JP4542976B2 - Image processing apparatus, image processing method, image processing program, and recording medium - Google Patents

Description

  The present invention relates to an image processing apparatus and an image processing method, and more particularly to an image processing technique for creating a question sentence from a document image.

There is a pen-type learning tool for improving memorization efficiency on the market. The black printed part to be memorized in textbooks, etc. is colored with a red (green) marker and covered with a transparent sheet of green (red), which is a complementary color of red (green). It is darkened to make it difficult to read the black printed part. However, in the case of the marker type that is commercially available, there is a drawback that when the black print portion described in the document is colored, the marker ink penetrates the document and stains the back side of the document. In addition, the colored portion is often uneven and visually unsightly. Furthermore, since the color density of the colored part of the marker is low, it is extremely difficult to make the black printed part colored in red (green) completely illegible under the perspective of a sheet with a green (red) transparent ground color. Have difficulty. Pens are also available that remove the attached marker when the items to be memorized have been memorized, but it is difficult to return to the state before coloring the marker.
Note that Patent Document 1 as a conventional technique includes a document including markings such as extracting a title from a document and acquiring an image of the original document when the image processing apparatus, the recording medium, and the program document are stored electronically. Various processing methods are disclosed.
Patent Document 2 uses a digital copying machine having a scoring and question sheet registration function to prepare a question sheet on which an answer is written, and enters different symbols in the question part and the answer part. A technique for dividing a question part and an answer part by using an OCR read by a copying machine is disclosed.
JP 2004-246597 A JP-A-8-186672

However, in the prior art disclosed in Patent Document 1, since the color to be marked is specified in advance, the writing instrument to be marked is not versatile, and the mark cannot be erased with an eraser or the like. There is a problem in terms.
Further, the prior art disclosed in Patent Document 2 is an invention that relies on the OCR technology, and there is a problem that it cannot be dealt with when the OCR misrecognizes characters.
In view of the above problems, the present invention provides an image processing apparatus capable of easily creating a question sentence in which a portion to be concealed is filled and an answer sentence of the concealment portion by marking the portion to be concealed on the document. And an image processing method.
Another object of the present invention is to provide an image processing apparatus and an image processing method capable of storing an original without making it dirty by marking with a writing tool or the like that can be erased with an eraser instead of special marking.

In order to solve this problem, the present invention provides an image reading unit that reads an original image having at least one mark, and extracts the mark from the image read by the image reading unit. Mark extraction means for storing position information of the mark, mark removal means for removing the mark from the read image read by the image reading means, and a mark removal image from which the mark has been removed by the mark removal means; A circumscribed rectangle extracting means for extracting a circumscribed rectangle of the read image by collating with the mark position stored by the mark extracting means, and a rectangle for storing a rectangular image corresponding to the rectangular area extracted by the circumscribed rectangle extracting means. Area storage that fills the coordinates corresponding to the rectangular area extracted by the image storage means and the circumscribed rectangle extraction means of the mark-removed image And means rude, and by instructing the rectangular area of the mark removal image filled by the Area Fill means, characterized in that within that region of the rectangular image is displayed.
The image processing apparatus of the present invention reads a document image in which a portion to be concealed is marked, and stores an image obtained by removing the marking from the image and the position information of the marking. Further, the circumscribed rectangle is extracted by comparing the marking position with the read image, and the circumscribed rectangle image is stored. Then, a question sentence is created by painting the circumscribed rectangle part, and the answer is displayed by clicking the part with a mouse or the like.

According to a second aspect of the present invention, vertical / horizontal determination means for determining whether the mark-removed image is vertical writing or horizontal writing, index adding means for attaching an index based on the appearance order of rectangular images stored in the rectangular image storage means, An image index is generated by converting the index into another character type, an image index is added to the rectangular image in a random order, and an answer image is generated by associating the index with the image index Answer image generating means, option / answer image generating means for generating an image together with the answer image using the rectangular image with the image index generated by the image index adding means as an option, and the rectangular area filled by the area filling means The index added by the index adding means A problem image generating means for generating a question image by drawing a white color, and a combining means for combining the option / answer image generated by the option / answer image generating means with the question image It is characterized by.
The present invention further determines whether the image from which the mark is removed is vertical writing or horizontal writing, and adds an index such as a number in the order of appearance of the rectangular image. Next, the index is converted into a character type such as a character to generate an image index, which is shuffled and added to a rectangular image, and an answer is created by associating the index with the image index. Then, the image index is shuffled and added to the rectangular image as an option, and a question image is created together with the answer.

According to a third aspect of the present invention, there is provided an image reading means for reading a document image having at least one mark and a sentence mark on which a sentence including the mark is marked, and the mark and the sentence mark from the image read by the image reading means. Mark extracting means for extracting and storing position information of the mark and sentence mark; mark removing means for removing the mark and sentence mark from the read image read by the image reading means; and A circumscribed rectangle extracting unit that extracts a circumscribed rectangle of the read image by comparing the mark removed image from which the mark and the sentence mark have been removed with the position of the mark and the sentence mark stored by the mark extracting unit, and the mark removed image Vertical / horizontal determination means for determining vertical writing or horizontal writing; Sentence information creation means for creating coordinate information of the start point and end point of the sentence mark, and a question image is generated by filling the areas corresponding to the answer start point and answer end point created by the sentence information creation means from the mark removal image And a synthesis means for extracting a rectangular area corresponding to the answer starting point and answer ending point from the mark-removed image and synthesizing it with the question image.
The present invention is the same as the first aspect until the circumscribed rectangle is extracted. After that, it is determined whether the image from which the mark is removed is vertical writing or horizontal writing, creates sentence information as a question sentence from the mark position, generates a filled image from the position information of the mark part in it, generates the question sentence and the answer. Synthesize.

An image reading step for reading a document image having at least one mark and a mark extraction step for extracting the mark from the image read by the image reading step and storing position information of the mark A mark removal step of removing the mark from the read image read by the image reading step, a mark removal image from which the mark has been removed by the mark removal step, and a mark position stored by the mark extraction step Circumscribing rectangle extracting step for extracting circumscribed rectangles of the read image by collating with each other, a rectangular image storing step for storing a rectangular image corresponding to the rectangular region extracted by the circumscribed rectangle extracting step, and Fill the coordinates corresponding to the rectangular area extracted by the circumscribed rectangle extraction step Includes a band filling step, and by instructing the rectangular region of the mark removal image filled by the Area Fill step, characterized in that the rectangular image is displayed.
The present invention has the same effect as that of the first aspect.
According to a fifth aspect of the present invention, a vertical / horizontal determination step for determining whether the mark-removed image is vertical writing or horizontal writing, an index adding step for adding an index based on the appearance order of the rectangular images stored in the rectangular image storage step, and the index Is converted into another character type to generate an image index, the image index is added to the rectangular image in a random order, and an answer image is generated by associating the index with the image index. An answer image generating step, an option / answer image generating step for generating an image together with the answer image using the rectangular image with the image index generated by the image index adding step as an option, and a rectangular area filled by the area filling step The index A problem image generation step of generating a problem image by drawing the index attached in the addition step in white; a synthesis step of combining the option / answer image generated by the option / answer image generation step and the question image; Is further provided.
The present invention has the same effect as that of the second aspect.

According to a sixth aspect of the present invention, there is provided an image reading step of reading a document image having a sentence mark in which at least one mark and a sentence including the mark are marked, and the mark and the sentence mark are extracted from the image read by the image reading step. A mark extraction step of extracting and storing position information of the mark and sentence mark; a mark removal step of removing the mark and sentence mark from the read image read by the image reading step; and the mark removal step A circumscribed rectangle extracting step of extracting a circumscribed rectangle of the read image by collating the mark removed image from which the mark and the sentence mark have been removed with the position of the mark and sentence mark stored in the mark extracting step, and the mark removed image Vertical or horizontal writing Corresponding to the answer start point and answer end point created by the sentence information creating step from the mark-removed image, and the sentence information creating step for creating the coordinate information of the start point and end point of the sentence mark And a synthesis step of generating a question image by filling a region to be extracted, extracting a rectangular region corresponding to the answer start point and the answer end point from the mark-removed image, and synthesizing it with the question image.
The present invention has the same effect as that of the third aspect.
A seventh aspect is characterized in that the image processing method according to any one of the fourth to sixth aspects is programmed so as to be controllable by a computer.
An eighth aspect is characterized in that the image processing program according to the seventh aspect is recorded in a computer-readable format.

  According to the present invention, a document image in which a portion to be concealed is marked is read, the image obtained by removing the marking from the image and the position information of the marking are stored, and the circumscribed rectangle is extracted by comparing the marking position with the read image. At the same time, the image of the circumscribed rectangle is memorized, and a question sentence is created by painting the circumscribed rectangle part, and the answer is displayed by clicking on the part with a mouse etc. Question sentences and answers can be created automatically.

Hereinafter, the present invention will be described in detail with reference to embodiments shown in the drawings. However, the components, types, combinations, shapes, relative arrangements, and the like described in this embodiment are merely illustrative examples and not intended to limit the scope of the present invention only unless otherwise specified. .
FIG. 1 is a block diagram showing an image processing apparatus according to an embodiment of the present invention. The image processing apparatus 100 is configured by a CPU 1 that controls the entire apparatus and realizes the image processing function of the present embodiment, a ROM, a RAM, and the like, stores a control program for the CPU 1, and executes the control program by the CPU 1. A memory 2 that holds information and the like that are created occasionally, a hard disk drive, a hard disk, and the like, and an image storage device 3 that stores and stores documents after image processing in the hard disk, and a scanner An image input device 4 configured to read an image such as a sheet original (included in a document) and store it in the memory 2, an external memory interface 5 which is an interface for connecting to an external CD-ROM, etc., a CRT, It consists of a liquid crystal display (LCD) or a printer, etc. It consists of an output device 6 that displays and prints the input original image, compressed image, binary image of the image processing result, and the like, and a keyboard, mouse, touch panel, etc., and information on image processing (including scanner operation information) And an input device 7 used for input.

  The image processing apparatus of the present invention is configured using a general-purpose processing apparatus such as a personal computer, and the image processing function is realized by software. A central processing unit (hereinafter referred to as CPU) 1 controls the entire apparatus and realizes the image processing function of this embodiment. The memory 2 is composed of a ROM, a RAM, and the like, and stores a control program of the CPU 1 and holds information and the like that are created when the control program is executed by the CPU 1. The image storage device 3 includes a hard disk drive, a hard disk, and the like, and stores and stores documents after image processing in the hard disk. The image storage device 3 is used to read a document recorded on the hard disk and store it in another storage device. The image input device 4 is configured by a scanner or the like, and reads an image such as a sheet original (included in a document) and stores it in the memory 2. The input device 7 includes a keyboard, a mouse, a touch panel, and the like, and is used for inputting information related to image processing (including scanner operation information). The output device 6 is configured by a CRT, a liquid crystal display (LCD), a printer, or the like, and displays and prints an original image, a compressed image, a binary image of an image processing result, or the like input from the image input device 4. . The external memory interface 5 is an interface for connecting to an external CD-ROM or the like. These are connected by an internal bus 8.

  FIG. 2A is a diagram showing an example of a document used for explaining the operation of the embodiment of the present invention. Hereinafter, it is assumed that the document 10 is read by the image input device 4. FIG. 2B is a diagram showing an example in which a portion to be concealed of the original 10 is marked. Then, as shown in FIG. 2B, for example, the areas marked with a red pen are set as a mark 101 “person”, a mark 102 “world company”, and a mark 103 “(independent creation)”, respectively. Any number of colors can be used, and marking with ink that disappears over time or with a pencil is convenient because no handwriting remains on the material.

  FIG. 3 is a flowchart for explaining the operation of the image processing apparatus according to the first embodiment of the present invention. First, a portion of the original 10 to be concealed is marked as shown in FIG. 2B (S201). Next, the marked material is read by an image reading device such as a scanner or a camera (S202). Hereinafter, the diagram of FIG. 2B is assumed to be handled as the input image 104. Moreover, the following similar results can be obtained by detecting the marked coordinates not only by the image reading apparatus but also by a coordinate detection apparatus such as a tablet. That is, an input image 104 can be created by displaying an electronic manuscript on a display and marking the electronic manuscript with a mouse or the like. The input image 104 can also be created by marking. Subsequently, the mark 101, the mark 102, and the mark 103 are extracted from the input image 104, and the extracted mark position 104, mark position 105, and mark position 106 are stored in the mark position storage buffer 107 in the memory 2 (S203). The mark extraction method uses an existing method described in Patent Document 1 or the like. Next, the mark 101, the mark 102, and the mark 103 are removed from the input image 104 (S204). The removal of the mark is also performed using the method described in Patent Document 1. Next, the input image 104 from which the mark has been removed is compared with the mark position stored in the mark position storage buffer 107, and a circumscribed rectangle is extracted from the input image 104 (S205). The circumscribed rectangle extraction is to find the minimum rectangle surrounding the character area included in the marking area. Since the circumscribed rectangle extraction method is also described in Patent Document 1 and the like, it is omitted here. The circumscribed rectangle information extracted here is stored in the rectangle storage buffer 108 in the memory 2. Information stored in the rectangular storage buffer 108 is as shown in FIG. Note that the start point and the end point represent the position coordinates (x, y) of the input image 104, and the start point and the end point represent the vertices of a rectangular diagonal line. In the coordinate system used in this embodiment, the upper left of the input image 104 in FIG. 2B is the origin, the right direction of the image is X, the lower direction is Y, and all character sizes are 10 * 10 for simplification. , Line breaks and spaces are not considered.

Subsequently, the rectangular images corresponding to the positions of the rectangular region 301, the rectangular region 302, and the rectangular region 303 in FIG. 4 stored in the rectangular storage buffer 108 are extracted from the input image 104 from which the marks have been removed, and the rectangular storage buffer 108 is extracted. (S206). At this time, there is a method in which the data corresponding to the rectangular area is not a rectangular image but character-recognized in the rectangular image by OCR or the like and stored in a text format. Then, an image of the input image 104 in which the coordinates corresponding to the rectangular areas 301, 302, and 303 stored in S205 are painted in black is stored as a material painted image 501 (S207).
Finally, the material filled image 501 is displayed on the monitor (S208). FIG. 5 is a diagram showing a material filled image 501. A rectangular area 301 is a material filled image 501a, a rectangular area 302 is a material filled image 501b, and a rectangular area 303 is a material filled image 501c. Then, when the user performs a mouse click within the stored rectangular area shown in FIG. 4 in the document fill image 501, a rectangular image corresponding to the rectangular area is displayed superimposed on the document fill image 501. For example, FIG. 6 is an example displayed when the coordinates (15, 15) are clicked with the mouse. That is, by clicking on the coordinates (15, 15) in the approximate center of the material filled image 501a with the mouse, the “person” 11 in the rectangular area 301 is displayed.

FIG. 7 is a flowchart for explaining the operation of the image processing apparatus according to the second embodiment of the present invention. First, the part of the original 10 to be hidden is marked as shown in FIG. 2B (S701). Next, the marked material is read by an image reading device such as a scanner or a camera (S702). Hereinafter, the diagram of FIG. 2B is assumed to be handled as the input image 104. In addition, the following similar result can be obtained by detecting the marked coordinates not only by the image reading apparatus but also by a coordinate detection apparatus such as a tablet, but the description thereof is omitted. Subsequently, the mark 101, the mark 102, and the mark 103 are extracted from the input image 104, and the extracted mark position 104, mark position 105, and mark position 106 are stored in the mark position storage buffer 721 in the memory 2 (S703). The mark extraction method uses an existing method described in Patent Document 1 or the like. Next, the mark 101, the mark 102, and the mark 103 are removed from the input image 104 (S704). The removal of the mark is also performed using the method described in Patent Document 1. Next, the input image 104 from which the mark has been removed is compared with the mark position stored in the mark position storage buffer 721, and a circumscribed rectangle is extracted from the input image 104 (S705). The circumscribed rectangle extraction is to find the minimum rectangle surrounding the character area included in the marking area. Since the circumscribed rectangle extraction method is also described in Patent Document 1 and the like, it is omitted here. The circumscribed rectangle information extracted here is stored in the rectangle storage buffer 722 in the memory 2. Information stored in the rectangular storage buffer 722 is as shown in FIG. Note that the start point and the end point represent the position coordinates (x, y) of the input image 104, and the start point and the end point represent the vertices on a diagonal line of the rectangle. In the coordinate system used in this embodiment, the upper left of the input image 104 in FIG. 2B is the origin, the right direction of the image is X, the lower direction is Y, and all character sizes are 10 * 10 for simplification. , Line breaks and spaces are not considered.
Subsequently, rectangular images corresponding to the positions of the rectangular area 801, the rectangular area 802, and the rectangular area 803 in FIG. 8A stored in the rectangular storage buffer 722 are extracted from the input image 104 from which marks are removed, and the rectangular image is extracted. The data is stored in the storage buffer 722 (S706). At this time, there is a method in which the data corresponding to the rectangular area is not a rectangular image but character-recognized in the rectangular image by OCR or the like and stored in a text format.

Next, it is determined whether the input image 104 is written vertically or horizontally (S707). Although this determination includes a general-purpose method used in OCR or the like, it can also be determined from the rectangular shape stored in the rectangular storage buffer 722. INDEX is attached to the rectangular image 801, the rectangular image 802, and the rectangular image 803 stored in the rectangular storage buffer 722. As for the order of INDEX, in horizontal writing, INDEX is given from the smallest Y coordinate of the starting point, and when equal, the X coordinate is smallest. In the case of vertical writing, INDEX is given from the smaller X coordinate of the starting point, and in the case of equality, the Y coordinate is smaller. In this embodiment, 1 is added to the rectangular image 801, 2 is added to the rectangular image 802, and 3 is added to the rectangular image 803, and the result is stored in the rectangular storage buffer 722 (S708). Subsequently, as shown in FIG. 8B, INDEX 1, 2, and 3 are set to other character types (in this embodiment, katakana) to form an image INDEX. Then, the image INDEX is shuffled. In this embodiment, the image INDEX corresponding to the rectangular image 801 is a, the image INDEX corresponding to the rectangular image 802 is c, and the image INDEX corresponding to the rectangular image 803 is a.
Next, the image INDEX is imaged. First, the background is white, and INDEX is pasted as a black image thereon. Of course, it is better to calculate and use the background color and the character color when calculating the circumscribed rectangular area in S705. Next, the image INDEX is combined to the left and the rectangular image to the right to create a rectangular image with the image INDEX (S709). Of course, when the input image 104 is vertically written, the image INDEX image is up and the rectangular image is down. In this embodiment, a period is also imaged in order to clarify the distinction between the image INDEX and the rectangular image.

Subsequently, an answer image is created from the images INDEX and INDEX having a correspondence relationship from the table of FIG. As in the case where the INDEX is imaged, the background is white, and the black images INDEX and INDEX are pasted on it. At this time, it is preferable to put a hyphen or the like between them so that the correspondence between the image INDEX and INDEX is easily understood. In step S <b> 707, since the input image 104 is determined to be horizontal writing, the horizontal image size of the option / answer image is set to be the same as the horizontal image size of the input image 104. Then, in the same manner as in S709, the background color is white, the character “option” is imaged and pasted, and then the rectangular image with the image INDEX is pasted in the order of the image INDEX. Subsequently, the characters “answer” are imaged and pasted, and then the answer images are pasted in INDEX order to create an option / answer image 1002 (S710). FIG. 9A shows an option / answer image thus created. Since the option / answer image in FIG. 9A has a large margin on the right side, the layout as shown in FIG. 9B can be made in consideration of the horizontal size of each indexed image.
An image in which the coordinate positions corresponding to the rectangular areas 801, 802, and 803 saved in S705 in the input image 104 are filled with black is used as a material filled image (see FIG. 10), and is further managed by the rectangular storage buffer 722. INDEX is drawn in white in a rectangle filled with black (S711). The image at that time is defined as a problem image 1001. Finally, the image is synthesized with the question image 1001 up and the option / answer image 1002 down (S712). It is further preferable to insert a dotted line 12 or the like as shown in FIG. 11 so that the question image 1001 and the option / answer image 1002 can be distinguished when they are combined.

FIG. 13 is a flowchart for explaining the operation of the image processing apparatus according to the third embodiment of the present invention. First, on the material as shown in FIG. 2A, a portion to be memorized by the user is marked with a red pen, for example, as shown in FIG. Subsequently, a sentence including a portion to be memorized is marked with a blue pen, for example, to create a sentence mark 1203, a sentence mark 1204, and a sentence mark 1205 (S1301). In the present embodiment, the question mark is marked with a red pen and the sentence mark is marked with a blue pen. Moreover, since the question mark is always a part of the sentence mark, a pen of the same color may be used. In particular, marking with ink that disappears over time or with a pencil is convenient because no handwriting remains on the material. Next, the marked material is read by an image reading device such as a scanner or a camera (S1302). Hereinafter, FIG. 12 is assumed to be handled as an image read by the image reading apparatus and an input image 1250.
Next, the question marks 1201, 1202 and sentence marks 1203, 1204, 1205 are extracted, and the extracted question marks 1206, 1207 and sentence marks 1208, 1209, 1210 are stored in the mark position storage buffer 1321 in the memory 2 (S1303). ). As the mark extraction method, an existing method described in Patent Document 1 is used. Next, the question marks 1201, 1202 sentence marks 1203, 1204, 1205 are removed from the input image 1250 (S1304). The removal of the mark is also performed using the method described in Patent Document 1.
Then, the input image 1250 from which the mark has been removed is compared with the mark position stored in the mark position storage buffer 1321, and circumscribed rectangle extraction is performed from the input image 1250 (S1305). The circumscribed rectangle extraction is to find the minimum rectangle surrounding the character area included in the marking area. Since the circumscribed rectangle extraction method is also described in Patent Document 1 and the like, it is omitted here. The circumscribed rectangle information extracted here is stored in the rectangle storage buffer 1322 in the memory 2. Information stored in the rectangular storage buffer 1322 is as shown in FIG. The question marks 1201 and 1202 correspond to the question rectangles 1401 and 1402, and the sentence marks 1203, 1204 and 1205 correspond to the sentence rectangles 1403, 1404 and 1405. Note that the start point and the end point represent the position coordinates (x, y) of the input image 1250, and the start point and the end point represent the vertices on the diagonal line of the rectangle. In the coordinate system used in the present embodiment, the upper left of the input image 1250 is the origin, the right direction of the image is X, the lower direction is Y, all character sizes are 10 * 10 for simplification, Not considered.

  Next, it is determined whether the input image 1250 is written vertically or horizontally (S1306). Although this determination includes a general-purpose method used in OCR or the like, it can also be determined from the rectangular shape stored in the rectangular storage buffer 1322. In the case of this embodiment, it is determined as horizontal writing. Next, the correspondence between the question rectangle and the sentence rectangle is examined (S1307). From the start point and end point of the question rectangle, and from the start point and end point of the sentence rectangle, it is understood that the question rectangle 1401 is a part of the minute rectangle 1403 and the question rectangle 1402 is a part of the minute rectangle 1404. Since the rectangular image and the rectangular sentence image have a one-to-one relationship, the sentence rectangle 1405 is the same sentence as the sentence rectangle 1403 or the sentence rectangle 1404. In S1306, it can be seen that the input image 1250 is horizontal writing. Therefore, if the Y-axis coordinates of the sentence rectangle 1403 and the sentence rectangle 1404 are closer, and if the Y-axis coordinates of the sentence rectangle 1403 and the sentence rectangle 1404 are equal, the end X The same sentence as the one with larger coordinates. In this embodiment, it can be determined that the sentence rectangle 1404 and the sentence rectangle 1405 are the same sentence. Using these, sentence information is created (see FIG. 15).

Next, a problem is created using the rectangular storage buffer 1322 and the input image 1250 (S1308). First, specify the problem image size. Any number of problem image sizes may be used, but if the same image horizontal size as the input image 1250 is set, a beautiful problem paper can be created. Then, a problem sheet is created from the problem information (FIG. 15) stored in the rectangular memory buffer. First, the area corresponding to the start point 1 and the end point 1 of the sentence 1501 is extracted from the input image 1250 with a mark, and the area corresponding to the answer start point and the answer end point is painted in black. Then attach it to the appropriate position of the problem image. There is a method of automatically determining the attachment position on the machine side, but a method of specifying by the user is also conceivable. Subsequently, an area corresponding to the start point 1 and the end point 1 and an area corresponding to the start point 2 and the end point 2 of the sentence 1502 are extracted from the marked input image 1250, and the areas corresponding to the answer start point and the answer end point are filled with black. In the case of a sentence extending over two lines like the sentence 1502, a line break is necessary, so that it is attached to an appropriate position in the same manner as the sentence 1501 while appropriately adding a line break in consideration of the problem image size (FIG. 16).
Finally, the rectangular areas indicated by the answer start point and answer end point of the sentences 1501 and 1502 are extracted from the marked input image 1250 and attached to the lower part of the question image. Of course, a better problem is created by adding an indexing step as in the second embodiment.

As described above, according to the present invention, a document image in which a portion to be concealed is marked is read, an image obtained by removing the marking from the image is stored, and marking position information is stored. Extract the rectangle, memorize the image of the circumscribed rectangle, create a question sentence by painting the circumscribed rectangle, and click on the portion with the mouse to display the answer. You can automatically create question sentences and answers.
In addition to the above, it is determined whether the image from which the mark is removed is vertical writing or horizontal writing, an index such as a number is added in the order of appearance of the rectangular image, and the index is converted into a character type such as a character to generate an image index. Then, shuffle it and add it to the rectangular image, create an answer by associating the index with the image index, select the option that shuffled the image index and added it to the rectangular image, and create the question image along with the answer , You can answer by selecting the answer from the options.
Also, after extracting the circumscribed rectangle, it is determined whether the image with the mark removed is vertical writing or horizontal writing, creating sentence information as a problem sentence from the mark position, and generating a filled image from the position information of the mark part in it Since the question sentence and the answer are combined, it is possible to extract the part to be the question sentence from the input image and create the question and the answer.
Further, by programming the image processing method of the present invention in accordance with an OS that can be controlled by a computer, any computer equipped with the OS can be controlled by the same processing method.

1 is a block diagram illustrating an image processing apparatus according to an embodiment of the present invention. (A) is a figure which shows an example of the original document used in order to demonstrate operation | movement of embodiment of this invention, (b) is a figure which shows an example which marked the part to conceal of the original document 10. FIG. 5 is a flowchart for explaining the operation of the image processing apparatus according to the first embodiment of the present invention. It is a figure which shows the format of the information stored in a rectangular storage buffer. It is a figure which shows a material filling image. It is a figure which shows the example displayed when a coordinate is clicked with the mouse | mouth. It is a flowchart explaining operation | movement of the image processing apparatus which concerns on the 2nd Embodiment of this invention. (A) (b) is a figure which shows the format of the information stored in a rectangular memory buffer. (A) and (b) are figures which show an example of an option and an answer image. It is a figure which shows the image which painted the coordinate position corresponding to a rectangular area in black. It is a figure which shows an example of a question image and an option and an answer image. It is a figure which shows an example of a question mark and a sentence mark. It is a flowchart explaining operation | movement of the image processing apparatus which concerns on the 3rd Embodiment of this invention. It is a figure which shows the format of the information stored in a rectangular storage buffer. It is a figure which shows an example of sentence information. It is a figure which shows an example of the synthesized image of a question sentence and an answer.

Explanation of symbols

  1 CPU, 2 memory, 3 image storage device, 4 image input device, 5 external memory interface, 6 output device, 7 input device, 8 bus, 100 image processing device

Claims (8)

Image reading means for reading a document image having at least one mark, mark extraction means for extracting the mark from the image read by the image reading means, and storing position information of the mark, and the image reading A mark removing means for removing the mark from the read image read by the means, and a mark removal image from which the mark has been removed by the mark removing means and a mark position stored by the mark extracting means A circumscribed rectangle extracting unit that extracts a circumscribed rectangle of the read image, a rectangular image storage unit that stores a rectangular image corresponding to the rectangular area extracted by the circumscribed rectangle extracting unit, and a circumscribed rectangle extracting unit of the mark removal image An area filling means for filling the coordinates corresponding to the extracted rectangular area,
An image processing apparatus characterized in that a rectangular image in the area is displayed by designating the rectangular area of the mark-removed image painted by the area filling means.   Vertical / horizontal determination means for determining whether the mark-removed image is vertical writing or horizontal writing, index adding means for adding an index based on the appearance order of rectangular images stored in the rectangular image storage means, and the index as another character type An image index adding means for generating an image index by converting the image index into a random order, and an answer image generating means for generating an answer image by associating the index with the image index An option / answer image generating means for generating an image together with the answer image using the rectangular image with the image index generated by the image index adding means as an option, and the index adding means for the rectangular area filled by the area filling means The index attached by A problem image generating means for generating a question image by drawing; and a combining means for combining the option / answer image generated by the option / answer image generating means with the question image. The image processing apparatus according to claim 1.   Image reading means for reading a document image having a sentence mark in which at least one mark and a sentence including the mark are marked, and extracting the mark and sentence mark from the image read by the image reading means, And mark extraction means for storing the position information of the sentence mark, mark removal means for removing the mark and sentence mark from the read image read by the image reading means, and the mark and sentence mark are removed by the mark removal means. The circumscribed rectangle extracting means for extracting the circumscribed rectangle of the read image by comparing the position of the removed mark removed image with the mark and sentence mark stored by the mark extracting means, and whether the mark removed image is written vertically or horizontally Vertical / horizontal determination means for determining the sentence and the sentence Sentence information creation means for creating coordinate information of the start point and end point of the mark, and a question image is generated by filling the area corresponding to the answer start point and answer end point created by the sentence information creation means from the mark removal image An image processing apparatus comprising: synthesis means for extracting a rectangular area corresponding to the answer start point and answer end point from the mark-removed image and synthesizing it with the question image. An image reading step for reading a document image having at least one mark, a mark extraction step for extracting the mark from the image read by the image reading step, and storing positional information of the mark, and the image reading step A mark removal step for removing the mark from the read image read by the step, a mark removal image from which the mark has been removed by the mark removal step, and a mark position stored by the mark extraction step, A circumscribed rectangle extracting step for extracting a circumscribed rectangle of the read image, a rectangular image storing step for storing a rectangular image corresponding to the rectangular area extracted by the circumscribed rectangle extracting step, and a circumscribed rectangle extracting step for the mark removal image Area fill that fills the coordinates corresponding to the extracted rectangular area Includes a step, the,
An image processing method according to claim 1, wherein the rectangular image is displayed by indicating the inside of the rectangular area of the mark-removed image that has been painted by the area painting step.   A vertical / horizontal determination step for determining whether the mark-removed image is vertical writing or horizontal writing, an index adding step for adding an index based on the appearance order of the rectangular images stored in the rectangular image storing step, and the index as another character type An image index is added to the rectangular image by generating an image index by conversion, and the image index is assigned in a random order; an answer image generating step for generating an answer image in association with the index and the image index; An option / answer image generating step for generating an image together with the answer image using the rectangular image with the image index generated by the image index adding step as an option; and the index adding step for the rectangular area filled by the area filling step. A problem image generation step of generating a question image by drawing the index attached in white, and a synthesis step of combining the question image with the option / answer image generated by the option / answer image generation step The image processing method according to claim 4, comprising: an image processing method according to claim 4.   An image reading step of reading a document image having a sentence mark in which at least one mark and a sentence including the mark are marked; and the mark and the sentence mark are extracted from the image read by the image reading step; And a mark extracting step for storing position information of the sentence mark, a mark removing step for removing the mark and the sentence mark from the read image read by the image reading step, and the mark and the sentence mark by the mark removing step. A circumscribed rectangle extraction step of extracting a circumscribed rectangle of the read image by comparing the position of the removed mark removed image with the mark and sentence mark stored in the mark extracting step, and whether the mark removed image is written vertically or horizontally Determine An area corresponding to the answer start point and answer end point created by the sentence information creating step from the mark removal image, and a sentence information creating step for creating coordinate information of the start point and end point of the sentence mark; An image processing method comprising: a synthesis step of generating a question image by painting, extracting a rectangular area corresponding to the answer start point and answer end point from the mark-removed image, and synthesizing with the question image.   An image processing program in which the image processing method according to claim 4 is programmed so as to be controllable by a computer.   A recording medium in which the image processing program according to claim 7 is recorded in a computer-readable format.

Images