JP2016092560A - Image processing apparatus and computer program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To create an appropriate composite image including characteristics of a background image and a target image.SOLUTION: An image processing apparatus comprises: an acquisition unit that acquires background image data and target image data; a region determination part that determines a composite region as a region in a background image to be composed a target image; a region division part that divides the composite region into a first kind region and a second kind region based on characteristics of a partial image in the composite region of the background image; a creation part that creates processed background image data that shows a processed background image in which the composite image obtained by composing the partial image and the target image is arranged in the composite region by determining the pixel value in the first kind region and the pixel value in the second kind region with a different processing; and an output part that outputs the processed background image data. A reflection rate of the partial image of first kind region of composite image is larger than that of the partial image of second kind region of composite image.SELECTED DRAWING: Figure 5

Description

  The present specification relates to image processing for combining a target image with a background image.

  Conventionally, various processing of image data representing an image has been performed. For example, a technique has been proposed in which a background image and a target image are combined and the combined image is printed (see, for example, Patent Document 1).

JP 2001-76127 A

  However, the actual situation is that no sufficient ingenuity has been made regarding the synthesis of images. For example, when the background image and the target image are combined, the characteristics of one image (for example, the background image) may be lost.

  This specification provides a technique for generating an appropriate composite image including the characteristics of the background image and the characteristics of the target image when the background image and the target image are combined.

  The present specification discloses the following application examples.

[Application Example 1] An image processing apparatus, an acquisition unit that acquires background image data representing a background image and target image data representing a target image, and an image processing device in which the target image is to be synthesized Based on the characteristics of the partial image in the composite area of the background image, an area determination unit that determines a composite area that is an area, and divides the composite area into a first type area and a second type area The partial image and the target image are synthesized by determining the region segmentation unit to be performed, and the pixel value in the first type region and the pixel value in the second type region by different processes. The generated composite image is a generation unit that generates processed background image data representing the processed background image arranged in the composite region, and reflects the partial image of the first type region of the composite image The rate is the second type of the composite image. The generation unit, and the processing, wherein the reflection rate of the partial image is larger than the reflection rate of the partial image in a region, and the reflection rate of the partial image is such that the value of each pixel in the partial image is reflected on the value of each pixel. And an output unit for outputting completed background image data.

  According to the above configuration, the first type region and the second type region are determined in the synthesis region based on the characteristics of the partial image of the background image. The reflection rate of the partial image in the first type region of the composite image is higher than the reflection rate of the partial image in the second type region of the composite image. As a result, an appropriate composite image including the characteristics of the partial image (that is, the background image) and the characteristics of the target image is generated.

  The technology disclosed in the present specification can be realized in various forms. For example, an image reading device, a printing device, an image processing method, a computer program for realizing these devices or methods, It can be realized in the form of a recording medium on which the computer program is recorded.

It is explanatory drawing which shows the system 900 as one Example. 5 is a flowchart of the operation of the system 900. It is explanatory drawing of the image data used by a present Example. It is a flowchart of a synthetic | combination process. It is a figure for demonstrating a synthetic | combination process. It is a figure which shows an example of the histogram of embossed image EI. It is a figure which shows an example of the processed background image CBI and the image of a comparative example. It is a flowchart of determination of the background residual ratio BR in 2nd Example. It is explanatory drawing of the table for determining the background residual ratio BR of 2nd Example. It is a flowchart of the synthetic | combination process of 3rd Example. It is a flowchart of the synthetic | combination process of 4th Example.

A. First embodiment:
FIG. 1 is an explanatory diagram showing a system 900 as an example. The system 900 includes a network 500 such as the Internet, a terminal device 100, a printer 200, and a server 300. The terminal device 100 and the server 300 are each connected to the network 500. For this reason, the terminal device 100 and the server 300 can communicate with each other.

  The terminal device 100 is a mobile terminal such as a smartphone, for example. The terminal device 100 includes a CPU 110 as a controller that controls the entire terminal device 100, a volatile storage device 120 such as a DRAM, a nonvolatile storage device 130 such as a flash memory, a display unit 140 such as a liquid crystal panel, and a liquid crystal An operation unit 150 such as a touch panel superimposed on the panel and a communication interface 180 for connecting to the network 500 (for example, a wireless communication interface compliant with the IEEE 802.11a / b / g / n standard) are provided. . The volatile storage device 120 is provided with a buffer area 122 for temporarily storing various intermediate data generated when the CPU 110 performs processing. The non-volatile storage device 130 stores a computer program 132. The computer program 132 is an application program provided in a form downloaded from a computer such as a server via the network 500, for example. Instead, the computer program 132 may be provided in a form stored in a DVD-ROM or the like.

  The CPU 110 executes the computer program 132 using the volatile storage device 120 and the nonvolatile storage device 130, thereby executing the processing of FIG. 2 described later in cooperation with the server 300.

  The printer 200 is a device that prints an image, and is a so-called inkjet printer. Instead, other types of printers (for example, so-called laser printers) may be employed.

  A server 300 as an image processing apparatus according to the present embodiment includes a CPU 310 as a controller that controls the entire server 300, a volatile storage device 320 such as a DRAM, a nonvolatile storage device 330 such as a hard disk drive, and a network 500. And a communication interface 380 (for example, a wired communication interface compliant with the IEEE 802.3 standard) for connection. The volatile storage device 320 is provided with a buffer area 322 for temporarily storing various intermediate data generated when the CPU 310 performs processing. The nonvolatile storage device 330 stores a computer program 332 and a template image data group 334. The computer program 332 is provided, for example, in a form stored on a DVD-ROM or the like, or downloaded from a computer such as another server via the network 500. The template image data group 334 includes a plurality of template image data used in image processing (FIG. 2) described later. Specifically, the template image data group 334 includes a plurality of template image data that are candidates for background image data to be described later and a plurality of template image data that are candidates for synthesis target image data to be described later. Yes.

  The CPU 310 executes the computer program 332 using the volatile storage device 320 and the nonvolatile storage device 330, thereby executing the processing of FIG. 2 described later in cooperation with the terminal device 100.

  FIG. 2 is a flowchart of the operation of the system 900. In the figure, processing by the terminal device 100 and processing by the server 300 are shown. The process of FIG. 2 is a process of generating print image data representing an image (also referred to as a print image) to be printed on the New Year's card pattern and printing the image on a postcard using the print image data. The process of FIG. 2 is started when the computer program 132 is activated in the terminal device 100.

  In S <b> 5, the CPU 110 transmits a request for a template image data group used for generating print image data to the server 300. In S <b> 10, the CPU 310 transmits a template image data group 334 to the terminal device 100 as a response to the request. As a result, the template image data group 334 is downloaded from the server 300 to the terminal device 100.

  In S15, the CPU 110 selects one background image data based on a user instruction. Specifically, the CPU 110 displays a UI screen (not shown) including a thumbnail image of an image represented by template image data that is a candidate for background image data on the display unit 140, and receives one background image from the user. Receives instructions to specify data. CPU 110 selects one background image data from a plurality of template image data in accordance with the instruction.

  FIG. 3 is an explanatory diagram of image data used in this embodiment. FIG. 3A shows an example of the background image BI represented by the background image data. The background image BI includes a background Bg and a plurality of objects Bo1 to Bo3. The objects Bo1 to Bo3 include, for example, a character Bo1, a drawing Bo2, and Bo3. In the example of FIG. 3A, the base Bg is not a simple solid image, but an image having a specific texture (texture), for example, an image including fine noise and wrinkles. The background image data is, for example, RGB image data that represents a color for each pixel with an RGB color system color value (RGB value).

  Further, the background remaining rate BR is associated with the background image data as attached data. The background residual ratio BR is a value used in a synthesis process to be described later, and is, for example, a value in the range of 0% to 60% (40% in the example of FIG. 3A). The background residual ratio BR is a value determined in advance for each background image data. For example, when the background Bg and the objects Bo1 to Bo3 in the background image BI are images having a specific texture such as the fine noise and wrinkles described above, or when the amount of edges in the background image BI is relatively large, The background residual ratio BR is determined to be a relatively large value (for example, 30% to 60%). When the background Bg and the objects Bo1 to Bo3 in the background image BI are simple solid images, or when the edge amount in the background image BI is relatively small, the background remaining rate BR is relatively A low value is determined (for example, 0% to 30%).

  In S20, the CPU 110 selects one compositing target image data based on a user instruction. Specifically, the CPU 110 displays a UI screen (not shown) including thumbnail images of images represented by a plurality of template image data that are candidates for the composition target image data on the display unit 140, and receives 1 from the user. An instruction to specify the image data to be combined is received. In accordance with the instruction, the CPU 110 selects one compositing target image data from a plurality of template image data.

  FIG. 3B shows an example of the compositing target image CI represented by the compositing target image data. In the present embodiment, the compositing target image CI is an image (also referred to as an imprint image) representing an imprint including a specific character (“Kagasho” in the example of FIG. 3B). The compositing target image CI includes a plurality of object pixels OP1 constituting an imprint and a plurality of non-object pixels BP1 not constituting an imprint. The synthesis target image data is, for example, RGB image data that represents a color for each pixel with an RGB color system color value (RGB value).

  Further, mask data is associated with the compositing target image data as attached data. The mask image FI (FIG. 3B) represented by the mask data has the same size as the template image (that is, the number of pixels in the vertical direction and the number of pixels in the horizontal direction). The mask image FI is a binary image for specifying the object pixel OP1 in the corresponding compositing target image CI. That is, each pixel of the mask image FI includes an ON pixel indicating that the corresponding pixel in the compositing target image CI is the object pixel OP1, and a corresponding pixel in the compositing target image CI is the non-object pixel BP1. One of the OFF pixels shown. In the mask image data of the present embodiment, an alpha channel that defines the transmittance for each pixel is used. For example, the value of the ON pixel is a value (for example, 255) indicating that the transmittance is 0%, and the value of the OFF pixel is a value (for example, 0) indicating that the transmittance is 100%. is there. The black area of the mask image FI in FIG. 3B indicates an area configured by ON pixels, and the white area indicates an area configured by OFF pixels.

  In S25, the CPU 110 determines a synthesis area CA in the background image BI based on a user instruction. The synthesis area CA is an area where the synthesis target image CI is synthesized, and is an area having the same size and shape as the synthesis target image CI. Specifically, the CPU 110 displays a UI screen (not shown) including the background image BI and the compositing target image CI superimposed on the background image BI on the display unit 140. In the UI screen, the user moves the compositing target image CI in the background image BI by pressing a portion of the display unit 140 where the compositing target image CI is displayed and executing a drag process. Then, on the UI screen, the user inputs a position determination instruction in a state where the composition target image CI is arranged at a desired position on the background image BI. When the position determination instruction is input, the CPU 110 determines the area in the background image BI where the composition target image CI is arranged on the UI screen as the composition area CA. In the following, the description will be continued by taking as an example the case where the synthesis area CA is determined at the position shown in FIG.

  In S <b> 30, the CPU 110 transmits a composition instruction to the server 300. In the compositing instruction, for example, information (for example, file name) for specifying the selected background image data and compositing target image data, and information for specifying the compositing area CA determined in S25 (for example, compositing) Coordinate information of the upper left vertex of the area CA).

  When server 300 receives the compositing instruction, in S35, CPU 310 acquires background image data from template image data group 334 stored in nonvolatile storage device 330 of server 300 according to the information included in the compositing instruction. . In S <b> 40, similarly, the CPU 310 acquires composition target image data from the template image data group 334 stored in the nonvolatile storage device 330 of the server 300. In S45, CPU 310 determines a synthesis area CA in background image BI according to the information included in the synthesis instruction. In S50, the CPU 310 executes a composition process.

  FIG. 4 is a flowchart of the synthesis process. FIG. 5 is a diagram for explaining the synthesis process. The composition process is a process for generating processed background image data representing a processed background image in which the composite image is arranged in the composition area CA. The synthesized image is an image obtained by synthesizing the synthesis target image CI (FIG. 3B) and the partial image (FIG. 3A) in the synthesis area CA of the background image BI.

  In S100, the CPU 310 extracts partial image data representing the partial image PI in the synthesis area CA from the background image data representing the background image BI. FIG. 5A shows an example of the partial image PI. In the example of FIG. 5A, the partial image PI includes a part of the objects Bo2 and Bo3 in the background image BI and a part of the background Bg.

  In S105, the CPU 310 generates embossed image data representing the embossed image EI using the partial image data. Specifically, the CPU 310 converts the color value (RGB value) of each pixel of the partial image PI into a luminance value Y using a predetermined conversion formula, thereby expressing the luminance for each pixel of the partial image PI. Generate data. The luminance value Y is calculated by using an RGB value (R, G, B), for example, by an equation of luminance value Y = ((0.298912 × R) + (0.586611 × G) + (0.114478 × B)). The luminance value Y takes any value from 0 to 256, for example. CPU 310 applies emboss filter FL (FIG. 4) to the luminance data of partial image PI to generate emboss image data. The pixel value after applying the emboss filter FL in FIG. 4 takes any value from − (3 × 255) to + (3 × 255), but is normalized to a value from 0 to 256. .

  FIG. 5B shows an example of the embossed image EI. In the embossed image EI, the value of a pixel constituting a region with poor features such as a single color region in the partial image PI is a value in the vicinity of the intermediate value (that is, 128). In the embossed image EI, the value of a pixel (also referred to as an edge pixel) constituting the edge of the partial image PI takes a value that is far from the intermediate value, for example, a value that is significantly larger or smaller than the intermediate value. In general, in the embossed image EI, the edge of the partial image PI is a double of a line constituted by pixels having a value larger than the intermediate value and a line constituted by pixels having a value smaller than the intermediate value. Extracted as a line. Further, in the embossed image EI, the texture (texture) included in the partial image PI is emphasized in a relief pattern. In other words, by generating the embossed image EI, features of the partial image PI such as edges and textures are extracted. Here, the texture is a pattern of a predetermined size or less expressing the above-mentioned paper wrinkles. The predetermined size is, for example, the number of pixels in the vertical direction and the horizontal direction is 10 pixels or less, and is about 1/100 of the number of pixels in the horizontal direction and the vertical direction of the compositing target image CI.

  FIG. 5B shows an example of the embossed image EI. In the embossed image EI, the feature of the partial image PI appears in a relief shape, but in FIG. 5B, the embossed image EI is illustrated in a simplified manner in order to avoid complication of the drawing. In the embossed image EI, the gray-only region indicates a region with poor features of the partial image PI, that is, a region with relatively few edges and textures. The black area indicates a characteristic area of the partial image PI and an area having a relatively large number of edges and textures. In this embossed image EI, an edge Eg2 of the object Bo2 of the partial image PI and an edge Eg3 of the object Bo3 appear. Further, the texture Be of the base Bg of the partial image PI appears in the embossed image EI.

  In S110, the CPU 310 generates a histogram of the embossed image EI. The histogram is data in which each pixel in the embossed image EI is classified into a plurality of classes according to the pixel value. In the present embodiment, in this embodiment, a histogram is generated with each of 256 gradation pixel values as one class.

  FIG. 6 is a diagram illustrating an example of a histogram of the embossed image EI. The peak Pc at the center corresponds to a portion other than the base Bg and the edges of the objects Bo2 and Bo3. The peaks (for example, Ps1 and Ps2) far from the center correspond to the edges. The closer the base Bg is to solid monochrome coating, the higher the peak Pc becomes. When the base Bg has a texture such as a pattern and is not solid black, the peak Pc is low and thick.

  In S115, the CPU 310 determines the background remaining rate BR. As described above, the background remaining rate BR is associated with background image data as attached data. The CPU 310 determines the background remaining rate BR (unit:%) associated with the background image data acquired in S35 as the background remaining rate BR used in this synthesis process.

  In S120, the CPU 310 determines a binarization threshold TH1. Specifically, the CPU 310 calculates the total pixel number AR of the embossed image EI, that is, the total pixel number AR of the partial image PI. Then, the CPU 310 determines, as the binarization threshold TH1, the minimum pixel value at which the ratio of the pixels having a value larger than the binarization threshold TH1 with respect to the total number of pixels AR is equal to or less than the background residual ratio BR. In other words, when AR pixels are arranged in descending order of the pixel value, the {AR × (BR / 100)}-th pixel is determined. Then, the determined pixel value is determined as the binarization threshold TH1.

  In S125, the CPU 310 generates area segment image data representing the area segment image AI. The area division image AI is a binary image that divides the synthesis area CA into an ON area constituted by ON pixels and an OFF area constituted by OFF pixels. In the area segmented image AI, the ratio of the number of pixels in the OFF area to the ON area (which can also be referred to as an area ratio) is substantially equal to the background remaining rate BR.

  Specifically, the CPU 310 generates region segmented image data by binarizing the value of each pixel of the embossed image EI using the binarization threshold TH1. In the binarization in this step, when the value of the pixel of interest in the embossed image EI is equal to or less than the binarization threshold TH1, the value of the pixel of interest is determined as the ON pixel that constitutes the ON region. When the value of the pixel of interest in the embossed image EI is larger than the binarization threshold TH1, the value of the pixel of interest is determined as an OFF pixel that constitutes an OFF region.

  FIG. 5C shows an example of the area segment image AI. The black area of the area segment image AI in FIG. 5C indicates the ON area, and the white area indicates the OFF area. As shown in FIG. 5C, the OFF area of the area segment image AI includes an area corresponding to the edge of the partial image PI. This is because the pixels belonging to the above-described peak Ps1 are classified as OFF pixels. In the area division image AI, an ON area and an OFF area are mixed in the area corresponding to the background Bg of the partial image PI according to the texture of the background Bg. This is because a part of the above-described peak Pc is classified as an ON pixel and the other part is classified as an OFF pixel.

  In S130 to S150, a process of synthesizing the synthesis target image CI (FIG. 3B) with the synthesis area CA in the background image BI (FIG. 3A) is executed. First, in S130, the CPU 310 selects one target pixel from a plurality of pixels in the synthesis area CA of the background image BI. In S135, the CPU 310 determines whether or not the pixel of the mask image FI (FIG. 3B) corresponding to the target pixel is an ON pixel. The corresponding pixel of the mask image FI is a pixel of the mask image FI that overlaps the target pixel when the mask image FI is arranged in the synthesis area CA of the background image BI.

  When the corresponding pixel of the mask image FI is an ON pixel (S135: YES), in S140, the CPU 310 determines that the pixel of the area segment image AI (FIG. 5C) corresponding to the target pixel is an ON pixel. Judge whether there is. The corresponding pixels of the area segment image AI are pixels of the area segment image AI that overlap the target pixel when the area segment image AI is arranged in the synthesis area CA of the background image BI.

  When the corresponding pixel of the area segment image AI is an ON pixel (S140: YES), in S145, the CPU 310 sets the value of the target pixel (RGB value) to the pixel of the compositing target image CI corresponding to the target pixel. Replace with value (RGB value). The corresponding pixel of the synthesis target image CI is a pixel of the synthesis target image CI that overlaps the target pixel when the synthesis target image CI is arranged in the synthesis area CA of the background image BI.

  When the corresponding pixel of the mask image FI is an OFF pixel (S135: NO), or when the corresponding pixel of the region segmented image AI is an OFF pixel (S140: NO), the CPU 310 does not execute S145. , The process proceeds to S150.

  In S150, the CPU 310 determines whether or not all the pixels in the synthesis area CA of the background image BI have been processed as the target pixel. If there is an unprocessed pixel (S150: NO), the CPU 310 returns to S130 and selects the unprocessed pixel as the next pixel of interest. When all the pixels have been processed (S150: YES), the CPU 310 ends the composition process. As a result of the synthesis process, processed background image data representing the processed background image CBI in which the synthesized image CMI obtained by synthesizing the partial image PI and the synthesis target image CI is arranged in the synthesis area CA is generated.

  FIG. 7 is a diagram illustrating an example of the processed background image CBI and the image of the comparative example. It can be seen that the composite image CMI is arranged in the composite area CA of the processed background image CBI in FIG. FIG. 5D shows an image (only a pixel (also referred to as a replacement pixel) that is replaced with the value of the corresponding pixel in the synthesis target image CI (FIG. 3A) in the synthesis image CMI. RI, also referred to as a replacement image) is shown.

  As can be seen from FIG. 5D, all the pixels in the synthesis area CA of the background image BI are not replaced with the pixels of the synthesis target image CI. As will be understood from the synthesis process of FIG. 4, the replacement pixel is a pixel in which the corresponding pixel of the mask image FI is ON and the area segment image AI is ON among all the pixels in the synthesis area CA. In other words, the replacement pixel is a pixel in which the corresponding pixel of the compositing target image CI is the object pixel OP1 and the corresponding pixel of the region segmented image AI is ON. The non-replacement pixel is a pixel in which at least one of the corresponding pixel of the mask image FI and the corresponding pixel of the region segmented image AI is OFF. Of the synthesis area CA, an area constituted by non-replacement pixels is also referred to as a first type area, and among the synthesis area CA, an area constituted by substitution pixels is also referred to as a second type area. As described above, the ratio of the number of pixels in the OFF region to the ON region of the region segmented image AI is determined based on the background remaining rate BR. Then, since the first type region and the second type region are determined based on the region division image AI and the predetermined mask image FI, the first type region with respect to the second type region is determined. It can be said that the ratio of the number of pixels is determined based on the background residual ratio BR.

  In the replacement image RI of FIG. 5D, the gray area OP indicates the second type area, and the white areas VP1 to VP3 indicate the first type area. The first type region includes a region VP1 corresponding to the edge of the partial image PI, a region VP2 generated according to the texture of the background Bg of the partial image PI, and a region VP3 corresponding to the OFF pixel of the mask image FI. It is out. That is, among the plurality of pixels in the partial image PI, the plurality of pixels corresponding to the object pixel OP1 of the synthesis target image CI are pixels that form an edge and pixels that form a texture (pattern of a predetermined size or less). And other pixels (mainly pixels in a solid monochrome region). Almost all the pixels constituting the edge are classified as non-replacement pixels. Pixels that make up a pattern of a predetermined size or less are classified as either replacement pixels or non-replacement pixels. Other pixels are classified as replacement pixels. In other words, among these pixels, the proportion classified into the first type region (that is, the proportion in which the pixel value is maintained in the composite image) constitutes pixels constituting the edge, a pattern having a predetermined size or less. Higher in the order of pixels to be processed and other pixels.

  When the processed background image data is generated, the CPU 310 transmits the generated processed background image data to the terminal device 100 via the network 500 in S55 of FIG. When the terminal device 100 receives the processed background image data from the server 300, in S60, the CPU 110 executes a printing process. Specifically, the CPU 110 generates print data using the processed background image data, and supplies the generated print data to the printer 200. The printer 200 prints the processed background image CBI according to the received print data. For example, the processed background image CBI shown in FIG. 7A is printed on a postcard.

  According to the first embodiment described above, the composite area CA is divided into the first type area and the second type area based on the characteristics of the partial image PI in the composite area CA of the background image BI. The More specifically, in order to extract features such as edges and textures of the partial image PI, an embossed image EI (FIG. 5B) is generated (S105 in FIG. 4), and the embossed image EI is binarized. As a result, the area segment image AI (FIG. 5C) is generated (S125 in FIG. 4). Then, based on the area division image AI and the mask image FI, the first type area and the second type area are determined (S135 and S140 in FIG. 4). As a result, the synthesis area CA is divided into a first type area and a second type area.

  The value of the pixel in the first type area in the synthesis area CA is maintained at the value of the pixel in the background image BI (that is, the partial image PI), and the pixel in the second type area in the synthesis area CA. Is replaced with the value of the corresponding pixel in the compositing target image CI, thereby generating a composite image CMI (S145 in FIG. 4). That is, the value of the pixel in the first type region and the value of the pixel in the second type region are determined by different processes. Then, processed background image data representing the processed background image CBI in which the composite image CMI is arranged in the composite area CA is output (S55 in FIG. 2).

  As a result, an appropriate composite image CMI including the characteristics of the partial image PI of the background image BI and the characteristics of the compositing target image CI is generated.

  More specifically, the CPU 310 extracts edges (for example, Eg2 and Eg in FIG. 5B) in the partial image PI by generating the embossed image EI. Then, the CPU 310 binarizes the embossed image EI to generate an area segmented image AI, thereby determining the extracted edge area as the first type area (FIG. 5C).

  As a result, it is possible to prevent the edge in the partial image PI of the background image BI from being lost from the composite image CMI. Therefore, it is possible to suppress the appearance of the composite image CMI, and thus the processed background image CBI, from being deteriorated.

  FIG. 7B shows an example of the image SBI of the comparative example. In the image SBI, when the composition target image CI is superimposed on the composition area CA of the background image BI, the value of the pixel in the composition area CA that overlaps the object pixel OP1 of the composition target image CI is replaced with the value of the object pixel. Thus, the image is generated. In the image SMI (also referred to as a simple synthesized image) in the synthesis area CA of the image SBI, the edge of the partial image PI may be lost due to the synthesis of the object (frame or character) of the synthesis target image CI. I understand. In contrast, it can be seen that the edge of the partial image PI is not lost in the composite image CMI of the processed background image CBI of the present embodiment.

  In addition, the CPU 310 extracts the texture (for example, the texture of the background Bg) shown in the partial image PI by generating the embossed image EI. That is, a pattern of a predetermined size or less included in the partial image PI is extracted. Then, the CPU 310 binarizes the embossed image EI to generate an area segmented image AI, whereby the first type area and the second type area are determined according to the extracted pattern of a predetermined size or less. .

  As a result, it is possible to suppress the loss of the texture shown in the partial image PI from the composite image CMI. Therefore, it is possible to suppress the appearance of the composite image CMI, and thus the processed background image CBI, from being deteriorated.

  It can be said that the processes from S105 to S125 in FIG. 4 are classification processes for classifying the pixels in the partial image PI into replacement pixels and non-replacement pixels. Then, the CPU 110 determines the value of the pixel in the first type region configured by the replacement pixel as the pixel value of the partial image PI, and the value of the pixel in the second type region configured by the non-replacement pixel. Is determined as the pixel value of the compositing target image CI, thereby synthesizing the partial image PI and the compositing target image CI (S135 to S145 in FIG. 4). As a result, it is possible to appropriately synthesize the partial image PI and the synthesis target image CI and generate an appropriate synthesized image including the features of the partial image PI and the characteristics of the synthesis target image CI.

  An embossing filter is used for this classification process. Thereby, since the edge and the texture in the partial image PI can be extracted appropriately, the first type region and the second type region are appropriately determined according to the edge and the texture in the partial image PI. Can do. Accordingly, it is possible to appropriately suppress the loss of the edge and texture in the partial image PI from the composite image CMI.

  In S15 of FIG. 2, the background image data representing the background image BI used for the composition process is selected from a plurality of template image data included in the template image data group 334. These background image data are associated with a background remaining rate BR as attached data (FIG. 3A). Then, as described above, the ratio of the number of pixels of the first type region to the second type region is determined based on the background remaining rate BR. In other words, when the first template image data is selected as the background image data, the ratio of the first type area to the second type area is determined to be the first ratio, and the second template When image data is selected, the ratio is determined as the second ratio. As a result, an appropriate composite image CMI can be generated according to the selected template image data.

  For example, a relatively large value of the background remaining rate BR (for example, 40%) is associated with template image data representing a template image including a background including a paper wrinkled texture. As a result, when the template image data is selected as the background image data, the ratio of the first type area in the composite image CMI is high in the composite image CMI, so the composite image CMI is printed on wrinkled paper. It becomes an image that expresses a faint impression as when a seal stamp is pressed. Further, a relatively low value of the background remaining rate BR (for example, 10%) is associated with template image data representing a template image that does not include a paper wrinkle-like texture and includes a solid background. As a result, when the template image data is selected as background image data, the ratio of the first type region in the composite image CMI is low in the composite image CMI, so the composite image CMI is printed on paper without wrinkles. As in the case of pressing a seal, the image represents a clear imprint with no blur.

B. Second Embodiment In the first embodiment, the background remaining rate BR is associated with background image data as attached data, but the background remaining rate BR may be determined according to the characteristics of the partial image PI. . FIG. 8 is a flowchart for determining the background remaining rate BR in the second embodiment. FIG. 9 is an explanatory diagram of a table for determining the background remaining rate BR of the second embodiment.

  In the second embodiment, the determination of the background remaining rate BR in S115 of FIG. 4 is performed by processing in the flowchart of FIG. In S200, the CPU 310 calculates the standard deviation σ of the pixel value of the embossed image EI. The standard deviation σ is calculated using the following formula (1). Vave in Expression (1) is an average value of the values of all the pixels of the embossed image EI. Further, n in the formula (1) is the total number of pixels of the embossed image EI. Vi in Expression (1) is a value of each of n pixels of the embossed image EI (1 ≦ i ≦ n). As the variation in the pixel value of the partial image PI increases, the variation in the value of each pixel in the embossed image EI also increases. Therefore, the standard deviation σ, which is a feature amount indicating the variation in the value of each pixel in the embossed image EI, It can also be said to be a feature amount indicating variation in pixel values of the image PI.

  In S210, the CPU 310 determines the background remaining rate BR based on the standard deviation σ. Specifically, the CPU 310 refers to a table that defines the correspondence relationship between the standard deviation σ and the background remaining rate BR shown in FIG. 9, and determines the background remaining rate BR corresponding to the calculated standard deviation σ. This table is preinstalled in the computer program 332. As shown in FIG. 9, when the standard deviation σ is 0, the background remaining rate BR is determined to be the minimum value BRmin, and when the standard deviation σ is greater than or equal to the specific value σs, the background remaining rate BR is the maximum The value BRmax is determined. In the range of 0 ≦ σ ≦ σs, the background remaining rate BR is determined so as to increase linearly as the standard deviation σ increases. The values of BRmin, BRmax, and σs are, for example, the background residual ratio BR is changed for a sample of the processed background image CBI that uses a plurality of samples of the background image BI that are different from each other and the synthesis target image CI sample. And experimentally determined by evaluating the appearance of the generated processed background image CBI sample. BRmin is, for example, 0%, and BRmax is, for example, 60%.

  According to the second embodiment described above, the background remaining rate BR is determined according to the characteristics of the partial image PI. That is, the ratio of the first type region to the second type region in the synthesis region CA is determined according to the characteristics of the partial image PI. As a result, according to the feature of the partial image PI, the extent to which the feature of the partial image PI remains in the composite image CMI can be appropriately determined.

  Specifically, the background residual ratio BR is increased as the standard deviation σ, which is a feature amount indicating variations in the pixel values of the partial image PI, is increased (FIG. 9). Therefore, the larger the standard deviation σ, the larger the ratio of the first type region to the second type region. That is, the extent to which the characteristics of the partial image PI are left in the composite image CMI is increased. In general, it is considered that it is preferable to increase the degree of remaining characteristics of the partial image PI in the composite image CMI as the variation in the pixel values of the partial image PI increases. The greater the variation in the pixel values of the partial image PI, the higher the possibility that the partial image PI contains more edges. In addition, the larger the variation in the pixel values of the partial image PI, the higher the possibility that a characteristic texture is included in the partial image PI. Therefore, the greater the variation in the pixel values of the partial image PI, the higher the possibility that the appearance of the composite image CMI can be improved by leaving the features of the partial image PI such as edges and textures in the composite image CMI. In addition, when the variation in the pixel values of the partial image PI is relatively low, for example, when the partial image PI is a solid-colored single color area, the significance of leaving the characteristics of the partial image PI in the composite image CMI is significant. Is relatively low. If the level of the characteristic of the partial image PI is excessively increased, the characteristic of the synthesis target image CI may be impaired in the synthesized image CMI. Therefore, the variation in the pixel values of the partial image PI is relatively low. In such a case, it may be preferable to make the degree of leaving the characteristics of the partial image PI relatively low. As can be seen from the above description, according to the second embodiment, a more preferable composite image CMI can be generated in accordance with variations in the pixel values of the partial image PI.

In the second embodiment, the standard deviation σ of the pixel value of the embossed image EI is used as the feature amount indicating the variation of the pixel value of the partial image PI. A standard deviation σ of pixel values, a variance σ ² , an average edge amount, and the like may be used.

C. Third Example In the first example, the background remaining rate BR is associated with the background image data as attached data. Instead, in the third embodiment, the background image data is associated with attached data as process designation information indicating whether simple synthesis processing or special synthesis processing is performed. For example, when the background or object in the background image BI is an image having a specific texture (texture) such as fine noise or wrinkles, the processing designation information indicating that the special composition processing is performed includes the background image BI. Is associated with the background image data. When the background Bg and the objects Bo1 to Bo3 in the background image BI are simple solid images, the process designation information indicating that simple composition processing is performed is background image data representing the background image BI. Is associated with.

  FIG. 10 is a flowchart of the synthesis process according to the third embodiment. In S300, the CPU 310 determines whether or not the background image BI to be used for the composition process is a target for the special composition process. Specifically, the CPU 310 refers to the process designation information associated with the background image data representing the background image BI to be used for the synthesis process. When the process designation information indicates that the special composition process is to be performed, the CPU 310 determines that the background image BI is the target of the special composition process. When the process designation information indicates that simple composition processing is to be performed, the CPU 310 determines that the background image BI is not a target for special composition processing.

  When the background image BI is the target of the special synthesis process (S300: YES), in S310, the CPU 310 performs the synthesis process of S100 to S150 of FIG. 4 described in the first embodiment as the special synthesis process. Run. In the third embodiment, the background remaining rate BR determined in S115 of FIG. 4 is a fixed value (for example, 40%). When the background image BI is not the target of the special synthesis process (S300: NO), in S320, the CPU 310 executes a simple synthesis process. The simple combining process is a process for generating image data representing the image SBI of the comparative example in FIG. That is, when the composition target image CI is superimposed on the composition area CA of the background image BI, the CPU 310 replaces the value of the pixel in the composition area CA that overlaps the object pixel OP1 of the composition target image CI with the value of the object pixel OP1. As a result, processed background image data is generated.

  According to the third embodiment, when the background image BI for which it is preferable to use the simple synthesis process is used, the simple synthesis process is executed, and when the background image BI for which the special synthesis process is preferably used is used. The special synthesis process is executed. Therefore, appropriate processed background image data can be generated in accordance with the background image BI to be used for the synthesis process.

D. Fourth Embodiment The synthesizing process of the fourth example is different from the synthesizing process of the first example in the process of generating the area segment image AI. FIG. 11 is a flowchart of the synthesis process according to the fourth embodiment. In S400, as in S100 of FIG. 4, the CPU 310 extracts partial image data representing the partial image PI in the synthesis area CA from the background image data representing the background image BI.

  In S405, the CPU 310 calculates a representative color value (RGB value) of the partial image PI. For example, an average color value (Rav, Gav, Bav) including an average value Rav of R component values of all pixels of the partial image PI, an average value Gav of G component values, and an average value Bav of B component values is Calculated as the representative color value of the partial image PI.

  In S410, the CPU 310 calculates a representative color value of the compositing target image CI. For example, like the representative color value of the partial image PI, the average color value of all the pixels of the compositing target image CI is calculated as the representative color value of the compositing target image CI.

  In S415, the CPU 310 calculates a distance D between the representative color value of the partial image PI and the representative color value of the compositing target image CI. Specifically, the Euclidean distance D in the RGB color space of these two color values is calculated.

  In S420, based on the distance D, the CPU 310 selects an area division pattern to be used from a plurality of predetermined area division patterns. The area division pattern is a binary image in which an OFF area and an ON area are divided, like the area division image AI of the first embodiment. In the area segmentation pattern, the OFF area and the ON area are arranged so as to express a pattern that appears in a faint impression, for example. In the present embodiment, as shown in FIG. 11, an area division pattern to be used is selected from three types of area division patterns AP1 to AP3. In the area division patterns AP1 to AP3 in FIG. 11, the black area indicates the ON area, and the white area indicates the OFF area. Between these area division patterns AP1 to AP3, the ratio of the number of pixels in the OFF area to the ON area is different. The ratio increases stepwise in the order of the patterns AP1, AP2, and AP3, and is, for example, 10%, 25%, and 40%.

  In the present embodiment, when the distance D is within a relatively small first range (for example, D <D1), the area division pattern AP1 in which the ratio of the number of pixels in the OFF area to the ON area is relatively small is selected. The When the distance D is in the middle second range (for example, D1 ≦ D <D2), the area division pattern AP2 having a medium ratio of the number of pixels in the OFF area to the ON area is selected. When the distance D is within a relatively large third range (for example, D2 ≦ D), the area division pattern AP3 having a relatively large ratio of the number of pixels in the OFF area to the ON area is selected. That is, as the distance D is larger, a region division pattern having a larger ratio is selected.

  In S425, the CPU 310 generates an area segment image AI using the selected area segment pattern. Specifically, the CPU 310 generates an area segment image AI having the same size as the partial image PI by arranging and arranging the selected area segment patterns by the size of the partial image PI (that is, the size of the synthesis area CA). To do. Therefore, the ratio of the number of pixels in the OFF area to the ON area in the area segment image AI is the same as the ratio in the selected area segment pattern.

  When the region segmented image AI is generated, the processing of S130 to S150 of FIG. 4 is executed in S430.

  According to the fourth embodiment described above, the ratio of the number of pixels in the OFF region to the ON region in the region segmented image AI is determined according to the characteristics of the partial image PI. More specifically, the ratio of the number of pixels in the OFF area to the ON area increases as the distance D, which is a feature amount indicating the difference between the color of the partial image PI and the color of the synthesis target image, increases. As a result, as the difference between the color of the partial image PI and the color of the synthesis target image increases, the ratio of the number of pixels of the first type area to the second type area in the synthesis area CA increases. As a result, a more preferable composite image CMI can be generated according to the difference between the color of the partial image PI and the color of the compositing target image CI.

  More specifically, when the difference between the color of the partial image PI and the color of the compositing target image CI is large, it is considered that the significance of leaving the characteristics of the partial image PI in the composite image CMI is increased. This is because when the feature of the partial image PI is left in the composite image CMI, the feature is easily noticeable. For this reason, when the difference between the color of the partial image PI and the color of the synthesis target image CI is large, the ratio of the number of pixels of the first type region to the second type region in the synthesis region CA may be increased. preferable. On the other hand, if the difference between the color of the partial image PI and the color of the compositing target image CI is small, the significance of leaving the characteristics of the partial image PI in the composite image CMI becomes small. It is preferable to suppress damage. For this reason, when the difference between the color of the partial image PI and the color of the synthesis target image CI is small, the ratio of the number of pixels of the first type region to the second type region in the synthesis region CA may be reduced. preferable. Therefore, according to the fourth embodiment, a more preferable synthesized image CMI can be generated according to the difference between the color of the partial image PI and the color of the synthesis target image CI.

E. Variations:
(1) In each of the above embodiments, in the composite image CMI, the pixel value in the first type region is the pixel value of the partial image PI of the background image BI, and the pixel value in the second type region is used. As the value, the value of the pixel of the compositing target image CI is used. Instead, a weighted average value of the corresponding pixel value of the partial image PI and the corresponding pixel value of the compositing target image CI may be used as the pixel value in the composite image CMI.

  For example, generally, the pixel values (Rm, Gm, Bm) in the composite image CMI are the corresponding pixel values (Rp, Gp, Bp) of the partial image PI and the corresponding pixel values of the compositing target image CI ( Rc, Gc, Bc) may be used to calculate using the following equation (2).

Rm = Wp × Rp + Wc × Rc
Gm = Wp × Gp + Wc × Gc
Bm = Wp × Bp + Wc × Bc (2)

  Here, Wp is a weight for the pixel of the partial image PI, Wc is a weight for the pixel of the compositing target image CI, and (Wp + Wc) = 1. Here, among the pixels in the composite image CMI, the weight Wp when determining the pixels in the first type region is set as the weight Wp1 of the first type region. Further, the weight Wp when determining the pixels in the second type area is set as the weight Wp2 of the second type area. In this case, the weight Wp1 of the first type region may be set larger than the weight Wp2 of the second type region (Wp1> Wp2). The weight Wp can be said to be the reflection rate of the partial image PI indicating the degree to which the value of each pixel of the partial image PI is reflected with respect to the value of each pixel in the composite image CMI. In this way, it is possible to generate an appropriate composite image CMI including the characteristics of the partial image PI (that is, the background image) and the characteristics of the compositing target image CI.

  Here, among the pixels in the composite image CMI, the weight Wc when determining the pixel in the first type region is set as the weight Wc1 of the first type region. Further, the weight Wc when determining the pixels in the second type region is set as the weight Wc2 of the second type region. More preferably, the weight Wp1 of the first type region is set larger than the weight Wc1 of the first type region, and the weight Wc1 of the second type region is set larger than the weight Wp1 of the second type region (Wp1). > Wc1 and Wc2> Wp2). For example, Wp1 = 0.8, Wc1 = 0.2, Wc2 = 0.8, and Wp2 = 0.2 are set. The weight Wc can be said to be a reflection rate of the compositing target image CI indicating the degree to which the value of each pixel of the compositing target image CI is reflected with respect to the value of each pixel in the composite image CMI. By doing so, it is possible to generate a composite image CMI in which the characteristics of the partial image PI (that is, the background image) and the characteristics of the compositing target image CI are more appropriately left.

  In each of the above embodiments, the value of the pixel in the first type region is calculated by using Equation (2) assuming that Wp1 = 1 and Wc1 = 0, and the value of the pixel in the second type region is , Wp2 = 0 and Wc2 = 1, which is equivalent to the case where the calculation is performed using Expression (2). In other words, each of the above embodiments satisfies (Wp1> Wp2) and also satisfies (Wp1> Wc1 and Wc2> Wp2).

(2) In the first embodiment, the background remaining rate BR is data attached to the background image data, but the background remaining rate BR may be data attached to the synthesis target image data. In this case, the ratio of the number of pixels of the first type region to the second type region is determined based on the background remaining rate BR associated with the synthesis target image data. In other words, when the first template image data is selected as the compositing target image data, the ratio of the first type area to the second type area is determined as the first ratio, When template image data is selected, the ratio is determined to be the second ratio.

  In the third embodiment, the process designation information indicating whether simple synthesis processing or special synthesis processing is performed is attached data of the background image data, but the background remaining rate BR is the value of the synthesis target image data. It may be attached data. In this case, in S300 of FIG. 10, the CPU 310 may determine whether to perform the special combining process or the simple combining process in accordance with the process designation information as the attached data of the combining target image data.

(4) In the composition processing (FIG. 2) of the first embodiment, an embossed image EI is generated by applying an emboss filter to the partial image PI, and the embossed image EI is binarized, thereby obtaining a region. A segmented image AI is generated. Instead, an edge image is generated by applying an edge extraction filter such as a Sobel filter or a Prewitt filter to the partial image PI, and an area segment image is generated by binarizing the edge image. Also good. This process is suitable, for example, when the composite image CMI is generated so as to leave the edge of the partial image PI and not leave the texture features included in the partial image PI.

(5) In the above embodiment, both the background image data and the compositing target image data are selected from a plurality of template image data. Alternatively, one or both of the background image data and the compositing target image data may be other image data, for example, scan data obtained by reading a user's original with a scanner.

(6) In the above-described embodiment, the use of the processed background image data is printing, but is not limited to this, and any use can be adopted. For example, the CPU 110 of the terminal device 100 may store the acquired processed background image data in the nonvolatile storage device 130 for future use. The processed background image data may be output to a device other than the terminal device 100, for example, another server connected to the network 500 or a user's personal computer (not shown).

(7) The image processing executed by the server 300 of the above embodiment is realized by a device of a different type from the server 300 (for example, the terminal device 100, a printer, a digital camera, a scanner, a multifunction device, a personal computer, a mobile phone). Also good. For example, the CPU 110 of the terminal device 100 may execute all image processing performed by the server 300 in the above embodiment. Further, the CPU of the printer 200 may execute all the image processing executed by the CPUs 110 and 310 of the terminal device 100 and the server 300 in the above embodiment.

(8) The server 300 may include a plurality of devices (for example, computers) that can communicate with each other via a network. The image processing executed by the server 300 can be executed in part by a plurality of devices. In this case, a plurality of devices function as one image processing device.

(9) In each of the above embodiments, a part of the configuration realized by hardware may be replaced with software, and conversely, part or all of the configuration realized by software is replaced with hardware. You may do it. For example, all or part of the image processing executed by the server 300 may be realized by a dedicated hardware circuit including a logic circuit.

  As mentioned above, although this invention was demonstrated based on the Example and the modification, Embodiment mentioned above is for making an understanding of this invention easy, and does not limit this invention. The present invention can be changed and improved without departing from the spirit and scope of the claims, and equivalents thereof are included in the present invention.

  DESCRIPTION OF SYMBOLS 100 ... Terminal device, 110 ... CPU, 120 ... Volatile memory device, 122 ... Buffer area, 130 ... Nonvolatile memory device, 132 ... Computer program, 140 ... Display 150, operation unit, 180 ... communication interface, 200 ... printer, 300 ... server, 310 ... CPU, 320 ... volatile storage device, 322 ... buffer area, 330 ... Nonvolatile storage device, 332 ... Computer program, 334 ... Template image data group, 380 ... Communication interface, 500 ... Network, 900 ... System

Claims (10)

An image processing apparatus,
An acquisition unit that acquires background image data representing a background image and target image data representing a target image;
An area determination unit that determines a synthesis area that is an area in the background image to be synthesized with the target image;
Based on the characteristics of the partial image in the synthesis area of the background image, the synthesis area is divided into a first type area and a second type area,
A composite image obtained by combining the partial image and the target image by determining the value of the pixel in the first type region and the value of the pixel in the second type region by different processes, A generating unit configured to generate processed background image data representing a processed background image arranged in the composite region, wherein the reflection rate of the partial image in the first type region of the composite image is the composite image Greater than the reflection rate of the partial image of the second type region, and the reflection rate of the partial image is such that the value of each pixel in the partial image is reflected with respect to the value of each pixel. A generator,
An output unit for outputting the processed background image data;
An image processing apparatus comprising: The image processing apparatus according to claim 1,
The area classification unit is an image processing apparatus that extracts an edge in the partial image and determines the extracted area of the edge as the first type area. The image processing apparatus according to claim 1, wherein:
The region classification unit includes a classification process for classifying a plurality of pixels in the partial image into a first type pixel and a second type pixel, and a process of applying a specific filter to the partial image. Perform the classification process,
The generation unit determines a value of a pixel in the first type region configured by the first type pixel as a value of a pixel of the partial image, and the second type of pixel configured by the second type pixel. An image processing apparatus that synthesizes the partial image and the target image by determining pixel values in two types of regions as pixel values of the target image. The image processing apparatus according to claim 3,
The area section is
By executing the classification process including the process of applying the specific filter to the partial image, the synthesis area is defined as the first type area and the first type according to a pattern having a predetermined size or less included in the partial image. An image processing apparatus that divides into two regions. The image processing apparatus according to claim 4,
The image processing apparatus, wherein the specific filter is an emboss filter. An image processing apparatus according to any one of claims 1 to 5,
The region classification unit is an image processing device that determines a ratio of the first type region to the second type region according to a feature in the partial image. The image processing apparatus according to claim 6,
The image processing apparatus according to claim 1, wherein the area classification unit increases the ratio of the first type area as the feature amount indicating the variation in the value of the pixel in the partial image increases. The image processing apparatus according to claim 6,
The image processing apparatus, wherein the area dividing unit increases the ratio of the first type area as the feature amount indicating the difference between the color of the partial image and the color of the target image increases. The image processing apparatus according to claim 1,
One of the acquired background image data and the target image data is selected from a plurality of template image data,
The area section is
When the first template image data is selected, the ratio of the first type area to the second type area is determined as the first ratio,
When the second template image data is selected, the image processing apparatus determines a ratio of the first type area to the second type area as a second ratio. A computer program,
An acquisition function for acquiring background image data representing a background image and target image data representing a target image;
A region determination function for determining a composite region that is a region in the background image to be combined with the target image;
A region dividing function for dividing the composite region into a first type region and a second type region based on a feature of a partial image in the composite region of the background image;
A composite image obtained by combining the partial image and the target image by determining the value of the pixel in the first type region and the value of the pixel in the second type region by different processes, A generation function for generating processed background image data representing a processed background image arranged in the composite region, wherein the reflection rate of the partial image in the first type region of the composite image is the composite image Greater than the reflection rate of the partial image of the second type region, and the reflection rate of the partial image is such that the value of each pixel in the partial image is reflected with respect to the value of each pixel. Generation function,
An output function for outputting the processed background image data;
A computer program that causes a computer to realize