in the background of the original image 101. Therefore, when automatic density correction is performed on image data, the first density correction is performed on the image data obtained by performing the document movement type reading operation, and the image data obtained by performing the sensor movement type reading operation. Is subjected to a second density correction different from the first density correction. In this case, it is preferable that the first density correction is lighter than the second density correction. What is important in this embodiment is that the type of preprocessing corresponding to the type of image reading operation selected by the scanner is selected, and the selected preprocessing is executed by the image preprocessing unit. . For example, when the document movement type reading operation is selected, execution of noise removal processing is selected as preprocessing, and first density correction is executed as the degree of automatic density correction. On the other hand, when the sensor movement type reading operation is selected, the execution of the noise removal process is not selected as the pre-processing, and the second density correction is executed as the degree of the automatic density correction. Here, the function of the pre-processing selection means for selecting the pre-processing according to the image reading operation selected by the image input means, for example, the scanner, and causing the image pre-processing means to execute the selected pre-processing, the pre-processing selecting step, A process selection function is executed.

  In the present embodiment, as an example of another technique that enables accurate extraction of information in an image, even if it is image data obtained by any type of reading operation of a plurality of types of reading operations, The information extraction means is also caused to execute information extraction processing corresponding to the image reading operation selected by the image input means such as a scanner. In other words, the distance between the original paper 102 and the sensor of the scanner is relatively constant during the document movement type reading operation, whereas the original paper 102 is removed from the original table (contact glass) during the sensor movement type reading operation. The image characteristics of the read image data are likely to fluctuate, such as floating or not having a pressure plate on the original paper 102. Therefore, at the time of the sensor movement type reading operation, as an example, a plurality of threshold values for determining the presence or absence of the dot detection signal in the information extraction means are set to prevent detection failure of the dot pattern 103.

  The present specification also discloses an image processing method in such an image processing apparatus, a computer program for image processing, and a storage medium for storing such a computer program.

3. Image processing apparatus, image processing method, computer program, and storage medium for storing computer program (1) Hardware configuration FIG. 13 is a block diagram showing hardware resources of an image processing apparatus. FIG. 14 is a vertical side view of a scanner as an image input unit. FIG. 15 is a block diagram of the dot pattern detection unit shown in FIG.

  As shown in FIG. 13, the image processing apparatus according to the present embodiment has a digital copying machine configuration in which a scanner 201 as an image reading apparatus, an image processing unit 202 having a digital circuit configuration, and a plotter 203 are controlled by a system controller 204. ing. The system controller 204 constitutes a part of the computer 1001, and utilizes computer functions constructed from a built-in CPU (Central Processing Unit) 204a, ROM (Read Only Memory) 204b, and RAM (Random Access Memory) 204c, The scanner 201, the image processing unit 202, and the plotter 203 are controlled according to an instruction from the operation display unit 205, and necessary information is returned to the operation display unit 205 for display.

  Here, the scanner 201 will be described in detail with reference to FIG. As shown in FIG. 14, an ADF (Automatic Document Feeder) 3 that is an automatic document feeder is provided on the scanner body 2 of the scanner 201 so as to be freely opened and closed by a support shaft 3a. The mode and the sensor movement type reading operation mode are freely selectable. The scanner body 2 is provided with a rectangular contact glass 4 on which a document sheet 102 (not shown in FIG. 14) is placed. The scanner body 2 facing the contact glass 4 includes an optical glass. A unit 5 is provided. The optical unit 5 is generally called a contact image sensor, and includes a light source (not shown) and a large number of image sensors (CCD: Charge Coupled Device) corresponding to a document reading width. And a line sensor (not shown) arranged in a line. In the optical unit 5, the arrangement direction of these image pickup elements is a main scanning direction B (a direction orthogonal to the paper surface of FIG. 14). A driving motor 6 that is a stepping motor is connected to the optical unit 5 by a pulley, a wire, or the like, and the optical unit 5 is movable in the sub-scanning direction A. This optical unit 5 is normally positioned on the upstream side in the sub-scanning direction A shown in FIG. 14 as a home position HP, and in the process of moving to the downstream side, the light source is placed on the original paper 102 placed on the contact glass 4. An original image 101 (not shown in FIG. 14) is irradiated, and a line sensor receives reflected light from the original paper 102, whereby the original image 101 is read and scanned.

  Such scanning of the original image 101 by the optical unit 5 is executed under the setting of the book mode which is a sensor movement type reading operation mode. The ADF mode, which is the mold reading operation mode, is also set as an operation mode so as to be switchable in an operation display unit 205 (see FIG. 13) described later. Under the setting of the ADF mode, the original image 101 is read and scanned by sequentially transporting the original paper 102 by the ADF 3 with the optical unit 5 set at the home position HP.

  The ADF 3 includes a document tray 7, a pickup roller 8, a pair of registration rollers 9, a transport drum 10, a pair of transport rollers 11, a pair of paper discharge rollers 12, and the like on the optical unit 5 positioned at the home position HP. Are sequentially conveyed in the sub-scanning direction A and discharged onto the paper discharge tray 13. The paper discharge tray 13 is provided on the upper surface of the document cover 14, and the document cover 14 can be opened and closed with respect to the contact glass 4. A stepping motor (not shown) is connected to the pickup roller 8, the registration roller 9, the conveyance drum 10, the conveyance roller 11, and the paper discharge roller 12 of the ADF 3 by a gear train or the like.

  Next, the image processing unit 202 will be described. The image processing unit 202 includes a preprocessing unit 210, a reproduction image processing unit 211, an information extraction unit 212, and an output control unit 213.

  As shown in FIG. 13B, the preprocessing unit 210 performs noise removal processing, automatic density correction processing, and the like. Details thereof will be described later. The pre-processing unit 210 receives image data from the scanner 101, performs necessary pre-processing on the received image data, and then converts the pre-processed image data into a reproduction image processing unit 211 or an information extraction unit. 212. Such a pre-processing unit 210 can be constructed by a digital circuit, a processor such as SIMD, or the like.

  The reproduction image processing unit 211 includes a filter processing unit 206, a scaling processing unit 207, a γ processing unit 208, and a gradation processing unit 209. Each of these units 206, 207, 208, and 209 is not different from an equivalent circuit configuration provided in a general digital copying machine, and a description thereof will be omitted.

  The output control unit 213 determines whether to output image data of an image read by the scanner 101 to the plotter 106 according to a determination result described later in the information extraction unit 212.

  As illustrated in FIG. 13C, the information extraction unit 212 includes a dot pattern detection unit 212a and an output prohibited document determination unit 212b. The dot pattern detection unit 212a and the output prohibition document determination unit 212b are hardware having a digital circuit configuration that detects and determines an output prohibition document such as a confidential document and prohibits copying. As another embodiment, the information extraction unit 212 may be configured by a processor such as SIMD.

  The dot pattern detection unit 212a detects a feature amount of the dot pattern 103 for confidential document determination or output prohibition document determination included in the image data of the original image 101, in this case, a process for detecting the dot density from the image data, and the detection A process of comparing the dot density, which is a feature quantity, with the reference feature quantity stored in the storage areas 253a, 254a, 255a (see FIG. 15) as the feature quantity of the output prohibited dot pattern, and the dot density, which is the detected feature quantity, And a process of determining whether or not the identity is recognized with the reference feature amount. By these processes, the function of the information extraction unit, the information extraction process, and the information extraction function for extracting specific information included in the image data is executed on the image data of the document image 101.

  In order to execute such various processes, the dot pattern detection unit 212a includes storage areas 253a, 254a, and 255a for storing and saving reference feature amounts as the feature amounts of output prohibited dot patterns. Further, for example, FIG. The hardware configuration illustrated in (a) is included. In other words, the dot 106 is detected by the dot detection unit 251 from the image data based on the original image 101 read by the scanner 201. As a detection method in this case, various conventional methods for detecting an image pattern by a digital circuit, for example, pattern matching can be used. As an example, the luminance data of the input image data or a specific signal of RGB, for example, the G signal is binarized and template data stored in the storage areas 253a, 254a, and 255a for the binary image data. Detects a match with. When the target dot is composed of a specific color, it is binarized by calculating the degree of coincidence with the specific color instead of binarizing the luminance signal and comparing the degree of coincidence with a threshold value. Alternatively, a binarization result is obtained by using a signal having a high sensitivity of a specific color from among RGB signals, for example, a B signal for a yellow dot. Next, the dot density determination unit 252 calculates the dot density within a certain area in the dots 106 detected by the dot detection unit 251. Such calculation processing is executed by a counter or an adder having a digital circuit configuration.

  Here, the dot pattern detection unit 212 a includes a base area dot number determination unit 253 and a message area dot number determination unit 254 subsequent to the dot density determination unit 252. These base area dot number determination section 253 and message area dot number determination section 254 have storage areas 253a and 254a, respectively. The base area dot number determination unit 253 is identical to the dot density of the base area 104 of the dot pattern 103 that prohibits output in a certain area as a reference for dot density calculation in the dot density determination unit 252 in the storage area 253a. A threshold value (threshold value for the reference feature) that is an allowable value for determination and a threshold value that is an allowable value for identity determination for the number of dots in the base region 104 of the dot pattern 103 that prohibits the output included in one original sheet 102 ( And a threshold value for the reference feature value). The message area dot number determination unit 254 has, in the storage area 254a, the identity of the dot pattern 103 that prohibits output within a certain area as a reference for dot density calculation in the dot density determination unit 252 to the dot density of the message area 105. A threshold value (threshold value for the reference feature value) that is an allowable value for determination and a threshold value that is an allowable value for determining the identity of the number of dots in the message area 105 of the dot pattern 103 that prohibits the output included in one original sheet 102 ( And a threshold value for the reference feature value).

  Therefore, the base area dot number determination unit 253 includes the density of the dots 106 calculated by the dot density determination unit 252 within a threshold relating to the dot density stored in the storage area 253a (threshold for the reference feature amount). If it is determined that the number of dots 106 of the same size detected by the dot detection unit 251 is accumulated, for example, in a counter. When the accumulated number of dots is included in the threshold value regarding the number of dots stored and stored in the storage area 253a (threshold value for the reference feature value), the base area 104 of the dot pattern 103 whose output is prohibited is displayed. It is determined that it exists, and the determination result is output to the output prohibited document determination unit 212b.

  On the other hand, in the message area dot number determination unit 254, the density of the dots 106 calculated by the dot density determination unit 252 is included in the threshold value relating to the dot density stored in the storage area 254a (threshold value for the reference feature amount). If it is determined that the number of dots 106 of the same size detected by the dot detection unit 251 is accumulated, for example, in a counter. When the accumulated number of dots is included in the threshold value regarding the number of dots stored and saved in the storage area 254a (threshold value for the reference feature amount), the message area 105 of the dot pattern 103 whose output is prohibited is displayed. It is determined that it exists, and the determination result is output to the output prohibited document determination unit 212b.

  Next, the output-prohibited document determination unit 212b receives the processing result of the dot pattern detection unit 212a, and executes an output-prohibited document determination process according to a predetermined determination criterion. For example, this determination criterion may be stored as a parameter in a storage area (not shown) included in the output prohibited document determination unit 212b, or may be set in such a storage area by an input from the operation display unit 205. For example, the output prohibited document determination unit 212b is an output prohibited document such as a confidential document when one of the base area 104 and the message area 105 of the dot pattern 103 exists in the original image 101 read by the scanner 201. Judge that there is. Alternatively, the output prohibited document determination unit 212b is an output prohibited document such as a confidential document when both the base area 104 and the message area 105 of the dot pattern 103 exist in the original image 101 read by the scanner 201. Judge that there is.

  Note that the output prohibition document determination condition in the output prohibition document determination unit 212b is set to determine that there is an output prohibition document when one of the base area 104 and the message area 105 of the dot pattern 103 exists. If so, it is not necessary to provide both the base area dot number determination unit 253 and the message area dot number determination unit 254 in the dot pattern detection unit 212a. For example, when only the base region 104 is used as the output prohibition document determination condition, the configuration shown in FIG. 15B including only the base region dot number determination unit 253 may be used. When only the message area 105 is used, the configuration shown in FIG. 15C including only the message area dot number determination unit 254 may be used.

  Further, when the document image 101 having the dot pattern 103 as illustrated in FIG. 6 and FIG. 7 in which the dot pattern 103 is composed of dots 106 having a single size is to be determined as an output-prohibited document, It is not necessary to provide both the base area dot number determination unit 253 and the message area dot number determination unit 254 in the subsequent stage of the dot density determination unit 252, and only the dot number determination unit 255 may be provided. The dot number determination unit 255 has a storage area 255a, and the storage area 255a has a dot density for the dot density of the dot pattern 103 that prohibits output within a certain area as a reference for dot density calculation in the dot density determination unit 252. A threshold value that is an allowance value for the identity determination (threshold value for the reference feature value) and a threshold value that is an allowance value for the identity determination for the number of dots of the dot pattern 103 that prohibits the output included in one document sheet 102 (reference feature) (Threshold for quantity) is stored and saved.

  Therefore, the dot number determination unit 255 includes the density of the dots 106 calculated by the dot density determination unit 252 within a threshold relating to the dot density stored in the storage area 255a (threshold for the reference feature amount). If it is determined, the number of dots 106 of the same size detected by the dot detection unit 251 is accumulated in, for example, a counter. If the accumulated number of dots is included in the threshold value regarding the number of dots stored and saved in the storage area 255a (threshold value for the reference feature value), it is determined that there is a dot pattern 103 whose output is prohibited. The determination result is output to the output prohibited document determination unit 212b.

  As described above, the output-prohibited document determination unit 212b executes means and a process for determining whether or not identity is recognized between the detected dot density, which is the feature quantity, and the reference feature quantity. That is, the output prohibited document determination unit 212b receives the processing result of the dot pattern detection unit 212a, and executes the determination process of the output prohibited document according to a predetermined determination criterion. As a determination criterion in this case, for example, when a determination result indicating that there is a dot pattern 103 whose output is prohibited is transmitted from the dot number determination unit 255, it is determined that the document is an output prohibited document such as a confidential document. The standard is adopted.

  As described above, the dot number determination units 253, 254, and 255 and the output-prohibited document determination unit 212b have been described. The image processing unit 202 having the digital circuit configuration in the present embodiment is a background image included in the image data of the original image 101. The dot pattern 103 embedded in the image data is detected from the image data, and the dot density, which is the feature amount of the detected dot pattern 103, is prohibited from the output stored in the storage areas 253a, 254a, 255a. And the process of determining the identity is executed. Thereby, in the present embodiment, the sameness between the detected feature amount of the dot pattern 103 and the feature amount of the dot pattern stored in the storage areas 253a, 254a, and 255a is confirmed regardless of the type of document image. Thus, it is possible to determine whether or not output of image data of the document image 101 should be prohibited.

  When the output prohibited document determination unit 212b determines that the document image 101 read by the scanner 201 is an output prohibited document, the output prohibited document determination unit 212b transmits to the system controller 204 that an output prohibited document has been detected. In response to this, the system controller 204 prohibits post-processing after detection of the output-prohibited document, that is, the copying operation in the plotter 203. Here, when it is determined that the document image 101 read by the scanner 201 is an output-prohibited document, that is, the feature amount of the detected dot pattern 103 and the feature amount of the dot pattern stored in the storage areas 253a, 254a, and 255a. When the system controller 204 determines that the identity is recognized, a process for prohibiting the image data output process, which is a post process, is executed. As a result, when the original image 101 read by the scanner 201 is an output prohibited document such as a confidential document, copying (reproduction) is prevented.

  In this embodiment, copying is prohibited as an example of prohibiting output of an output-prohibited document. However, this is merely an example. For example, scanner distribution is performed on image data of an original image 101 read by the scanner 201. It is also possible to adopt an output prohibition mode that prohibits. In this case, as a delivery method, it is possible to send an output prohibited document as a text or an attached document by e-mail, facsimile transmission, or data transmission. Such various distribution modes are prohibited.

  As another embodiment, the system controller 204 performs post-processing after detection of an output-prohibited document in a state where the copy result of the original document image 101 cannot withstand use as a copy or in an unreadable state. The image signal of the original document image 101 may be changed so as to be output from As an example, a process of changing the image signal to a certain pixel value and filling it in is performed. In this case, it is possible to paint in gray (128 in 256 gradations), white, black, or the like. As another example, the image signal may be changed so that some pattern is repeatedly generated.

  Alternatively, a communication processing unit connected to, for example, the system controller 204 indicates that a copy-prohibited document has been read or output as an aspect of suppressing output even if the output of the output-prohibited document itself is not prohibited. Notification may be made to the external remote control unit 215 via 214. The remote operation unit 215 as a notification destination in this case is an example, but is preferably a broad computer used by various managers such as a personal computer, a mobile computer, a mobile phone, and the like.

  In contrast, when the original image 101 read by the scanner 201 is not an output-prohibited document such as a confidential document, a normal copying operation is executed. That is, the image data of the original image 101 read by the scanner 201 is processed by the image processing unit 202, and the result is output by the plotter 203.

(2) Processing Process The example in which the dot pattern detection unit 212a and the output prohibited document determination unit 212b are configured as a processor such as a digital circuit or SIMD has been described above. On the other hand, in implementation, the computer 1001 detects the dot pattern based on the computer software installed in the HDD 216 or the like that can be used as the firmware in the ROM 204b of the system controller 204 as an example, and can be used by the system controller 204 as another example. The functions of the unit 212a and the output prohibited document determination unit 212b may be realized. In this case, it is preferable that all or a part of the computer software installed in the HDD 216 or the like is copied to the RAM 204c included in the system controller 204 and executed.

  Here, the processing contents of the dot pattern detection unit 212a and the output prohibited document determination unit 212b executed by the computer 1001 based on the computer software will be described based on the flowchart shown in FIG. The dot pattern detection unit 212a and the output prohibition document determination unit 212b having the above-described digital circuit configuration indicate that the image data of the original image 101 read by the scanner 201 is a confidential document or the like with reference to the dot density included in the dot pattern 103. Judgment was made whether the document is an output-prohibited document. On the other hand, the flowchart shown in FIG. 16 includes the dot pattern 103 for determining whether the image data of the original image 101 read by the scanner 201 is an output prohibited document such as a confidential document. Not only the dot density but also the inter-dot distance, that is, the distance d between the dots 106 illustrated in FIG. 8 is referred to. The processing content of the flowchart shown in FIG. 16 is as follows.

  First, the processing routine shown in FIG. 16 is executed at a certain time, and it is repeatedly determined whether the document image 101 is read by the scanner 201 and image data is input (step S101).

  When it is determined that the original image 101 is read by the scanner 201 and the image data is input (Y in step S101), for example, the input image data is copied into the image memory included in the memory represented by the comprehensive concept of the RAM 204c. (Step S102).

  On the other hand, if it is not determined that the original image 101 has been read by the scanner 201 (N in step S101), whether the original image 101 has been read by the scanner 201 and image data has been input. Wait for the decision.

  Then, the dot 106 is detected from the image data copied to such an image memory (step S103). As a detection method in this case, various conventional methods for detecting an image pattern by computer processing, for example, pattern matching can be used.

  Next, the dot density within a certain area of the detected dots 106 is calculated by the calculation function of the CPU 204a (step S104).

  Here, in the present embodiment, data for determining the dot density and data for determining the distance between dots are stored in the storage area of the nonvolatile memory or battery backup memory included in the RAM 204c. Is saved. As data for dot density determination, a threshold value (threshold value for a reference feature value) that is an allowable value for identity determination with respect to the dot density of the dot pattern 103 that prohibits output within a certain area that is a reference for dot density calculation, A threshold value (threshold value for the reference feature value) that is an allowable value for identity determination with respect to the number of dots of the dot pattern 103 that prohibits output included in one document sheet 102 is stored and saved. As data for determining the inter-dot distance, a threshold value regarding the peak position and peak value of the inter-dot distance frequency (threshold value for the reference feature value) is stored and saved.

  The threshold value for the reference feature amount used for the two types of determinations of the dot density determination and the inter-dot distance determination is the output prohibited document with the document image 101 having the dot pattern 103 including the base region 104 and the message region 105. And a document image 101 having the dot pattern 103 as illustrated in FIGS. 6 and 7 in which the dot pattern 103 is composed of dots 106 having a single size. Assuming that, a set of a plurality of types of threshold values is stored and saved in a storage area in the RAM 204c. In other words, assuming that the document image 101 having the dot pattern 103 including the base area 104 and the message area 105 is to be determined as an output-prohibited document, the storage area in the RAM 204c has a reference for dot density calculation. The threshold value (threshold value for the reference feature value) that is an allowable value for the identity determination with respect to the dot density of the base area 104 of the dot pattern 103 that prohibits the output within a certain area, and the output included in one document sheet 102 are prohibited. A threshold value (threshold value for the reference feature value) that is an allowable value for determining the identity of the dot pattern 103 with respect to the number of dots in the base region 104, and a threshold value for the peak position and peak value of the inter-dot distance frequency (threshold value for the reference feature value). Is stored and saved, and output within a certain area, which is the basis for dot density calculation, is prohibited A threshold value (threshold value for a reference feature) that is an allowable value for determining the identity of the dot density in the message area 105 of the dot pattern 103 to be output, and the message area 105 of the dot pattern 103 that prohibits the output included in one document sheet 102. A threshold value (threshold value for the reference feature value) that is an allowable value for the identity determination with respect to the number of dots and a threshold value regarding the peak position and peak value of the inter-dot distance frequency (threshold value for the reference feature value) are stored and stored. Further, it is assumed that the document image 101 having the dot pattern 103 as illustrated in FIG. 6 and FIG. 7 in which the dot pattern 103 is composed of dots 106 of a single size is to be determined as an output-prohibited document. In the storage area in the RAM 204c, a threshold value (threshold value for the reference feature value) that is an allowable value for determining the identity with respect to the dot density of the dot pattern 103 that prohibits output within a certain area serving as a reference for dot density calculation, A threshold value (threshold value for the reference feature value) that is an allowable value for the identity determination with respect to the number of dots of the dot pattern 103 that prohibits output included in one manuscript paper 102, and a threshold value for the peak position and peak value of the inter-dot distance frequency ( And a threshold value with respect to the reference feature amount).

  In subsequent steps S105 to S110, the dot density and the inter-dot distance detected as the feature amount of the original image 101 inputted in step S101 and the feature amount of the dot pattern of the output prohibited document such as a confidential document are set. It is determined whether identity is observed.

  Specifically, in step S105, the dot density within a certain area of the detected dots 106 calculated in step S104 is a threshold (reference feature) regarding the dot density of the dot pattern 103 stored and stored in the storage area in the RAM 204c. It is determined whether it is included in the threshold value for the quantity. The dot density of the dot pattern 103 in this case means the dot density of the base region 104 in the dot pattern 103, the dot density of the message region 105 in the dot pattern 103, or the dot density of the dot pattern 103.

  Then, as a result of the determination in step S105, the dot density within a certain area of the detected dots 106 calculated in step S104 is a threshold (reference) for the dot density of the dot pattern 103 stored and stored in the storage area in the RAM 204c. If the determination result is affirmative (Y in step S105), the number of dots of the detection dots 106 is accumulated, for example, in the resist region of the RAM 204c. If the determination result is negative (N in step S105) after the temporary storage process (step S106), the processes in steps S107 and S108 are executed without passing through the process in step S106. The

  Here, when the determination result in step S105 is negative, for the feature quantity called dot density, the dot density detected as the feature quantity of the document image 101 and the feature quantity of the dot pattern of the output-prohibited document. It means that there is no identity between them. On the other hand, the cumulative number of detected dots 106 temporarily stored in the registration area of the RAM 204c in step S106 is the dot density detected as the feature quantity of the original image 101 in the subsequent step S109 for the feature quantity called dot density. And the dot pattern feature quantity of the output prohibited document are used for determining whether or not the same is recognized. Therefore, when the determination result in step S105 is affirmative, that is, between the dot density detected as the feature amount of the original image 101 and the feature amount of the dot pattern of the output-prohibited document for the feature amount of dot density. If it is determined that the identity is not recognized (N in step S105), the process in step S106 is omitted because it does not make sense to execute the process in step S106.

  The processing from step S107 and step S108 to the subsequent step S110 is performed between the dot density detected as the feature amount of the document image 101 and the feature amount of the dot pattern of the output prohibited document with respect to the feature amount called inter-dot distance. This is a process for determining whether or not identity is recognized. Therefore, in this embodiment, when determining whether or not the same is recognized between the dot density detected as the feature amount of the original image 101 and the feature amount of the dot pattern of the output prohibited document, a feature called dot density is used. Both feature quantities and feature quantities such as inter-dot distances are used.

  In step S107, the inter-dot distance d between adjacent dots 106 included in the detected dot pattern 103 is calculated by arithmetic processing in the CPU 204a. For example, the inter-dot distance d between adjacent dots 106 included in the detected dot pattern 103 calculated in step S107 is obtained by calculating the center coordinates of the two adjacent dots 106 by the calculation function of the CPU 204a. The distance between these central coordinate positions can be calculated. In step S107, such inter-dot distances d are calculated for a plurality of sets of adjacent dots 106, and the calculated inter-dot distances d are stored, for example, in the resist area of the RAM 204c.

  In step S108, the inter-dot distance frequency is calculated by a calculation process in the CPU 204a. The inter-dot distance frequency calculated in step S108 is, for example, the peak value and peak position obtained based on the plurality of sets of inter-dot distances d calculated in step S107. These peak values and peak positions are calculated by the calculation function of the CPU 204a. As is apparent from the graph of FIG. 9, the peak position obtained based on the plurality of sets of inter-dot distances d is estimated to be the actual inter-dot distances d for the plurality of dots 106 detected in step S103. Is done. In this case, the peak value indicates the degree to which the plurality of dots 106 detected in step S103 are concentrated at the peak position. For example, in the case of the dot pattern 103 illustrated in FIG. 10A, the dot-to-dot distance frequency for the plurality of dots 106 detected in step S103 is a value represented by the graph illustrated in FIG. Take. In this case, the peak value of the inter-dot distance frequency is a certain value or more. On the other hand, in the case of the dot pattern 103 illustrated in FIG. 10B, the dot-to-dot distance frequency for the plurality of dots 106 detected in step S103 is the distribution value shown in the graph illustrated in FIG. The distribution value is flatter than that, and the peak value is lower. Therefore, it is possible to determine whether the dot pattern 103 is the dot pattern 103 as illustrated in FIG. 10A or the dot pattern 103 as illustrated in FIG. 10B by referring to the peak value. Thus, the dot pattern 103 illustrated in FIG. 10B can be excluded from the determination target.

  As described above, in the present embodiment, the number of detection dots 106 is accumulated and temporarily stored, for example, in the registration area of the RAM 204c by the arithmetic processing of the CPU 204a (step S106). Further, the dot distance frequency composed of the peak value and the peak position of the dot distance is obtained by the arithmetic processing of the CPU 204a, and is temporarily stored, for example, in the registration area of the RAM 204c (step S108). Therefore, in the present embodiment, as the processing of the subsequent step S109, whether the number of detected dots 106 temporarily stored in the resist area of the RAM 204c in step S106 is within the threshold stored in the storage area of the RAM 204c, or In step S107, two types of determination processing are performed to determine whether the dot-to-dot distance frequency consisting of the peak-to-dot distance peak value and the peak position temporarily stored in the resist area of the RAM 204c is within the threshold value stored and stored in the storage area in the RAM 204c. Execute. As a result of the determination process in step S109, when the two types of determination results are both negative, that is, the number of detection dots 106 is not within the threshold value, the peak value and peak position of the inter-dot distance are used. If the inter-dot distance frequency is not within the threshold (N in step S109), the process is terminated. On the other hand, if one of the two types of determination results is affirmative as a result of the determination process in step S109, that is, whether the number of detection dots 106 is within the threshold or the distance between dots is When the dot-to-dot distance frequency including the peak value and the peak position is within the threshold (Y in step S109), the feature amount of the dot pattern 103 included in the document image 101 and the feature amount of the dot pattern of the output prohibited document Identity will be recognized between them. That is, as an example, any one of the dot pattern 103 including only the base area 104 of the dot pattern 103, the message area 105 of the dot pattern 103, and the dot 106 of a single size in the document image 101 read by the scanner 201. Or more than one. In this case, therefore, the process proceeds to the output prohibited document determination process in step S110.

  In the output-prohibited document determination process in step S110, the output-prohibited document determination process is executed according to a preset determination criterion. This determination criterion may be stored as a parameter in a storage area in the RAM 204c, for example, or may be temporarily set in such a storage area in the RAM 204c by an input from the operation display unit 205. In this way, in the output-prohibited document determination process in step S110, if one of the base area 104 and the message area 105 of the dot pattern 103 exists in the original image 101 read by the scanner 201, the output is prohibited as a confidential document. Judge as a document. Alternatively, as another example, if both the base area 104 and the message area 105 of the dot pattern 103 exist in the original image 101 read by the scanner 201, it is determined that the document is an output-prohibited document such as a confidential document. To do. Alternatively, as another example, when the dot pattern 103 exists in the document image 101 read by the scanner 201, it is determined that the document is an output prohibited document such as a confidential document.

  In this way, processing for detecting the feature amount of the dot pattern 103 for confidential document determination or output prohibition document determination included in the image data of the original image 101 from the image data, and the detected feature amount is the feature amount of the output prohibition dot pattern. And a process for comparing with the reference feature quantity stored in the storage area in the RAM 204c and a process for determining whether or not the detected feature quantity and the reference feature quantity are identical. Become. Thus, in the present embodiment, regardless of the type of document image, the document image is confirmed by confirming the identity between the detected feature value of the dot pattern 103 and the feature value of the dot pattern stored in the storage area. It is possible to determine whether output of the image data 101 should be prohibited.

  FIG. 17 is a flowchart showing another example of processing contents of the dot pattern detection unit 212a and the output prohibited document determination unit 212b executed by the computer 1001 based on computer software.

  First, the processing routine shown in FIG. 17 is executed at a certain time, and it is repeatedly determined whether or not the original image 101 is read by the scanner 201 and image data is input (step S201).

  When it is determined that the original image 101 is read and image data is input by the scanner 201 (Y in step S201), the input image data is copied into an image memory included in a memory represented by a comprehensive concept such as the RAM 204c, for example. (Step S202).

  On the other hand, when it is not determined that the original image 101 is read by the scanner 201 and the image data is input (N in step S201), whether the original image 101 is read by the scanner 201 and the image data is input. Wait for the decision.

  Then, the dot 106 is detected from the image data copied to such an image memory (step S203). As a detection method in this case, various conventional methods for detecting an image pattern by computer processing, for example, pattern matching can be used.

  Next, the specific dot pattern 103 is detected by the arithmetic function of the CPU 204a. That is, a reference dot pattern serving as a reference for determining an output-prohibited document such as a confidential document is stored in storage areas existing inside and outside the system controller 204, such as the HDD 216, the ROM 204b, and the RAM 204c. It is detected whether the same dot pattern 103 as the reference dot pattern is present in the image data of the original image 101 read by 201. As a detection method in this case, various conventional methods for detecting an image pattern by computer processing, for example, pattern matching can be used.

  Next, the dot pattern density within a certain area in the dot pattern 103 detected in step S204 is calculated by the arithmetic function of the CPU 204a (step S205). This calculation is performed by obtaining the number of dot patterns 103 detected in step S204 within a certain area.

  Next, a storage area existing inside and outside the system controller 204, such as the HDD 216, the ROM 204b, and the RAM 204c, stores a reference dot pattern density serving as a reference for determining an output prohibited document such as a confidential document. Therefore, in step S206, it is determined whether or not the dot pattern density obtained by the calculation process in step S205 matches the reference dot pattern density. As a result, if the determination result is affirmative (Y in step S206), for example, the number of dot patterns that match the storage area of the RAM 204c is accumulated (step S207), and the process proceeds to step S208, and the determination result is negative. If yes, the process proceeds to step S208 without performing step S207. In step S208, for example, it is determined whether the number of dot patterns accumulated in the storage area of the RAM 204c is within a certain threshold (step S208), and output prohibited document determination processing is executed according to the result (step S209). ).

  In this way, processing for detecting the feature amount (form of the dot pattern 103) of the dot pattern 103 for confidential document determination or output prohibited document determination included in the image data of the original image 101 from the image data, and the detected feature amount For example, a process for comparing with a reference feature amount (density per unit area, cumulative number) of a specific dot pattern 103 as a feature amount of an output-prohibited dot pattern, a detected feature amount, and a reference The process of determining whether or not the identity is recognized with the feature amount is executed. Thus, in the present embodiment, regardless of the type of document image, the document image is confirmed by confirming the identity between the detected feature value of the dot pattern 103 and the feature value of the dot pattern stored in the storage area. It is possible to determine whether output of the image data 101 should be prohibited.

(3) Pre-processing according to the type of reading operation selected by the scanner 201 The image processing apparatus according to the present embodiment uses a scanner 201 to perform a plurality of types of reading operations, that is, a document moving type reading operation and a sensor moving type reading. The operation can be selected. Then, with respect to the image data obtained by such a scanner 201, information in the image can be extracted with high accuracy regardless of which kind of reading operation the image reading operation is obtained. Yes.

  As a technique for this, in the present embodiment, for example, a pre-processing unit 210 that constitutes an image pre-processing unit that selectively executes one or more pre-processing on image data obtained by the scanner 201 is provided. An image pre-processing step is executed. Such an image preprocessing function by the image preprocessing means can be realized not only by the preprocessing unit 210 having a digital circuit configuration but also by processing of the computer 1001 in accordance with computer software. As such preprocessing, for example, noise removal processing and automatic density correction processing are selected.

  When performing the document movement type reading operation, if dust adheres to the reading window of the contact glass 4, vertical stripes may occur in the read image due to the influence of the dust. Therefore, in such a case, for example, after the preprocessing called noise removal processing is executed by the filter processing unit 206 having a digital circuit configuration, information extraction processing for determining whether or not specific information is included in the image data is executed. It is desirable to do. That is, an image caused by dust tends to appear as an image having a lower density (thin) than the original. Therefore, in the noise removal processing by the filter processing unit 206, when a thin line with low density is continuously streaked in the sub-scanning direction, the pixel value corresponding to the line is changed to the “white” level. Is. On the other hand, during the sensor movement type reading operation, noise caused by dust on the contact glass 4 appears only at each point, and thus does not become conspicuous noise. Therefore, the linear noise removal process is not performed.

  Further, even if the image data based on the same document image 101 is between the image data obtained by performing the document movement type reading operation and the image data obtained by performing the sensor movement type reading operation, A subtle contrast is produced in the background of the original image 101. Therefore, when automatic density correction is performed on the image data by the γ processing unit 208 of the preprocessing unit 210, the first density correction is performed on the image data obtained by performing the document movement type reading operation, and the sensor movement is performed. A second density correction different from the first density correction is performed on the image data obtained by performing the mold reading operation. In this case, it is preferable that the first density correction is lighter than the second density correction.

  What is important in the present embodiment is that the type of preprocessing corresponding to the type of image reading operation selected by the scanner 201 is selected, and the selected preprocessing is executed by the preprocessing unit 210 that functions as an image preprocessing unit. It is to let you. For example, when the document moving type reading operation is selected, execution of noise removal processing in the filter processing unit 206 is selected as preprocessing by the preprocessing unit 210, and automatic density correction executed in the γ processing unit 208 is performed. The first density correction is executed as the degree. On the other hand, when the sensor movement type reading operation is selected, the execution of the noise removal processing in the filter processing unit 206 is not selected as the preprocessing by the preprocessing unit 210 and is executed in the γ processing unit 208. Second density correction is executed as the degree of automatic density correction.

(4) Information extraction processing according to the type of reading operation selected by the scanner 201 In this embodiment, even image data obtained by any type of reading operation of a plurality of types of reading operations can be accurately performed. As an example of another technique that enables extraction of information in an image, an information extraction process corresponding to the image reading operation selected by the scanner 201 is performed on the dot pattern detection unit 212a that executes the function of the information extraction unit. We are also doing it. That is, during the document movement type reading operation, the distance between the document sheet 102 and a sensor (line sensor) (not shown) provided in the optical unit 5 of the scanner 201 is relatively constant, whereas during the sensor movement type reading operation. The image feature of the read image data is likely to fluctuate depending on whether the document sheet 102 is floating from the contact glass 4 or there is a pressure plate on the document sheet 102. Therefore, during the sensor movement type reading operation, as an example, the dot detection unit 251 (see FIG. 15) of the dot pattern detection unit 212a sets a plurality of thresholds for determining the presence or absence of a dot detection signal, and detects the dot pattern 103. I try to prevent leaks.

3 is a front view illustrating a document image. FIG. FIG. 6 is a schematic diagram illustrating an example of a copy of an original image in which a dot pattern embedded in the original image is seen floating. FIG. 10 is a schematic diagram illustrating another example of a copy of an original image in which a dot pattern embedded in the original image is raised. FIG. 3 is a schematic diagram illustrating an enlarged dot pattern illustrated in FIG. 2. It is a schematic diagram which expands and shows the dot pattern illustrated in FIG. FIG. 6 is a front view illustrating a document image and its dot pattern as another embodiment. FIG. 7 is a front view illustrating a document image different from FIG. 6 and its dot pattern. It is a schematic diagram which illustrates the dot pattern which makes the distance between dots d. It is a graph which expresses the distance distribution between adjacent dots, taking the distance between dots on the horizontal axis and taking the appearance frequency on the vertical axis. It is a schematic diagram which illustrates a dot pattern. It is a front view which illustrates a dot pattern as another embodiment. FIG. 12 is a schematic diagram illustrating an example of a document image having a dot pattern illustrated in FIG. 11. 1 is a block diagram showing hardware resources as an embodiment of an image processing apparatus having a digital copying machine configuration according to the present invention; FIG. It is a vertical side view of a scanner. FIG. 14 is a block diagram of a dot pattern detection unit shown in FIG. 13. It is a flowchart which shows an example of the flow of the process performed by the computer software corresponding to the process in the dot pattern detection part and output prohibition document determination part which are shown in FIG. It is a flowchart which shows another example of the flow of the process performed by the computer software corresponding to the process in the dot pattern detection part and output prohibition document determination part which are shown in FIG.

Explanation of symbols

103 Specific information, certain characteristics, dot arrangement (dot pattern)
201 Image input means (scanner)
214 Image pre-processing means (pre-processing unit)

Claims (3)

An image input means capable of selecting a plurality of types of image reading operations;
Information extraction means for extracting specific information contained in the image data for the image data obtained by the image input means;
And the information extraction means executes an information extraction process according to an image reading operation selected by the image input means,
The information extraction unit changes a threshold for extracting specific information included in the image data in accordance with the image reading operation selected by the image input unit,
Threshold value for extracting the specific information, Ri threshold der for determining the presence or absence of dot detection signal,
The plural types of image reading operations by the image input means include an operation of obtaining image data by moving the sensor with respect to the document, and an operation of obtaining image data by moving the document with respect to the sensor,
When an operation for obtaining image data is selected by moving the sensor with respect to the document, detection omission is prevented more than when an operation for obtaining image data is selected by moving the document with respect to the sensor. An image processing apparatus characterized by changing a threshold value. An image input process capable of selecting a plurality of types of image reading operations;
For the image data obtained in the image input step, an information extraction step for extracting specific information included in the image data;
And in the information extraction step, an image processing method for executing information extraction processing according to the image reading operation selected in the image input step,
The information extraction step changes a threshold for extracting specific information included in the image data according to the image reading operation selected in the image input step,
Threshold value for extracting the specific information, Ri threshold der for determining the presence or absence of dot detection signal,
The plurality of types of image reading operations in the image input step include an operation of obtaining image data by moving a sensor with respect to a document, and an operation of obtaining image data by moving a document with respect to the sensor,
When an operation for obtaining image data is selected by moving the sensor with respect to the document, detection omission is prevented more than when an operation for obtaining image data is selected by moving the document with respect to the sensor. An image processing method characterized by changing a threshold value. It is installed in a computer provided in an image processing apparatus having an image input function capable of selecting a plurality of types of image reading operations, and image data obtained by the image input function is selected in the computer by the image input function. A computer-readable computer program for executing an information extraction function for extracting specific information included in the image data in accordance with an image reading operation,
The information extraction function changes a threshold for extracting specific information included in the image data in accordance with the image reading operation selected by the image input function,
Threshold value for extracting the specific information, Ri threshold der for determining the presence or absence of dot detection signal,
The plurality of types of image reading operations in the image input function include an operation of obtaining image data by moving a sensor with respect to a document, and an operation of obtaining image data by moving a document with respect to the sensor,
When an operation for obtaining image data is selected by moving the sensor with respect to the document, detection omission is prevented more than when an operation for obtaining image data is selected by moving the document with respect to the sensor. A computer program characterized by changing a threshold value.

Images