JP2007143047A - Image processing system and image processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce a load of image processing performed by an operator in the case of carrying out print with high quality. <P>SOLUTION: The feature information of an image is specified according to an image analysis and the determination of an operator with respect to input image data. The database of a printer is stored with sample information including the feature information of the image and an image processing parameter used for the image as the history of image processing performed in the past. The sample information matched with the feature information of the image is retrieved from the database. When the proper sample information is found out, the image processing using the image processing parameter is executed. When any proper sample information does not exist, the database of the printer of the other store is retrieved. When any proper sample information does not exist in any database, the image processing parameter is set by an operator. The image processing parameter set by the operator is registered in the database as new sample information. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an image processing system and an image processing program for performing image processing in advance when printing an image.

  2. Description of the Related Art Digital photo printers that perform print processing on photographic paper based on image data obtained by a digital camera or image data obtained by optically reading a developed photographic film are known (for example, Patent Documents). 1). The digital photo printer includes an image processing unit that appropriately corrects the color tone, light / dark density, and the like before print processing, and can create a good-looking print photo even for an image that has failed to be captured, for example.

  When performing image processing, select image processing items for which processing details such as brightness correction, contrast correction, sharpness correction, white balance correction, saturation correction, etc. have been determined in advance, and images prepared for each image processing item The strength of correction is determined by selecting a processing parameter. In order to create a high-quality printed photograph, it is necessary to perform appropriate image processing depending on, for example, the brightness and hue of the image due to the difference in weather, and depending on the shooting conditions that differ for each image Appropriate image processing items and parameters are determined. For this reason, advanced image processing with selection of image processing parameters has been performed by skilled operators.

Image processing is especially difficult for people. There are individual differences in human skin color, depending on ethnic and regional differences, the skin color is similar to brownish brown, reddish yellow, or bluish white. Different. In addition, the skin color changes depending on how the light hits a person, whether a woman has makeup, or the like. Furthermore, the appropriate color varies depending on the background scenery such as forests and lakes, and the appropriate color varies depending on the preference of the photographer and the cultural custom.
JP-A-2005-184627 (particularly paragraphs 0028 to 0043)

  However, it takes a lot of work to perform different image processing according to the person in the image, the photographer's preference, and the like. Further, it is impossible to improve the print quality simply by performing certain image processing in order to increase the work efficiency.

  The present invention has been made in consideration of the above problems, and provides an image processing system and an image processing program therefor that can reduce the burden of image processing by an operator and improve the efficiency of image processing while maintaining high print quality. The purpose is to do.

  An image processing system according to the present invention includes image processing means for performing image processing on image data in accordance with image processing parameters, a database in which sample information including image feature information and image processing parameters applied to the image is stored, and a novel When the image data is input, sample information retrieval means for retrieving sample information that matches the feature information of the image from the database, and sample information that matches the feature information is obtained by the sample information retrieval means In accordance with the image processing parameters included in the sample information, the image processing instruction means for causing the image processing means to execute image processing of new image data and the sample information matching the characteristic information cannot be obtained by the sample information searching means. Feature information of the new image and the image processing parameters used for that image. The new sample information including bets, characterized in that a sample information registration means for registering in the database.

  Further, the image processing apparatus is characterized by comprising image analysis means for analyzing the image data and calculating the feature information.

  The image analysis means includes face detection means for detecting a human face in the image, and skin color identification means for identifying the skin color of the detected face.

  Further, the image analysis means extracts photographing information added to image data as image feature information.

  The database is provided in each of a plurality of image processing apparatuses having the image processing means and the sample information searching means, and the plurality of image processing apparatuses are connected to each other via a network. To do.

  The database is provided in a server having the sample information searching means, and the server is connected to an image processing apparatus having the image processing means via a network, and the server is connected to the image processing apparatus from the image processing apparatus. When sample information is requested, the sample information is retrieved from the database, and the sample information is transmitted to the image processing apparatus.

  According to the image processing system of the present invention, sample information including image feature information and image processing parameters is accumulated as history data indicating the contents of image processing performed by an operator having specialized knowledge. For images that have the same characteristics as the processed image, image processing performed in the past is performed as a basic process, and small differences that differ from image to image can be corrected by fine-tuning parameters. That is, the operator's workload can be reduced compared to the case where the image processing parameters are set from the beginning. As a result, the work efficiency of image processing is improved and costs such as labor costs can be reduced. Further, even an operator with insufficient expertise can easily perform image processing to finish a high-quality image equivalent to a skilled operator. In addition, for image types that have not been subjected to image processing in the past, the used image processing parameters are registered in the database together with the image feature information, so a rich database that can handle various images can be realized, The operator's workload can be further reduced.

  Moreover, since the feature information obtained by analyzing the image data is used as the feature information of the image, the feature information of the image that is difficult for the operator to visually determine can be included in the sample information, and an appropriate image processing parameter can be set. The accuracy of extraction can be increased.

  Also, when analyzing image data, the skin color is identified from the extracted person's face, so when performing image processing on an image of a person, the skin color is common as image feature information. Since sample information can be searched, it is possible to improve the accuracy of extracting appropriate image processing parameters for performing appropriate image processing on an image in which a person is captured.

  In addition, data relating to photographing conditions recorded as magnetic data on an APS-compatible photographic film, data relating to photographing conditions added to image data in a digital camera, and the like as part of the image information makes it possible to change images depending on the photographing environment. It is possible to extract image processing parameters that can appropriately correct the brightness and hue of the image.

  In addition, by constructing a network with a photo printer having a database, even if the photo printer has a database that does not store appropriate sample information, it is possible to retrieve sample information from the database of another photo printer. High-quality printing can be performed by image processing performed in advance at the time of image printing. In addition, the efficiency of print processing can be improved by automating advanced image processing. Also, by connecting the photo printer with a server that collectively manages the database, the sample information is immediately accumulated and the database is updated to the latest state, so the frequency of occurrence of mismatching of image feature information is reduced. it can.

  In FIG. 1, printers 10 are installed in a plurality of print service stores. The printer 10 is a digital photo printer, and functions as an image processing device that performs image processing on input image data and corrects image quality, and a printer that prints on photographic paper based on the image data subjected to image processing. It has. Note that when printing on photographic paper, a series of operations such as printing of an image for exposing the photographic paper based on the image data, development of the photographic paper, washing with water, drying, and the like can be performed by one unit. In each print service store, a storage medium in which image data is stored by a photographed photographic film or a digital camera is deposited from a general customer, and a service for printing an image on photographic paper using the printer 10 is performed.

  The printer 10 has a database 11. The printer 10 can record new data generated in the printer 10 in the database 11 and can read the already recorded data when necessary. The printers 10 of each store are connected to each other by a dedicated line and can transmit and receive data to and from each other. For example, data can be copied from the database 11 of a specific printer 10 to the database of another printer 10.

  In FIG. 2, the printer 10 includes a communication unit 15 for transmitting and receiving data to and from other printers 10, a program storage unit 17 in which a system program executed by the CPU 16 is stored, and input image data. A print execution unit 18 for optically recording an image on a photographic paper based on the above, and performing a development process, a drying process, etc. of the photographic paper; and an input operation unit 19 provided with a keyboard for an operator to perform an input operation; An image display unit 20 for displaying an image to be printed, and an image data input unit 21 for inputting image data from various storage media.

  The CPU 16 executes a system program stored in the program storage unit 17 and functions as an image analysis unit 22, an image processing unit 23, a system control unit 24, a sample search unit 25, and a sample registration unit 26. The image analysis unit 22 analyzes the input image data and performs histogram calculation, average color calculation, person detection, background identification, and the like. When detecting a person, the image analysis unit 22 detects a skin color region from the image, and identifies a face by detecting both eyes existing in the skin color region. Also, when identifying the background, the colors used in the background are often discriminated at the pixel level, the background type candidates corresponding to the colors are picked up, and the operator selects the background type from the candidates. Let For example, if the background color is blue, the sky, sea, lake, etc. are picked up as candidates, and if it is green, forests, turf, parks, etc. are picked up. When the background is an artifact, the type of artifact may be selected by the operator.

  The image processing unit 23 performs image processing for correcting image quality when creating a printed photograph. The printer 10 has several tens of types of image processing items as image processing patterns. The image processing unit 23 executes the image processing items selected by the operator or the system control unit 24, and performs image processing on the image data. I do. When a specific image processing item is executed, an image processing parameter determined for each image processing item is selected. For example, in the image processing item for correcting the brightness level, the parameter value that determines the intensity of increase / decrease in the brightness of the image is the image processing parameter, and in the image processing item for correcting the color balance, the colors of the three primary colors are strengthened. A parameter value that determines whether or not to enhance the color of the complementary color is an image processing parameter.

  The system control unit 24 controls each unit of the printer 10 according to a sequence determined by the program. The system control unit 24 stores the image data input from the image data input unit 21 in the database 11 and reads the image data from the database 11 when printing an image.

  In FIG. 3, the database 11 stores input image data and sample information D1 serving as a history of image processing performed on specific image data. The sample information D1 includes original image data that has been subjected to image processing, image feature information that represents image features, and image processing parameters that have been used for image processing. Image feature information includes skin color type information indicating the skin color type of a person in the image, shooting time zone information indicating the time zone in which the image was shot, background type information indicating the type of the photographed landscape and the background of the person, and image Information that characterizes the contents of the image, such as the number of persons in the image, the photographer, and the nationality of the person in the image, is included. The image feature information includes not only image features but also shooting data added and recorded on a photographic film and shooting data incorporated into image data recorded by a digital camera. Further, information obtained by the image analysis unit 22 is included. And information input at the operator's discretion.

  The image processing parameter is a parameter value for each image processing item applied to specific image data for which image feature information has been obtained. For example, parameter values of image processing items such as level correction, color balance correction, sharpness correction, and contrast correction. Each parameter value varies depending on the skin color, race, and shooting environment of the person in the image. If the image feature information is the same, the parameter values are also substantially the same.

  The sample search unit 25 searches the database 11 for sample information D1 that matches the feature information of the image to be printed, and extracts the corresponding sample information D1 when the image processing is performed in advance with the image print processing. In addition, when the sample information D1 corresponding to the database 11 of the own store does not exist, the sample search unit 25 transmits the characteristic information of the image to be printed to the printer 10 of another store. For example, when the image search unit 25 receives image feature information from the printer 10 at another store, the sample search unit 25 searches the database 11 at the store for sample information D1 that matches the feature information, and extracts the corresponding sample information. To do. The extracted sample information D1 is transmitted to the printer 10 at another store. If the corresponding sample information D1 does not exist, information indicating that there is no corresponding data is transmitted to another printer 10.

  When the target sample information D1 is not found in the database 11 of the own store or the other store, the sample registration unit 26 performs image data of the image to be printed, feature information obtained from the image data, and the image Then, the image processing parameters set by the operator are registered in the database 11.

  Next, the operation of the printer 10 will be described. In FIG. 4, a general customer brings photographed film, a memory card used in a digital camera, a CDROM on which image data is recorded, etc. to a store A which is a print service store, and requests printing of an image. . The photographic film is developed. The printer 10 is provided with an image data input unit 21 having a film scanner, a memory card reader, and a CDROM drive, and an image of the photographic film is read by the film scanner and input as image data. Image data is read from the memory card or CDROM drive.

  The system control unit 24 stores the image data input from the film scanner and the image data read from the memory card or CD ROM drive in the database 11. The operator performs an operation of selecting image data to be printed via the input operation unit 19. The image analysis unit 22 analyzes the image data selected for printing. The image analysis unit 22 performs histogram calculation, average color calculation, person detection, background identification, etc., obtains the number of people in the image, and identifies the skin color. In addition, the shooting location, shooting date, and the like of an image are read out from shooting data magnetically recorded as additional information of the photo film or shooting data incorporated in the image data. In addition to the image feature information derived by the image analysis unit 22, the operator inputs information on the person in the image, the race related to the customer's home region, and the like as the image feature information.

  When the image feature information is determined by the image analysis unit 22 and the operator, the sample search unit 25 searches the database 11 for sample information that matches the feature information. The sample search unit 25 compares the feature information of the image to be printed with the feature information of the image registered as sample information in the database 11 and extracts sample information that matches the feature information. When there are multiple types of sample information extracted, the operator calls the image data of the corresponding sample information, compares the image to be printed with the registered image, and based on the most similar image. Sample information can be selected.

  On the other hand, when sample information that matches the feature information is not found from the database 11 of the own store, the sample search unit 25 transmits the image feature information to a plurality of printers 10 at other stores. The feature information transmitted at this time is only information related to the race of the person shown in the image. The printer 10 at another store searches for sample information that matches the race information. The corresponding number of sample information that matches the race information registered in the database 11 of the other store is transmitted to the printer 10 of the own store. The sample search unit 25 transmits all feature information other than race information to the printer 10 in the store where the number of hits of sample information that matches the person information is the highest. Thus, for example, if the person appearing in the image to be printed is white, the sample information is retrieved from a database in which a large amount of sample information of the image showing the white is registered. In the printer 10 at another store, similar to the printer 10 at the own store, sample information that matches this is searched based on the image feature information.

  When the sample information that matches the feature information of the image is found in the printer 10 of another store, this is transmitted to the printer 10 of the own store. On the other hand, when sample information that matches the feature information is not found, it is notified that there is no corresponding sample information. If there is no sample information that matches the image feature information in the database 11 of the own store and the other store, an image processing parameter is set by the operator for the image to be printed.

  When sample information that matches the image characteristic information is found from the database 11 of the own store or another store, the image processing parameters included in the sample information are extracted. The image processing unit 23 executes image processing according to the extracted parameters. When sample information that matches the feature information of the image is not found from the database 11 of the own store and the other store, the image processing unit 23 executes image processing according to the image processing parameters set by the operator. When sample information that matches the image feature information is found, the operator only needs to perform fine adjustments to the image processing parameters. Compared to when the operator himself sets the image processing parameters from the beginning, The amount of work is greatly reduced.

  If the sample registration unit 26 does not obtain sample information that matches the feature information of the image and there is image data in which all image processing parameters are set by the operator, the sample registration unit 26 collects the image data and the feature obtained from the image data. New sample information composed of information and image processing parameters set by the operator is created and registered in the database 11.

  When one image is processed, the next image data is read out, feature information is determined in the same manner, and sample information is searched. When the corresponding sample information is obtained, image processing to which the image processing data is applied is performed. If the corresponding sample information is not obtained, image processing parameters are set by the operator, and new registration of sample information is performed. The image data that has undergone image processing is sent to the print execution unit 18. In the print execution unit 18, exposure printing is performed on the photographic paper based on the image data, and the photographic paper is developed, washed with water, dried, and the like, and a print photograph is created while sorting for each order.

  In addition, this invention is not restricted to the said embodiment, A suitable change is possible. For example, when searching for sample information from a database of another store, if the image shows a person, the race information is given priority, and the database has a large number of registered sample information related to an image showing a person of the same race. Although sample information that matches the information is searched, if it is a landscape image, priority is given to the representative color representing the most frequently used color or the average color that averages the colors of the entire image. Information may be searchable. For example, when a snow scene image is brought into a store in a region where snow does not fall, the parameters of the snow scene image used in a store in a region with a lot of snow can be used. In addition, if ID information for customer identification is added as image feature information and search is performed with priority given to sample information of images brought in by the same customer in the past, a print photo that suits the customer's preference is provided. It is also possible.

  If sample information that matches all the feature information of the image cannot be obtained, the sample information may be extracted in the descending order of the matching feature information, and the operator may select appropriate sample information. In addition, as shown in FIG. 5, a server 30 that collectively manages a database is provided, and each server and a printer are connected via a network, so that search and new registration of sample information can be centrally managed by the server 30. With this system, there is no need to query a plurality of databases when searching for sample information, and sample information can be searched quickly.

1 is a conceptual diagram of a print system. It is a block diagram of a printer. It is a conceptual diagram of a database. 6 is a flowchart illustrating a work flow when printing is performed. 1 is a conceptual diagram of a print system including a server.

Explanation of symbols

10 Printer 11 Database 15 Communication Unit 16 CPU
DESCRIPTION OF SYMBOLS 17 Program memory | storage part 18 Print execution part 19 Input operation part 20 Image display part 21 Image data input part 22 Image analysis part 23 Image processing part 24 System control part 25 Sample search part 26 Sample registration part

Claims (7)

Image processing means for performing image processing on image data in accordance with image processing parameters;
A database in which sample information including image feature information and image processing parameters applied to the image is stored;
Sample information retrieval means for retrieving, from the database, sample information that matches the feature information of the image when new image data is input;
An image processing instruction unit for causing the image processing unit to perform image processing of new image data in accordance with an image processing parameter included in the sample information when sample information matching the feature information is obtained by the sample information searching unit; ,
Sample information for registering new sample information including feature information of a new image and image processing parameters applied to the image in the database when sample information matching the feature information is not obtained by the sample information search means An image processing system comprising registration means.   2. The image processing system according to claim 1, further comprising image analysis means for analyzing image data and calculating the feature information.   The image processing system according to claim 2, wherein the image analysis unit includes a face detection unit that detects a human face in the image and a skin color identification unit that identifies a skin color of the detected face.   4. The image processing system according to claim 2, wherein the image analysis unit extracts shooting information added to image data as image feature information. The database is provided in each of a plurality of image processing apparatuses having the image processing means and the sample information search means,
The image processing system according to claim 1, wherein the plurality of image processing apparatuses are connected to each other via a network. The database is provided in a server having the sample information search means,
The server is connected to an image processing apparatus having the image processing means via a network,
5. The server according to claim 1, wherein when the sample information is requested from the image processing apparatus, the server retrieves the sample information from the database and transmits the sample information to the image processing apparatus. The image processing system described in 1. Image processing means for performing image processing on image data in accordance with image processing parameters;
Sample that retrieves sample information that matches feature information of a new image from a database in which sample information including image feature information and image processing parameters applied to the image is stored when new image data is input Information retrieval means;
An image processing instruction unit for causing the image processing unit to perform image processing of new image data in accordance with an image processing parameter included in the sample information when sample information matching the feature information is obtained by the sample information searching unit; ,
Sample information for registering new sample information including feature information of a new image and image processing parameters used for the image in the database when sample information matching the feature information is not obtained by the sample information search means An image processing program for causing a computer to function as registration means.

Images