JP4136918B2 - Image processing apparatus, image processing method, and image processing program - Google Patents

Description

  The present invention relates to an image processing apparatus, an image processing method, and an image processing program for performing predetermined processing, for example, painting-like processing, on image data.

2. Description of the Related Art Conventionally, various image processing processes such as a color conversion process for changing a pixel color and a distortion conversion process for distorting an image are performed on a read image data by a scanner. For example, from the input image data and the result of dividing the image data into regions, the pixels are divided into groups for each divided region, group features are calculated and stored for each group, and based on the stored group features By processing the group features with reference to the group features of the surrounding groups, and constructing new pixels from the processed group features, the pixel data corresponding to the stroke of the painting is used as the unit of conversion for the image data. An image processing / synthesizing apparatus has been developed that automatically assigns new and abundant expressions (see Patent Document 1). In this case, since the features are grouped for each group, the features are processed in units of the group, and the pixels are assigned to the group so as to save the processed features, so that the group can be regarded as a brush. It is possible to process an original image such as a photographic image into various brush-touch images.
JP-A-10-11569 (page 3-9, FIG. 1-14)

  However, in the above-described image processing / synthesizing apparatus, the user can express his / her style of painting, but even if he likes the processed image processed by others, the processing procedure for creating the processed image from the original image is not known. For this reason, it is difficult to process an image so that it becomes a style of a work processed by others. In particular, for beginners (amateurs) who do not have sufficient knowledge to use the image processing / synthesizing device, the image processing / synthesizing device cannot be used. It cannot be processed.

  An object of the present invention is to provide an image processing apparatus, an image processing method, and an image processing program capable of easily processing an image so that the processed image is styled when someone likes a processed image processed by others. It is.

An image processing apparatus according to the present invention is an image processing apparatus that performs image-like processing on image data, and includes a target image to be subjected to image processing and a processed image created by performing image processing on an original image. Corresponding processing description describing the processing procedure when creating a processed image from the original image and processing description specifying information for specifying the processing description, the acquisition means for acquiring the processed image specifying information for specifying stores in advance Te, a specific information storage means for storing in advance in association with processing image identification information of processed image created by the processing described specified by the process description specifying information and the processing description specification information, the specific by referring to the information storage means to identify the processing description specifying information associated with the acquired processed image identification information by the acquisition unit, identified by the specified processing description specification information And image processing means for image processing the target image acquired by the acquisition unit using a physical description, the image processing means, of the group area for each group area by dividing the target image into a plurality of groups regions Based on the feature quantity calculation means for calculating the feature quantity, the group feature storage means for storing the feature quantity calculated by the feature quantity calculation means for each group area, and the feature quantity stored in the group feature storage means An integration unit that integrates a plurality of group regions to create a plurality of integration regions, an integration region storage unit that stores the integration region integrated by the integration unit, and an integration region that is stored in the integration region storage unit A processing unit that performs an image processing process on the target image as a unit, and the integration unit uses the feature amount stored in the group feature storage unit, A region in which the difference in feature amount between adjacent group regions is smaller than the first threshold and the number of group regions is equal to or greater than the first number is integrated as a first integrated region, and is further stored in the group feature storage unit. A feature amount difference between adjacent group regions is smaller than the first threshold, the number of group regions is equal to or greater than a second number less than the first number, and an integrated region A second threshold whose average value of a difference between a group region feature amount of a predetermined ratio or more and a group region feature amount in an adjacent integrated region is equal to or greater than the first threshold value The larger area is integrated as the second integrated area .

In the image processing apparatus according to the present invention, a target image to be subjected to image processing and processed image specifying information for specifying a processed image created by performing image processing on the original image are acquired, and the processed image is acquired from the original image. The process description describing the process procedure for creating the process is stored in advance in association with the process description specifying information for specifying the process description, and the process specified by the process description specifying information and the process description specifying information A process for specifying and specifying the process description specifying information associated with the processed image specifying information with reference to the specific information storage means that stores the processed image specifying information of the processed image created in accordance with the description in advance Since the target image is image-processed using the process description specified by the description specifying information, the processed image specifying information of the processed image processed by another person is used as the style of the processed image. Uni target image can be automatically processed, can be easily processed image so that the painting style of the processed image when you like is processed image by others processed.
Further, the target image is divided into a plurality of group areas, the feature amount of the group area is calculated for each group area, the calculated feature amount is stored for each group area, and a plurality of features are stored based on the stored feature quantities. Group areas are integrated to create multiple integrated areas, and image processing is performed using the integrated areas as a unit, so you can extract areas with a high degree of unity and contrast that are necessary for design, A region having gradation can be integrated into one region, and when a plurality of expressions are given to the target image, the areas to which each expression is given can be extracted simply and prioritized without unnaturalness.
Furthermore, using the stored feature values, a region where the difference between the feature values of adjacent group regions is equal to or smaller than the first threshold and the number of group regions is equal to or greater than the first number is defined as a first integrated region. Since they are integrated, an area with a high sense of unity in a wide range can be extracted as the first integrated area, and an area with gradation can be integrated into one area. Further, using the stored feature amount, a difference in feature amount between adjacent group regions is smaller than a first threshold, and the number of group regions is equal to or more than a second number smaller than the first number, and A second value in which an average value of a difference between a group region feature amount of a predetermined ratio or more among a plurality of group regions located at the boundary of the integration region and a group region feature amount in an adjacent integrated region is equal to or greater than a first threshold value; Since the region larger than the threshold is integrated as the second integrated region, a region with a high contrast effect in a narrow range can be extracted as the second integrated region.

  The image processing device further includes processed image generation means for generating processed image specifying information of the processed image created by the image processing means, and the specific information storage means is processed image specifying information generated by the processed image generating means. Is preferably stored in association with the process description specifying information of the process description used when creating the processed image.

  In this case, processed image specifying information for specifying the processed image is generated with respect to the generated processed image, and the generated processed image specifying information is the processing description specifying information of the processing description used when the processed image is created. Since the correspondence is newly stored, the created processed image can be automatically made into a database.

  The acquisition unit acquires the processed image created by the user and the processing description used when the processed image is created, and the image processing device generates processed image specifying information of the processed image acquired by the acquisition unit, and The processing description generating means for generating the processing description specifying information of the processing description acquired by the acquiring means is further provided, and the specifying information storage means uses the processed image specifying information generated by the processing description generating means as the processed image specifying information. It is preferable to newly store the information in association with the process description specifying information generated together.

  In this case, the processed image created by the user and the processing description used when creating the processed image are acquired, and processed image specifying information for specifying the processed image is generated for the acquired processed image. In addition, processing description specifying information for specifying the processing description is generated for the acquired processing description, and the generated processed image specifying information is associated with the processing description specifying information generated together with the processed image specifying information. Therefore, the processing description of the processed image created by the user can be automatically made into a database.

  The acquisition unit acquires the processed image created by the user and the processing description integrated with the processed image, and the generation unit for processing description generates processed image specifying information of the processed image acquired by the acquiring unit. In addition, it is preferable to extract a process description from the processed image and generate process description specifying information of the extracted process description.

  In this case, the processed image created by the user and the processing description integrated with the processed image are acquired, and processed image specifying information for specifying the processed image is generated for the acquired processed image. At the same time, the processing description is extracted from the processed image, and processing description specifying information for specifying the processing description is generated for the extracted processing description. Therefore, be aware of the distinction between the processed image and the processing description. Therefore, the processed image and the processing description can be handled, and the burden on the user can be reduced.

  It is preferable that the image processing apparatus includes a server computer that is communicably connected to the client computer via a predetermined network, and the acquisition unit acquires the target image selected by the user and the processed image specifying information from the client computer.

  In this case, the user can use the computer he or she uses as a client computer and receive an image processing service by a server computer, so that the user can easily enjoy the image processing service using a public network such as the Internet as a network. be able to.

  The integration unit preferably repeats the process of creating the first and second integrated regions a plurality of times, and increases the first threshold according to the number of repetitions. In this case, since the first threshold value increases in accordance with the number of repetitions of the process, it is possible to reliably extract a region with a high sense of unity and a region with a high contrast effect necessary for the design.

  The integration unit preferably decreases the first number according to the number of repetitions. In this case, since the first number decreases according to the number of repetitions, it is possible to more reliably extract a region with a high sense of unity necessary for the design.

  It is preferable that the integration unit creates the second integrated area from the group areas that are not integrated after calculating the first integrated area for each repetition process. Moreover, it is preferable that the integrated area storage means stores intermediate results of repeated integration processes, including all or a part of the number of repetitions.

  The feature amount preferably includes an average color for each group region, a position on the image for each group region, and the number of pixels for each group. In this case, since an integrated area can be created according to the color difference using the average color for each group area as a feature quantity, it is possible to accurately extract areas with a high unity and a high contrast effect necessary for the design. .

An image processing method according to the present invention is an image processing method for performing image- like processing on image data using a computer, wherein the computer performs image processing on a target image to be subjected to image processing and an original image. and an acquisition step of acquiring and processing image specifying information for specifying the processing image created by the computer, a processing description that describes a processing procedure for creating a processed image from the original image, the process described stores in advance in association with the processing description specification information for specifying a correspondence between processed image specifying information processed image created by the processing described specified by the process description specifying information and the processing description specification information referring to the specific information storage means for storing in advance Te, the obtained processed image processing description specific information associated with the identification information in the acquiring step JP And, viewed contains an image processing step of the image processing target image obtained in said obtaining step by using the process described specified by the identified processing description specification information, the image processing step, a target image a plurality of groups regions A feature amount calculating step for calculating the feature amount of the group region for each group region, a group feature storing step for storing the feature amount calculated in the feature amount calculating step for each group region, and the group feature An integration step of creating a plurality of integrated regions by integrating a plurality of group regions based on the feature values stored in the storage step, an integrated region storage step of storing the integrated regions integrated in the integration step, and the integration Image processing is performed on the target image in units of the integrated region stored in the region storage step. And the integration step uses the feature amount stored in the group feature storage step, and the feature amount difference between adjacent group regions is smaller than a first threshold and the number of group regions Are integrated as a first integrated region, and the feature amount stored in the group feature storage step is used to calculate a difference in feature amount between adjacent group regions. Less than the threshold value, and the number of group areas is equal to or greater than a second number smaller than the first number, and adjacent to the feature amount of the group area of a predetermined ratio or more among a plurality of group areas located at the boundary of the integrated area A step of integrating, as a second integrated region, a region in which the average value of the differences from the feature amounts of the group regions in the integrated region is larger than the second threshold equal to or greater than the first threshold. Including

An image processing program according to the present invention is an image processing program for performing painting-like processing on image data, and is a processing created by subjecting a target image to be subjected to image processing and an original image to image processing. An acquisition means for acquiring processed image specifying information for specifying an image and a processing description describing a processing procedure for creating a processed image from an original image are used as processing description specifying information for specifying the processing description. stores in advance in association, a specific information storage means for storing in advance in association with processing image identification information of processed image created by the processing described specified by the process description specifying information and the process description specifying information, identify the process description specifying information associated with the processed image specifying information acquired by referring to the acquisition means to the specifying information storing means, the specified process description certain information Cause the computer to function as an image processing means for image processing the target image acquired by the acquisition unit using a processing description specified by the image processing means, a group by dividing the target image into a plurality of groups area region A feature amount calculating means for calculating the feature amount of the group area for each group; a group feature storing means for storing the feature amount calculated by the feature amount calculating means for each group area; An integration unit that integrates a plurality of group regions based on the feature quantity, creates a plurality of integrated regions, an integrated region storage unit that stores the integrated region integrated by the integration unit, and stores the integrated region in the integrated region storage unit Processing means for performing an image processing process on the target image in units of the integrated region, wherein the integration means includes the group feature description. Using the feature amount stored in the means, a region where the difference in feature amount between adjacent group regions is smaller than a first threshold and the number of group regions is equal to or greater than the first number is defined as a first integrated region. Further, using the feature quantities stored in the group feature storage means, the difference in feature quantities between adjacent group areas is smaller than the first threshold, and the number of group areas is the first number. The average of the difference between the feature amount of the group region in the adjacent integrated region and the feature amount of the adjacent integrated region among the plurality of group regions located at the boundary of the integrated region that is equal to or smaller than the second number A region whose value is greater than a second threshold value equal to or greater than the first threshold value is integrated as a second integrated region .

  According to the present invention, the target image can be automatically processed using the processed image specifying information of the processed image processed by another person so that the processed image becomes the style of the processed image. In this case, the image can be easily processed so that the image of the processed image is obtained.

  Hereinafter, an image processing server which is an image processing apparatus according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a configuration of an image processing system using an image processing server according to an embodiment of the present invention.

  The image processing system shown in FIG. 1 includes a client computer 1 and an image processing server 2, and the client computer 1 and the image processing server 2 are connected to each other via a network 3 so that they can communicate with each other. The client computer 1 is used for transmitting a target image to be processed by the user and processed image specifying information for specifying a processed image created by performing image processing on the original image. The target image and the processed image specifying information are transmitted to the image processing server 2 via the network 3.

  Here, various image data can be used as the target image, and image data expressed in the RGB color system, XYZ color system, L * a * b * color system, or the like can be used. Alternatively, a plurality of pieces of image data may be used. The target image includes image data and photographic image data created by the user, the processed image includes image data processed by others other than the user, and the processed image specifying information provides the image processing service. Including a unique ID number that is assigned to the processed image and managed by the operator or the like, and is displayed together with the processed image in an actual exhibition or a predetermined homepage, for example. Therefore, when the user likes the processed image displayed together with the processed image specifying information, the user can acquire the processed image specifying information of the favorite processed image by storing the processed image specifying information.

  The image processing server 2 acquires the target image and the processed image specifying information, specifies the processing description describing all the processing procedures when creating the processed image from the original image from the processed image specifying information, and uses the specified processing description. After subjecting the target image to image processing using all the described processing procedures, the processed image is transmitted to the client computer 1 via the network 3. The client computer 1 displays the received processed image and the user can obtain the processed image.

  As the network 3, for example, the Internet is used, and data and the like are transmitted and received between the client computer 1 and the image processing server 2 according to TCP / IP (Transmission Control Protocol / Internet Protocol). The network 3 is not particularly limited to the Internet, and may be another network such as an intranet, or a network that is a combination of various networks such as the Internet or an intranet. Further, the client computer 1 and the image processing server 2 may be connected to each other by a dedicated line or the like instead of the network.

  The client computer 1 includes a ROM (Read Only Memory), a CPU (Central Processing Unit), a RAM (Random Access Memory), an external storage device composed of a hard disk drive, etc., a recording medium drive composed of a CD-ROM drive, etc. A computer, an input device composed of a keyboard, a mouse, etc., a communication device composed of a modem or a router, etc., and a normal computer equipped with a display device composed of a CRT (cathode ray tube) or a liquid crystal display device, etc. Is done. The ROM stores a BIOS (Basic Input / Output System) and the like in advance, the RAM is used as a work area of the CPU, and the hard disk drive stores an OS (Operating System) and a predetermined client program in advance. By executing these programs, they function as the communication unit 11, the control unit 12, the image data acquisition unit 13, the display unit 14, and the operation unit 15.

  The control unit 12 controls operations of the communication unit 11, the image data acquisition unit 13, the display unit 14, and the operation unit 15. The operation unit 15 is used to input various user operation commands, processed image specifying information, and the like, and outputs a target image acquisition command, processed image specifying information, and the like input by the user to the control unit 12. When the user inputs a target image acquisition command using the operation unit 15, the image data acquisition unit 13 is controlled by the control unit 12, and photographic image data or the like from an image data output device (not shown) such as a scanner or a digital camera. Are obtained and output to the control unit 12. The control unit 12 transmits the target image and the processed image specifying information to the image processing server 2 using the communication unit 11. Further, the control unit 12 receives the processed image transmitted from the image processing server 2 using the communication unit 11 and causes the display unit 14 to display the processed image.

  In addition, the user desires to register the processed image created by the user together with the processing description used when creating the processed image in the image processing server 2, and the user inputs the processed image and the processing description registration command using the operation unit 15. In this case, the image data acquisition unit 13 acquires the processed image created by the user and the processing description used when the processed image is generated, and outputs the acquired processed image to the control unit 12. The control unit 12 transmits the processed image and the processing description to the image processing server 2 using the communication unit 11.

  Here, the processing description is integrated into the processed image, and for example, when the processed image is created in an image format having a comment field, all the processes used for the processed image creation process in the comment field The procedure can be written as a process description or superimposed on the image data itself as a digital watermark. In this case, the processed image is preferably described by vector information in order to sufficiently reduce the data amount of the processing description. The process description may be in a form separate from the processed image. In this case, the process description and the processed image are acquired separately.

  The image processing server 2 includes a normal server computer including an external storage device including a ROM, a CPU, a RAM, a hard disk drive, and the like, and a communication device including a router. The ROM stores a BIOS (Basic Input / Output System) and the like in advance, the RAM is used as a work area of the CPU, the hard disk drive stores an OS (Operating System), an image processing program, and the like. By executing the processing program or the like, it functions as the communication unit 21, the image processing unit 22, the specific information storage unit 23, the processed image generation unit 24, and the processing description generation unit 25.

  The communication unit 21 acquires the target image and the processed image specifying information transmitted from the client computer 1 and outputs them to the image processing unit 22. The specific information storage unit 23 stores in advance a plurality of process descriptions that describe all the processing procedures when creating a processed image from the original image in association with the process description specifying information for specifying the process description, A plurality of stored process description specifying information and processed image specifying information of the processed image created by the process description specified by the process description specifying information are stored in advance in association with each other, and function as a database. For example, when the processed image having the processed image specifying information “98765” is created by the processing description including the seven processes A to G having the processing description specifying information “1234”, the processing description specifying information “12345” is set. Correspondingly, the processes A to G are stored as process descriptions in a table format, and the processed image specifying information “9876” is stored in a table format for the process description specifying information “12345”.

  The image processing unit 22 reads the processing description specifying information associated with the processed image specifying information acquired by the communication unit 21 with reference to the specifying information storage unit 23, and is associated with the read processing description specifying information. The process description is read out, and the target image acquired by the communication unit 21 is processed using the read process description to create a new processed image. The communication unit 21 transmits the processed image created by the image processing unit 22 to the client computer 1 via the network 3.

  The processed image generation unit 24 generates processed image specifying information for specifying the processed image with respect to the processed image created by the image processing unit 22. The specific information storage unit 23 newly stores the processed image specifying information generated by the processed image generation unit 24 in association with the processing description specifying information of the processing description used when creating the processed image, and updates the database.

  When the processed image created by the user and the processing description integrated with the processed image are transmitted from the client computer 1, the communication unit 21 integrates the processed image created by the user with the processed image. And process description embedded in. The processing description generation unit 25 generates processed image specifying information for specifying the processed image with respect to the processed image acquired by the communication unit 21, extracts a processing description from the processed image, and extracts the processing Processing description specifying information for specifying the processing description is generated for the description. The specific information storage unit 23 newly stores the processed image specifying information generated by the processing description generating unit 25 in association with the processed description specifying information generated together with the processed image specifying information. The processing description acquired by the communication unit 21 in association with the information is newly stored, and the database is updated.

  On the other hand, when the processed image created by the user and the processing description used when creating the processed image are individually transmitted from the client computer 1, the communication unit 21 transmits the processed image created by the user and the processed image. Acquire individually the process description used at the time of creation. The processing description generation unit 25 generates processed image specifying information for specifying the processed image with respect to the processed image acquired by the communication unit 21, and applies the processing description acquired by the communication unit 21 to the processing description. Process description specifying information for specifying the process description is generated. Also in this case, as described above, the specific information storage unit 23 updates the database by newly storing the processed image specifying information, the process description specifying information, and the process description in association with each other.

  In the present embodiment, the communication unit 21 corresponds to an example of an acquisition unit, the specific information storage unit 23 corresponds to an example of a specific information storage unit, the image processing unit 22 corresponds to an example of an image processing unit, and a processed image The generating unit 24 corresponds to an example of a processed image generating unit, and the processing description generating unit 25 corresponds to an example of a processing description generating unit.

  Next, image processing by the image processing system configured as described above will be described. FIG. 2 is a flowchart for explaining image processing by the image processing system shown in FIG.

  First, when the user operates the operation unit 15 to input the target image acquisition command and the processed image specifying information, the control unit 12 of the client computer 1 in step S <b> 11 receives the target image acquisition command and the processed image specifying information from the operating unit 15. And the target image is acquired using the image data acquisition unit 13 in accordance with the target image acquisition command. Next, in step S <b> 12, the control unit 12 transmits the target image and the processed image specifying information to the image processing server 2 using the communication unit 11.

  At this time, in step S <b> 21, the communication unit 21 of the image processing server 2 receives the target image and the processed image specifying information transmitted from the client computer 1 and outputs them to the image processing unit 22. Next, in step S <b> 22, the image processing unit 22 reads out the processing description specifying information associated with the processed image specifying information acquired by the communication unit 21 from the specifying information storage unit 23 and specifies the processing description specifying information. . Next, in step S <b> 23, the image processing unit 22 reads out the process description associated with the specified process description specifying information from the specified information storage unit 23. Next, in step S <b> 24, the image processing unit 22 performs image processing on the target image acquired by the communication unit 21 using the read processing description, and creates a new processed image having a style determined by the processing description.

  Next, in step S25, the image processing unit 22 notifies the processed image generation unit 24 that a new processed image has been created, requests processed image specifying information of the processed image, and generates the processed image generation unit 24. Generates processed image specifying information for specifying the processed image for the requested processed image and outputs the generated processed image specifying information to the specific information storage unit 23. Here, various generation methods can be used as the method for generating the processed image specifying information as long as a unique one can be generated for each processed image. For example, the specific information storage unit 23 manages the processed images with serial numbers. If it is, the unused numbers are used in ascending order.

  Next, in step S26, the specific information storage unit 23 newly stores the processed image specifying information generated by the processed image generating unit 24 in association with the processing description specifying information of the processing description used when the processed image is created. And update the database. As a result, the created processed image can be automatically made into a database. The data stored in the specific information storage unit 23 is not limited to the processed image specifying information, and the processed image and / or the target image may be stored and databased.

  Finally, in step S <b> 27, the image processing unit 22 outputs the processed image for which the storage of the processed image specifying information is completed to the communication unit 21, and the communication unit 21 transmits the processed image to the client computer 1. At this time, in step S <b> 13, the communication unit 11 of the client computer 1 receives the processed image transmitted from the image processing server 2 and outputs it to the control unit 12. Finally, in step S <b> 14, the control unit 12 displays the processed image on the display unit 14. In this way, when there is a processed image that the user likes, if the user can know the processed image specifying information of the processed image, the user can simply send the processed image specifying information to the image processing server 2 and can select his / her favorite image. Processing can be added.

  Through the above processing, the target image and the processed image specifying information are acquired, the processing description specifying information associated with the processed image specifying information is specified with reference to the specific information storage unit 23, and the specified processing description specifying information is specified. Since the target image can be image-processed using the processing description specified by the processing image, the target image is automatically processed using the processed image specifying information of the processed image processed by another person so that the processed image becomes the style of the processed image. When a user likes a processed image processed by another person, the image can be easily processed so that the processed image has a style.

  Next, a process description registration process executed as a part of the image processing process will be described. FIG. 3 is a flowchart for explaining the process description registration process by the image processing system shown in FIG.

  First, when the user operates the operation unit 15 to input a process description registration command, the control unit 12 of the client computer 1 uses the image data acquisition unit 13 in accordance with the process description registration command from the operation unit 15 in step S31. To obtain a processed image in which the process description is integrated. Next, in step S <b> 32, the control unit 12 transmits to the image processing server 2 the processed image in which the process description is integrated using the communication unit 11.

  At this time, in step S41, the communication unit 21 of the image processing server 2 receives the processed image in which the processing description transmitted from the client computer 1 is integrated and outputs the processed image to the processing description generating unit 25. Next, in step S <b> 42, the process description generation unit 25 extracts the process description from the processed image acquired by the communication unit 21 and outputs the process description to the specific information storage unit 23.

  Next, in step S 43, the process description generation unit 25 generates process description specifying information for specifying the process description for the extracted process description, and outputs the process description specifying information to the specified information storage unit 23. Here, as a method for generating the process description specifying information, various generation methods can be used as long as a unique one can be generated for each process description. For example, the specific information storage unit 23 manages the process descriptions by serial numbers. If it is, the unused numbers are used in ascending order.

  Next, in step S44, the specific information storage unit 23 updates the database by newly storing the process description extracted in step S42 in association with the process description specifying information generated in step S43.

  Next, in step S45, the process description generation unit 25 generates processed image specifying information for specifying the processed image with respect to the processed image received in step S41, and outputs the processed image specifying information to the specific information storage unit 23. Finally, in step S46, the specific information storage unit 23 updates the database by storing the processed image specifying information generated by the processing description generating unit 25 in association with the processing description specifying information stored in step S44. As a result, the processing description of the processed image created by the user can be automatically made into a database. Note that the processed image specifying information and the processing description specifying information may be transmitted to the client computer 1 together with the registration completion notification in order to let the user know the processed image specifying information and the processing description specifying information generated by the above processing. Good.

  Further, in the present embodiment, since the processed image can be managed in a state where the processed image specifying information and the processing description specifying information as ID numbers are assigned to all the processed images and the processing descriptions, the image processing server 2 A copyright management system can be constructed. As a result, when the user transmits the processed image specifying information to the image processing server 2, the processed image can be provided to the user after performing the charging process. It is also possible to charge for it and to manage the copyright fee.

  Next, the image processing unit 22 shown in FIG. 1 will be described in more detail. FIG. 4 is a block diagram showing a configuration of an example of the image processing unit 22 shown in FIG. 4 includes an image data reception unit 31, a group feature calculation unit 32, a divided image data storage unit 33, a group feature storage unit 34, an integrated region calculation unit 35, an integrated region storage unit 36, and image processing. The unit 37 is provided.

  The image data receiving unit 31 outputs the target image acquired by the communication unit 21 to the group feature calculation unit 32 and specifies the process description specifying information associated with the processed image specifying information acquired by the communication unit 21. The information is read from the information storage unit 23, and the process description associated with the read process description specifying information is read and output to the group feature calculation unit 32, the integrated region calculation unit 35, and the image processing unit 37.

  The group feature calculation unit 32 divides the input target image into a plurality of group regions by a known region division process according to the process description output from the image data reception unit 31, and the feature amount of the group region for each group region And the feature amount of each group area is stored in the group feature storage unit 34, and a plurality of divided image data obtained by dividing the target image is stored in the divided image data storage unit 33.

  Here, as the area dividing process, various area dividing processes can be used, and a method of extracting a brush stroke from a painting (Nakajima et al .: “Analysis of a painting image by a computer—automatic division into an oil painting image touch”), IEICE Spring National Convention Preliminary Proceedings, D (1994)) and 5-dimensional K-means algorithm (Izumi et al .: “Study on segmentation method using color information and position information”, IEICE Spring National Convention Preliminary Lecture Collection, D680 (1991)), Japanese Patent Application Laid-Open No. 10-11569, etc. can be used. The feature amount of the group area includes at least the average color for each group area, the position on the image for each group area, and the number of pixels for each group, and the division obtained by the above-described area division processing as necessary A feature amount that can be extracted from the region may be used.

  The divided image data storage unit 33 stores the divided image data in a table format in association with the area number of the group area. The group feature storage unit 34 stores the feature amount of each group area in association with the area number of the group area in a table format.

  The integrated area calculation unit 35 integrates a plurality of group areas based on the feature amounts of each group area stored in the group feature storage unit 34 in accordance with the process description output from the image data reception unit 31. An area is created and stored in the integrated area storage unit 36. Specifically, the integrated area calculation unit 35 sets, as the first integrated area, an area in which the difference in feature amount between adjacent group areas is smaller than the first threshold and the number of group areas is equal to or greater than the first number. Integrate. In addition, the integrated region calculation unit 35 has a feature amount difference between adjacent group regions smaller than the first threshold, the number of group regions is equal to or more than a second number less than the first number, and An average value of a difference between a feature amount of a group region of a predetermined ratio or more among a plurality of group regions located at a boundary and a feature amount of a group region in an adjacent integrated region is larger than a second threshold value equal to or greater than a first threshold The area is integrated as a second integrated area.

  At this time, the integrated area calculation unit 35 repeats the process of creating the first and second integrated areas a plurality of times, increases the first threshold according to the number of process repetitions, and increases the first threshold according to the number of process repetitions. Reduce the number. In this example, the average color of each group area is used as the feature amount of the group area, and the first and second integrated areas are created based on the color difference between adjacent group areas. Note that the feature amount used for the integrated region creation process is not particularly limited to the above example. For example, a predetermined function f (c, t) having the color component c and the texture component t of the group region as variables is used. The function f (c, t) may be defined as a feature amount, and the difference may be used as a reference.

  The integrated area storage unit 36 stores the created integrated area. Specifically, the integrated area storage unit 36 stores the area numbers of a plurality of group areas constituting each integrated area in association with the area numbers of the integrated areas in a table format.

  The image processing unit 37 reads one or a plurality of integrated regions as a preset expression region from the integrated region storage unit 36 in accordance with the processing description output from the image data receiving unit 31, and the feature amount of the group region constituting the read integrated region Are read from the group feature storage unit 34. Next, the image processing unit 37 creates a preset expression so that the expression selected by the process description is selected from a plurality of preset expressions, and the created preset expression is stored in the divided image data storage unit 33. By adding to the plurality of divided image data, the target image is processed into a painting-like image or the like and output to the communication unit 21. At this time, even when a plurality of integrated regions are selected as one region, the image processing unit 37 refers to the integrated region stored in the integrated region storage unit 36, and more area is selected from the selected integrated regions. An area is selected in units of integrated areas in preference to large integrated areas.

  In this example, the group feature calculation unit 32 corresponds to an example of a feature amount calculation unit, the group feature storage unit 34 corresponds to an example of a group feature storage unit, and the integrated region calculation unit 35 corresponds to an example of an integration unit. The integrated area storage unit 36 corresponds to an example of an integrated area storage unit, and the image processing unit 37 corresponds to an example of a processing unit.

  FIG. 5 is a flowchart for explaining the image processing by the image processing unit 22 shown in FIG. First, in step S51, the image data receiving unit 31 acquires the target image and the processed image specifying information from the communication unit 21, outputs the target image to the group feature calculating unit 32, and associates the target image with the processed image specifying information. The processing description of the specified processing description specific information is read from the specific information storage unit 23 and output to the group feature calculation unit 32, the integrated region calculation unit 35, and the image processing unit 37.

  Next, in step S52, the group feature calculation unit 32 divides the target image into a plurality of group regions according to the process description output from the image data reception unit 31, and calculates the feature amount of the group region for each group region. Then, the feature amount of each group region is stored in the group feature storage unit 34, and a plurality of divided image data is stored in the divided image data storage unit 33.

  Next, in accordance with the process description output from the image data receiving unit 31, the integrated area calculation unit 35 first increments the process repeat count I by adding 1 to the process repeat count I in step S53. Here, the initial value of the processing repetition count I is set to 0 in advance, and in the case of the first processing, the processing repetition count I is set to 1. In addition, as the total number of repetitions R that is the maximum number of processing repetitions I, a value of 2 or more is set in advance, for example, 10 is set.

  Next, in step S54, the integrated region calculation unit 35 sets a certain group region as the target group region, and the color difference D that is the difference between the average color of the group region adjacent to the target region and the average color of the target region is the first. It is determined whether or not D is smaller than the threshold T, that is, whether or not D <T. If the color difference D is not smaller than the first threshold T, step S54 is repeated to compare the color difference with other adjacent group areas. If the color difference D is smaller than the first threshold T, the process proceeds to step S55, and the group area having the color difference D smaller than the first threshold T is stored in a table format in association with the attention area. Integrate areas.

  Here, if the reference threshold is B, it is expressed as T = I × B / R. At this time, if the brightness range of the original image is L, B = L / 80, and the number of processing repetitions I increases. The first threshold value T is set so as to increase according to. The first threshold value T is not particularly limited to the above formula, and various changes are possible. For example, 80 in the above formula may be changed to any numerical value within the range of 10 to 200. In addition, a region having a high flatness, that is, a long and narrow region, may be excluded from the comparison target among the group regions, and the group region may be prevented from being integrated by overcoming the region.

  Next, in step S56, the integrated region calculation unit 35 determines whether or not the comparison processing in step S54 has been completed for all of the adjacent group regions, and the comparison processing is performed for all of the adjacent group regions. If not completed, step S54 is repeated to compare the feature values with other adjacent group areas. If the comparison process is completed for all adjacent group areas, the process proceeds to step S57. .

  Next, in step S57, the integrated area calculation unit 35 determines whether or not all group areas have been processed as attention areas. If all group areas have not been processed as attention areas, step S54 is repeated. When the feature amounts are compared using the other group areas as the attention area, and all the group areas are processed as the attention area, the process proceeds to step S58.

  Next, in step S58, the integrated area calculation unit 35 includes areas where the number N of group areas constituting the area is equal to or more than the first number G / (U × I + 1) among the areas integrated by the above processing. If there is no region that satisfies N ≧ G / (U × I + 1), the process proceeds to step S60. If there is a region that satisfies N ≧ G / (U × I + 1), step S59 is performed. , The region satisfying N ≧ G / (U × I + 1) is created as the first integrated region, and the region numbers of the plurality of group regions constituting each first integrated region are assigned to the first The information is stored in the integrated area storage unit 36 in a table format in association with the area number of the integrated area.

  Here, G is the total number of group areas into which the original image is divided, for example, 6435 is used, U is the number of unit areas, for example, 120 is used, and I is the above processing repetition. Is the number of times. In this case, a region with a high sense of unity can be extracted as a first integrated region in a wide range, and the first threshold T increases as the number of processing repetitions I increases, and the first number is the number of processing repetitions. Since I decreases as I increases, it is possible to extract a region having a high sense of unity over a wide range by repeating processing as a first integrated region with priority according to the number of repetitions. The first number is not particularly limited to the above formula and can be variously changed. For example, the number U of unit areas in the above formula can be changed to an arbitrary value within the range of 10 to 100, or the processing can be performed. A constant value unrelated to the number of repetitions I may be used.

  Next, in step S60, the integrated region calculation unit 35 has the number N of group regions constituting the region out of the remaining regions that are not extracted as the first integrated region by the above processing is equal to or greater than the second number. The average A of the difference between the average color of 80% or more of the group areas among the group areas located at the boundary of the integrated areas and the average color of the group areas belonging to the other integrated areas adjacent to each other is the second It is determined whether or not the threshold value is greater than the threshold value. If there is no region that satisfies these conditions, the process proceeds to step S62. If there is a region that satisfies these conditions, the process proceeds to step S61 and these conditions are set. An area to be satisfied is created as a second integrated area, and the area numbers of a plurality of group areas constituting each second integrated area are associated with the area numbers of the second integrated area in a table format. It is stored in the case area storage unit 36.

  In this example, for example, 3 is used as the second number, and 20 (about 6 times the reference threshold B) is used as the second threshold. Note that the second number and the second threshold are not particularly limited to the above example, and two or four or more numbers are used as the second number, or the reference threshold B is set as the second threshold. Any value can be used from the range of 1 to 20 times. In this case, a region having a high contrast effect in a narrow range can be extracted as the second integrated region, and the first threshold T increases as the number of processing repetitions I increases. A region having a high contrast effect can be extracted as a second integrated region with a priority according to the number of repetitions.

  Next, in step S62, the integrated area calculation unit 35 determines whether or not the processing repetition count I matches the total repetition number R. If the processing repetition count I has not reached the total repetition count R, the process proceeds to step S53. Then, the subsequent processing is repeated, and when the processing repetition count I has reached the total number of repetitions R, the integrated region creation processing is terminated and the processing proceeds to step S63. Note that even when all the group areas are integrated into the first and second integrated areas before the process repetition count I reaches the total number of repetitions R, the integrated area creation process is terminated and the process proceeds to step S63. May be. By this processing, it is possible to classify into a first integrated region, a second integrated region, and other integrated regions with high priority (for example, the number of repetitions <3).

  Next, in step S63, first, the image processing unit 37 reads and reads one or a plurality of integrated regions from the integrated region storage unit 36 as preset expression regions in accordance with the process description output from the image data receiving unit 31. The feature amount of the group area constituting the integrated area is read from the group feature storage unit 34. In this case, since the image processing unit 37 selects a region in units of integrated regions, it is possible to switch the representation such as switching the representation at a boundary with less unnaturalness or adding the representation only to a region with a high sense of unity. Further, the designated area can be finely adjusted by changing the designation of the priority order according to the number of repetitions.

  In addition, as a method for automatically specifying the appropriate priority for each integrated area, the integrated area result that is integrated by the number of repetitions when the number of group areas included in the integrated area is no longer increased or extremely reduced Can be used. In addition, the image processing unit 37 selects not only the integrated region stored in the integrated region storage unit 36 but also the feature amount stored in the group feature storage unit 34 as a preset expression region in units of group regions. In this case, it is possible to select a region in detail.

  Thereafter, the image processing unit 37 uses a feature amount processing process disclosed in Japanese Patent Application Laid-Open No. 10-11569 or the like so as to obtain an expression selected by a process description from a plurality of preset expressions. The preset expression is created, and the created preset expression is added to the plurality of divided image data stored in the divided image data storage unit 33, thereby processing the target image into a painting-like image or the like, and the communication unit 21. Output to.

  At this time, since the image processing can be performed using the feature amount of the group area, instead of performing the image processing in units of pixels, the group area is determined in the same manner as the technique disclosed in Japanese Patent Laid-Open No. 10-11569. It can be expressed by vector information composed of the size and direction of the area, the curvature, etc., and it is more natural for human beings and a work can be created by freely changing the enlargement ratio.

  FIG. 6 is a diagram schematically illustrating an example of a divided image, an integrated image, and a processed image. FIG. 6A shows a divided image stored in the divided image data storage unit 34, and each ellipse in the drawing corresponds to a group region. FIG. 6B shows an integrated image stored in the integrated area storage unit 36. In this example, an integrated area R1 corresponding to the background and an integrated area R2 corresponding to the person are created. Finally, (c) of FIG. 6 shows a processed image created by the image processing unit 22, and a blurred expression is added to the integrated region R1 corresponding to the background, or an integrated region R2 corresponding to a person. A clear expression can be added.

  FIG. 7 is a diagram illustrating an example of a target image, FIG. 8 is a diagram illustrating an example of an integrated image, and FIG. 9 is a diagram illustrating an example of a processed image. When the target image shown in FIG. 7 is processed into a picture-like image by the image processing unit 22 shown in FIG. 4, an integrated image in which each integrated area is set is created as shown in FIG. 8, for example, the area T1 is repeated. The region is a region with a high sense of unity extracted at the first time, the region T2 is a region with a high contrast effect extracted at the first repetition, and the region T3 is extracted at the third repetition This is a highly unified area. Thereafter, a painting-like expression was added to each integrated region, and the processed image shown in FIG. 9 could be created.

  With the above processing, in this example, the target image is divided into a plurality of group areas, the feature amount of the group area is calculated for each group area, and the calculated feature amount is stored and stored for each group area. Multiple group areas are integrated based on feature values to create multiple integrated areas, and image processing is performed in units of the integrated areas. A high area can be extracted and a gradation area can be integrated into one area. As a result, when giving a plurality of expressions to the target image, it is possible to easily and easily extract a region to which each expression is given.

  The configuration of the image processing unit 22 is not particularly limited to the above example, and various modifications are possible. For example, an image processing technique disclosed in Japanese Patent Laid-Open No. 10-11569 or the like without using an integrated region. May be used.

  In the present embodiment, the image is processed by the image processing server 2 using the client / server model. However, the client computer 1 processes the image, or all processing is executed by one computer. Various changes are possible.

It is a block diagram which shows the structure of the image processing system using the image processing server by one embodiment of this invention. It is a flowchart for demonstrating the image processing process by the image processing system shown in FIG. It is a flowchart for demonstrating the process description registration process by the image processing system shown in FIG. It is a block diagram which shows the structure of an example of the image process part shown in FIG. It is a flowchart for demonstrating the image process by the image process part shown in FIG. It is a figure which shows typically an example of a divided image, an integrated image, and a process image. It is a figure which shows an example of a target image. It is a figure which shows an example of an integrated image. It is a figure which shows an example of a process image.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Client computer 2 Image processing server 21 Communication part 22 Image processing part 23 Specific information storage part 24 Processed image generation part 25 Process description generation part 31 Image data reception part 32 Group feature calculation part 33 Division | segmentation image data storage part 34 Group feature Storage unit 35 Integrated region calculation unit 36 Integrated region storage unit 37 Image processing unit

Claims (7)

An image processing apparatus that performs image-like processing on image data,
An acquisition means for acquiring a target image to be subjected to image processing and processed image specifying information for specifying a processed image created by performing image processing on the original image;
A processing description describing a processing procedure for creating a processed image from an original image is stored in advance in association with processing description specifying information for specifying the processing description, and the processing description specifying information and the processing description specifying Specific information storage means for preliminarily storing the processed image specific information of the processed image created by the process description specified by the information,
The process description specifying information associated with the processed image specifying information acquired by the acquiring means with reference to the specifying information storage means is specified, and the process description specified by the specified process description specifying information is used. Image processing means for image processing the target image acquired by the acquisition means ,
The image processing means includes
A feature amount calculating means for dividing the target image into a plurality of group regions and calculating a feature amount of the group region for each group region;
Group feature storage means for storing the feature quantity calculated by the feature quantity calculation means for each group region;
Integration means for creating a plurality of integrated areas by integrating a plurality of group areas based on the feature quantities stored in the group feature storage means;
Integrated area storage means for storing the integrated area integrated by the integration means;
Processing means for performing image processing on the target image in units of integrated areas stored in the integrated area storage means,
The integration unit uses the feature amount stored in the group feature storage unit, and the difference in feature amount between adjacent group regions is smaller than a first threshold and the number of group regions is equal to or greater than the first number. Are integrated as a first integrated region, and using the feature amount stored in the group feature storage means, the difference in feature amount between adjacent group regions is smaller than the first threshold, The number of group areas is equal to or greater than a second number less than the first number, and a group area in an integrated area adjacent to a predetermined amount or more of a group area feature amount among a plurality of group areas located at the boundary of the integrated area An image processing apparatus that integrates a region having an average value of a difference from a region feature amount that is greater than a second threshold that is equal to or greater than the first threshold as a second integrated region . Further comprising processed image generation means for generating processed image specifying information of the processed image created by the image processing means;
The specified information storage means newly stores the processed image specifying information generated by the processed image generating means in association with the processing description specifying information of the processing description used when creating the processed image. Item 12. The image processing apparatus according to Item 1. The acquisition means acquires a processed image created by a user and a processing description used when the processed image is created,
The processing description generation information for generating the processing description specific information of the processing description acquired by the acquisition means and the processing image generation information for the processed image acquired by the acquisition means is further provided,
The specific information storage means stores the processed image specifying information generated by the processing description generating means in association with the processing description specifying information generated together with the processed image specifying information, and corresponds to the processing description specifying information. The image processing apparatus according to claim 1, wherein the processing description acquired by the acquisition unit is stored. The acquisition means acquires a processed image created by a user and a processing description integrated with the processed image,
The processing description generation unit generates the processed image specifying information of the processed image acquired by the acquiring unit, extracts the processing description from the processed image, and generates the processing description specifying information of the extracted processing description The image processing apparatus according to claim 3. The image processing device includes a server computer communicably connected to a client computer via a predetermined network,
The image processing apparatus according to claim 1, wherein the acquisition unit acquires a target image and processed image specifying information selected by a user from the client computer. An image processing method for performing image- like processing on image data using a computer,
An acquisition step in which the computer acquires a target image to be subjected to image processing and processed image specifying information for specifying a processed image created by performing image processing on the original image;
The computer stores in advance a process description describing a processing procedure for creating a processed image from an original image in association with process description specifying information for specifying the process description, and the process description specifying information and The processing image specifying information acquired in the acquisition step is referred to by referring to the specific information storage means that stores the processing image specifying information of the processing image created by the processing description specified by the processing description specifying information in advance in association with the processing image specifying information. identify the process description specifying information associated with, seen containing an image processing step of the image processing target image obtained in said obtaining step by using the process described specified by the specified process description specifying information,
The image processing step includes
A feature amount calculating step of dividing the target image into a plurality of group regions and calculating a feature amount of the group region for each group region;
A group feature storage step of storing the feature amount calculated in the feature amount calculation step for each group region;
An integration step of integrating a plurality of group regions based on the feature amounts stored in the group feature storage step to create a plurality of integration regions;
An integrated region storage step for storing the integrated region integrated in the integration step;
A processing step of performing image processing on the target image in units of the integrated region stored in the integrated region storage step,
In the integration step, the feature amount stored in the group feature storage step is used, a difference in feature amount between adjacent group regions is smaller than a first threshold value, and the number of group regions is equal to or more than the first number. The regions are integrated as a first integrated region, and the feature amount stored in the group feature storage step is used to make a difference in feature amount between adjacent group regions smaller than the first threshold, and the group region The number of group regions in the integrated region adjacent to the feature amount of the group region equal to or greater than a predetermined ratio among the plurality of group regions located at the boundary of the integrated region, the number of which is equal to or greater than the second number less than the first number. image, wherein the average value of the difference between the feature quantity includes the step of integrating the area greater than the first threshold value or more second threshold as a second combined area Engineering method. An image processing program for performing painting-like processing on image data,
An acquisition means for acquiring a target image to be subjected to image processing and processed image specifying information for specifying a processed image created by performing image processing on the original image;
A processing description describing a processing procedure for creating a processed image from an original image is stored in advance in association with processing description specifying information for specifying the processing description, and the processing description specifying information and the processing description specifying Specific information storage means for preliminarily storing the processed image specific information of the processed image created by the process description specified by the information,
The process description specifying information associated with the processed image specifying information acquired by the acquiring means with reference to the specifying information storage means is specified, and the process description specified by the specified process description specifying information is used. Causing the computer to function as image processing means for performing image processing on the target image obtained by the obtaining means ;
The image processing means includes
A feature amount calculating means for dividing the target image into a plurality of group regions and calculating a feature amount of the group region for each group region;
Group feature storage means for storing the feature quantity calculated by the feature quantity calculation means for each group region;
Integration means for creating a plurality of integrated areas by integrating a plurality of group areas based on the feature quantities stored in the group feature storage means;
Integrated area storage means for storing the integrated area integrated by the integration means;
Processing means for performing image processing on the target image in units of integrated areas stored in the integrated area storage means,
The integration unit uses the feature amount stored in the group feature storage unit, and the difference in feature amount between adjacent group regions is smaller than a first threshold and the number of group regions is equal to or greater than the first number. Are integrated as a first integrated region, and using the feature amount stored in the group feature storage means, a difference in feature amount between adjacent group regions is smaller than the first threshold, The number of group areas is equal to or greater than a second number smaller than the first number, and a group area in an integrated area adjacent to a predetermined amount or more of a group area feature amount among a plurality of group areas located at the boundary of the integrated area An image processing program that integrates, as a second integrated region, a region in which an average value of differences from a region feature amount is greater than a second threshold that is equal to or greater than the first threshold .