JP2024080365A - Image processing device, control method thereof, and program - Google Patents

Abstract

[Problem] To provide an image processing device that suppresses color changes such as filling in the background of characters when the image is converted to PDF (high compression), and a control method and program for the same. [Solution] The image processing device is characterized by comprising a detection unit (303) that detects a character area from input multi-value image data, an extraction unit (304) that extracts characters from the detected character area and obtains a unit character area within the character area in which a single character exists, an assignment unit (308) that assigns a conversion color determined based on the color distribution of the background color within the character area other than the extracted characters to a "unit character area" within the character area in a reduced multi-value image obtained by reducing the multi-value image data, and outputs the converted color, and a color conversion unit (311) that replaces a color equivalent to the background color in the character area in an image obtained by compressing the output image, with the conversion color. [Selected Figure] Figure 3

Description

The present invention relates to an image processing device capable of compressing an input multi-valued image while suppressing color changes, and a control method and program for the same.

In recent years, the spread of scanners has led to the digitization of paper documents. Generally, color images have large file sizes, so currently, methods of compressing images using JPEG compression, etc., are widespread. However, while JPEG compression is a very effective method for compressing natural images such as photographs, when text areas are compressed using JPEG compression, image degradation known as "mosquito noise" occurs.

For this reason, an image compression method has been proposed, as disclosed in Patent Document 1. In this method, the input image is divided into a "text region," a "photographic region," and a "background region." The "text region" is then binarized, after which the "photographic region" is compressed using MMR compression, and the "background region" is compressed using JPEG compression. This makes it possible to represent full-color images in small file sizes while maintaining the quality of the "text region."

Here, data obtained by converting "MMR compressed data (also referred to as "MMR data")," "JPEG compressed data (also referred to as "JPEG data")," and "text color data in text areas" into PDF format is referred to as "PDF (high compression)." Furthermore, the text color in the "text area" is extracted from the input image or a reduced multi-value image obtained by reducing the input image, and this is referred to as "text color extraction."

JP 2013-125994 A

As mentioned above, the "photo area" and "background area" are reduced in size and compressed using JPEG. Compared to conventional JPEG compression of the entire image, JPEG compression of only the areas other than the "text area" prevents image degradation in the "text area," resulting in a higher compression rate and smaller file size.

However, areas other than "text areas" are compressed at a higher rate than before, and because the image is reduced, "block noise" is more likely to occur than before. For this reason, PDF (high compression) may cause color changes to be more noticeable than the conventional JPEG compression of the entire image. In particular, when filling in the background of a "text area," the color filling process of the "text area" is also performed on a reduced multi-value image, which can cause apparent color changes.

The object of the present invention is to provide an image processing device that suppresses color changes when filling in the background of characters in PDF (high compression), as well as a control method and program for the same.

In order to achieve the above object, the present invention is an image processing device capable of extracting a rectangular area containing characters as a character area, and is characterized by comprising: a detection unit that detects a character area from input multi-value image data; a cut-out unit that cuts out characters from within the character area detected by the detection unit and obtains a unit character area within the character area, which is an area containing a single character; an assignment unit that assigns a conversion color determined based on the color distribution of the background color within the character area other than the characters cut out by the cut-out unit to the unit character area within the character area in a reduced multi-value image obtained by reducing the multi-value image data, and outputs the converted color; and a color conversion unit that replaces a color equivalent to the background color in the character area in an image obtained by compressing the output image by the assignment unit with the converted color.

The present invention has the effect of realizing an image processing device, a control method and a program for suppressing color changes when filling in the background of text in a PDF (high compression) by extracting the background color of the text area and replacing it with a specific color.

FIG. 1 is a configuration diagram of an image processing system. FIG. 2 is a diagram illustrating the configuration of an MFP according to the first embodiment. FIG. 2 is a functional configuration diagram of a data processing unit 215 in the first embodiment. 5 is a flowchart showing a process of a data processing unit 215 in the first embodiment. FIG. 2 is a schematic explanatory diagram of an image sample in the first embodiment. FIG. 4 is an explanatory diagram of a color conversion table in the first embodiment. 5 is an explanatory diagram of character color extraction in the first embodiment. FIG. 5 is an explanatory diagram of character color extraction in the first embodiment. FIG. 10 is a flowchart showing a JPEG data generation process in the second embodiment.

The following describes in detail the embodiments of the present invention with reference to the drawings. However, the configurations described in the following embodiments are merely examples, and the scope of the present invention is not limited to the configurations described in the embodiments. First, the first embodiment of the present invention will be described.

First Embodiment
<Image Processing System>
Fig. 1 is a configuration diagram of an image processing system according to a first embodiment. As shown in Fig. 1, in the image processing system, an MFP (multifunction peripheral) 101 and a client PC 102 are connected by a network 103 so that required information can be communicated between them. Dotted lines 104 and 105 in Fig. 1 indicate a process flow. The dotted line 104 indicates a process in which a user uses the scanner function of the MFP 101 to read a paper document as a scanned image. At this time, the user can operate an operation unit (203 in Fig. 2) of the MFP 101, which will be described later, to set a destination (e.g., the client PC 102) to which the scanned image is to be sent, and various settings related to scanning and sending.

These various settings include color mode, file format (e.g., JPEG, TIFF, PDF, PDF (high compression)), etc. In this embodiment, we will explain the case where PDF (high compression) is specified as the data format. Details of the PDF (high compression) processing will be described later.

The dotted line 105 indicates the process in which the MFP 101 generates data using software or hardware based on the various specified settings and transmits it to the specified destination. Here, the image transmitted to the client PC 102 is transmitted in a common file format such as PDF, so that it can be viewed using a general-purpose viewer that the client PC 102 has.

<Configuration of MFP>
2 is a diagram showing the configuration of the MFP 101 in the first embodiment. The MFP 101 has a scanner unit 201 as an image input device, a printer unit 202 as an image output device, a control unit 204 that controls the entire MFP, and an operation unit 203 that serves as a user interface. The control unit 204 is a controller that is connected to the scanner unit 201, the printer unit 202, and the operation unit 203, and also has a function of inputting and outputting image information and device information by connecting to a LAN 209.

The CPU 205, RAM 206, operation unit I/F 207, network I/F 208, ROM 210, storage unit 211, and data processing unit 215 are connected to a system bus 216 inside the control unit 204, and are capable of communicating required information between them. The CPU 205 is a processor that controls the entire system. The RAM 206 is a system work memory for the operation of the CPU 205, and also functions as an image memory for temporarily storing image data. The ROM 210 is a boot-type ROM in which programs such as a system boot program are stored. The storage unit 211 is a non-volatile storage medium such as a hard disk drive, and stores system control software, image data, etc. The CPU 205 executes the system control programs stored in the storage unit 211 to realize various functions of the MPF 101.

The operation unit I/F 207 is an interface with the operation unit (UI) 203, and outputs image data to be displayed on the operation unit 203 to the operation unit 203. The operation unit I/F 207 also has a function of sending information on user operations on this device via the operation unit 203 to the CPU 205. The network I/F 208 connects this device to a LAN 209, and inputs and outputs data to and from outside the device. For example, it is used to transmit compressed data in PDF format to another device, and to receive compressed data in PDF format from another device. The data processing unit 215 will be described later.

The ImageBusI/F 212 is a bus bridge that connects the system bus 216 and an image bus 217 that transfers image data at high speed, and converts the data structure. The image bus 217 is composed of, for example, a PCI bus, IEEE 1394, etc. The RIP unit 213 and the device I/F 214 are connected to the image bus 217. The RIP unit 213 performs rendering processing, which is processing that analyzes PDL (page description language) code and develops it into a bitmap image of a specified resolution. The device I/F 214 is connected to the scanner unit 201 via a signal line 218, and is also connected to the printer unit 202 via a signal line 219, and performs synchronous/asynchronous conversion of image data.

<Data Processing Unit 215>
The data processing unit 215 generates a PDF (highly compressed) by performing processes such as area determination, compression processing, and PDF file generation. The generated PDF (highly compressed) is transmitted to a specified destination (e.g., the client PC 102) via the network I/F 208 and LAN 209. The data processing unit 215 can also decompress compressed data received via the network I/F 208 and LAN 209. The decompressed image is transmitted to the printer unit 202 via the RIP unit 213 and device I/F 214 and printed.

<Configuration of data processing unit 215>
Next, the configuration of an image processing device that generates PDF (high compression) from input data, which is realized by the data processing unit 215, will be described with reference to Fig. 3. Fig. 3 shows each processing unit for generating output data (PDF (high compression)) from input data (RGB multi-value image data). Each processing unit in Fig. 3 is realized by the CPU 205 reading and executing a computer program stored in the ROM 210 or the storage unit 211. Note that a part or all of them can also be configured with hardware such as an ASIC.

The data processing unit 215 has a gray conversion unit 301, a binarization unit 302, a first area determination unit 303, a second area determination unit 304, an MMR compression unit 305, a character color extraction unit 306, a reduction unit 307, a character area filling unit 308, and a JPEG compression unit 309. The data processing unit 215 further has a color conversion unit 311 and a final stage PDF generation unit 310.

First, an example of input data (RGB multi-value image data) read by the scanner unit 201 is shown as "500" in Figure 5. The code "500" is an example of an image including the character string "EF" shown as "501", the character string "EF" on a colored background as "502", the white character string "EF" shown as "503", and the image "504" including a photograph.

<Gray conversion unit 301>
The gray conversion unit 301 generates gray multi-value image data from the input data (RGB multi-value image data) read by the scanner unit 201. In this embodiment, a luminance Y defined by the following equation (Equation 1) in a general YUV color space is used as a gray signal, but the gray conversion is not limited to this.

Y = 0.299 x R + 0.587 x G + 0.114 x B ... (Formula 1)

<Binarization Unit 302>
The binarization unit 302 generates binary image data from the gray multi-value image data obtained from the gray conversion unit 301. The binarization method in this embodiment calculates a single threshold value from a histogram obtained from the gray multi-value image data. The gray multi-value image data is then binarized using this calculated threshold value. The reference numeral "510" in Fig. 5 indicates the result of binarization by the binarization unit 302.

<First Area Determination Unit 303>
The first area determination unit 303 detects a "character area" and a "photo area" from the binary image data generated by the binarization unit 302. This provides "character area information (X, Y, W, H)," information indicating an inverted character area (hereinafter, "inverted character area information"), and "photo area information (X, Y, W, H).""X,Y" in the "character area information (X, Y, W, H)" represents the "coordinate position (X, Y)" of the upper left corner of the rectangular area that is the detected character area, and "W, H" represent the "width" (horizontal length) and "height" (vertical length) of the rectangular area, respectively. Reference numeral "520" in FIG. 5 indicates a detection example in which a "character area" and a "photo area" are detected by the first area determination unit 303. Reference numerals "521,""522," and "523" indicate "character area information (X, Y, W, H)," and reference numeral "524" indicates "photo area information (X, Y, W, H)." The code "523" indicates that it is an "inverted character area" since it has "inverted character information".

This process is disclosed in a known area identification method (for example, JP-A-06-068301) and is carried out using this method. For example, an overview of the case where area determination is performed on binary image data 510 will be described. (1) "Black pixel blocks" are detected by tracing the contours of black pixels. As a result, six "black pixel block 1", "black pixel block 2", "black pixel block 3", "black pixel block 4", "black pixel block 5", and "black pixel block 6" are obtained as shown by reference numeral "530" in Figure 5.

(2) The obtained black pixel blocks are classified into "characters," "inverted characters," or "photographs" using at least one of the following: size, shape, and black pixel density. For example, a "black pixel block" whose aspect ratio is close to "1" and whose size is within a set range is determined to be a "black pixel block" that constitutes "characters." A pixel block that is rectangular in shape and has a high black pixel density is determined to be an "inverted character." The remaining "black pixel blocks" are then determined to be "black pixel blocks" that constitute a "photograph." As a result, "black pixel block 1," "black pixel block 2," "black pixel block 3," and "black pixel block 4" at reference number "530" in Figure 5 are determined to be "characters," "black pixel block 5" is determined to be "inverted characters," and "black pixel block 6" is determined to be a "photograph."

(3) If the distance between the "black pixel blocks" that make up a "character" is within a predetermined distance (for example, "3 pixels"), the "black pixel blocks" are classified into the same group. Then, the "circumscribed rectangular area" that includes all of the "black pixel blocks" classified into the same group is determined to be a "character area". As a result, the "black pixel block 1" and "black pixel block 2", and the "black pixel block 3" and "black pixel block 4" in reference number "530" in FIG. 5 are determined to be close to each other and are determined to be "character areas". By the above processes from (1) to (3), the determination results are output that the areas indicated by reference numbers "521" and "522" are "character areas", the area indicated by reference number "523" is "inverted character area", and the area indicated by reference number "524" is "photographic area". The above is the process performed by the first area determination unit 303.

<Second Area Determination Unit 304>
The second area determination unit 304 executes a "character segmentation process" on the area determined to be a "character area" by the first area determination unit 303. This results in "unit character area information (x, y, w, h)." Reference numeral "540" in FIG. 5 indicates the result of performing the character segmentation process on each of the areas indicated by reference numerals "521,""522," and "523." Here, reference numerals "541" to "546" indicate "unit character area information (x, y, w, h)." The character segmentation process is a process of segmenting the circumscribed rectangle of each character as a "character segmentation rectangle" based on the horizontal projection and the vertical projection in the "character area."

<MMR compression unit 305>
The MMR compression unit 305 inputs the binary image data binarized by the binarization unit 302, and performs MMR compression on the area determined to be a "text area" by the first area determination unit 303. The compressed MMR data is input to a PDF generation unit 310, which will be described later.

<Reduction Unit 307>
The reduction unit 307 reduces the input data (RGB multi-value image data) read by the scanner unit 201 to generate a "reduced multi-value image." The generated "reduced multi-value image" is temporarily stored in the RAM 206. In this embodiment, "reduction" refers to resolution conversion to a lower resolution, and for example, resolution conversion is performed using the bicubic method.

<Character Region Filling Unit 308>
The character region filling unit 308 refers to the "binary image data" binarized by the binarization unit 302 and the "character region information (X, Y, W, H)" obtained from the first region determination unit 303, and calculates the average value of the "background color" in the "character region". The "background color" value and the calculated average value of the "background color" are temporarily stored in the RAM 206. Also, the unit character region information (x, y, w, h) obtained from the second region determination unit 304 is referred to, and the calculated average value of the "background color" is assigned to the "unit character region" of the reduced multi-value image. In other words, the "unit character region" in the "character region" of the reduced multi-value image is filled (assigned) with the calculated "background color", and a filled "filled reduced multi-value image" is generated. This improves the compression rate of the subsequent JPEG compression unit 309.

<JPEG compression unit 309>
A JPEG compression unit 309 JPEG-compresses the "filled reduced multi-valued image" generated by the character area filling unit 308. The JPEG data is input to a PDF generation unit 310, which will be described later.

<Character Color Extraction Unit 306>
The character color extraction unit 306 executes the following process by referring to the "character area information (X, Y, W, H)" and "unit character area information (x, y, w, h)" obtained from the first area determination unit 303 and the second area determination unit 304. That is, by referring to the "character area information" and the "unit character area information", the "representative color" of each character in the "character area" is extracted while matching the position of the "black portion" of the binary image data generated by the binarization unit 302 with the position of the "reduced multi-value image" generated by the reduction unit 307.

Figure 7 is a schematic diagram for explaining the processing of the character color extraction unit 306. Reference characters "701", "702", and "703" are "reduced multi-value image 150 (dpi)" generated by the reduction unit 307, and reference characters "704", "705", and "706" are "binary image 300 (dpi)" generated by the binarization unit 302. Reference characters "707", "708", and "709" are "binary image data (300 dpi)" and character colors for each character, and reference characters "710", "711", and "712" indicate the character colors after character color extraction.

Here, the "unit character area information (x, y, w, h)" of the "binary image data (300 dpi)" indicated by the reference numeral "704" is referenced, and the color value corresponding to the "black portion" of the binary image is obtained from the "reduced multi-value image (150 dpi)" indicated by the reference numeral "701". Here, the explanation will be supplemented with reference to FIG. 8. Reference numeral "801" in FIG. 8 is an enlarged view of the "binary image (300 dpi)" indicated by the reference numeral "704", and reference numeral "802" is an enlarged view of the "reduced multi-value image (150 dpi)" indicated by the reference numeral "701". The images indicated by the reference numerals "801" and "802" are illustrated with the same size of one pixel.

Therefore, code "802", which is "150 dpi", has a width and height that are "1/2" of code "801", which is "300 dpi". The text color extraction unit 306 references color value "807", which corresponds to the positions of codes "803" to "806", which are the "black pixels" of code "801". In this way, the text color extraction unit 306 obtains color values corresponding to all the black pixels of code "801" and calculates the average color value (text color).

Reference numeral "707" indicates the average color of each character obtained in this way. As shown in reference numeral "707", the character "E" has "average color (R, G, B) = (80, 80, 80)" and the character "F" has "average color (R, G, B) = (75, 75, 75)". The reason why the average colors of the characters "E" and "F" are different is as follows. The "reduced multi-value image (150 dpi)" is a reduced version of data read by a scanner. Therefore, the color values of the characters "E" and "F" vary from pixel to pixel, and the halftone state that occurs at the edge of the character due to reduction is different between the characters "E" and "F". In order to align the color values of each character, the character color extraction unit 306 replaces them with a similar color, for example, a single "representative color" if it is within a range of a predetermined luminance difference or color difference.

In this embodiment, a method is adopted in which the average color of a character with a large number of "black pixels" in the binary image data before the average color calculation is selected as the "representative color". In other words, when comparing the number of pixels that make up "E" with the number of pixels that make up "F", the number of pixels that make up "E" is greater. As a result, the "average color (R, G, B) = (80, 80, 80)" of the character "E" is selected as the "representative color". The symbol "710" indicates that the "representative color (R, G, B) = (80, 80, 80)" of both the characters "E" and "F" is selected. Similarly, the "unit character area information (x, y, w, h)" of the "binary image data 300 (dpi)" indicated by the symbol "705" is referenced, and the color value corresponding to the "black portion" of the binary image data is obtained from the "reduced multi-value image 150 (dpi)" indicated by the symbol "702".

Reference number "708" is the average color for each character obtained in this way. As shown in reference number "708", the character "E" has "average color (R, G, B) = (100, 80, 100)" and the character "F" has "average color (R, G, B) = (95, 75, 95)". Reference number "711" indicates that both the characters "E" and "F" have "representative color (R, G, B) = (100, 80, 100)".

Similarly, the "unit character area information (x, y, w, h)" of the "binary image data 300 (dpi)" indicated by the reference symbol "706" is referenced, and the color value corresponding to the "black portion" of the binary image data is obtained from the "reduced multi-value image 150 (dpi)" indicated by the reference symbol "703". Reference symbol "709" is the average color for each character obtained in this manner. As indicated by the reference symbol "709", the character "E" has an "average color (R, G, B) = (140, 140, 140)" and the character "F" has an "average color (R, G, B) = (135, 135, 135)". Furthermore, reference symbol "712" indicates that both the characters "E" and "F" have "representative color (R, G, B) = (140, 140, 140)". In this manner, the character color extraction unit 306 extracts the character color for each "character area".

<Color conversion unit 311>
The color conversion unit 311 refers to the "background color" and its average value stored in the RAM 206 by the character area filling unit 308, and replaces the color corresponding to the "background color" in the character area in the image compressed by the JPEG compression unit 309 with the average value. Specific processing by the color conversion unit 311 will be described later.

<PDF Generation Unit 310>
A PDF generating unit 310 combines the MMR data compressed by the MMR compressing unit 305, the character color obtained by the character color extracting unit 306, and the JPEG data compressed by the JPEG compressing unit 309 and color-converted by the color converting unit 311. Then, the PDF generating unit 310 converts the data into a PDF format to generate a PDF (highly compressed).

As described above, the first area determination unit 303 (detection unit) detects a character area from the input multi-value image data. The second area determination unit 304 (cut-out unit) cuts out characters from the detected character area and obtains a unit character area, which is an area in which a single character exists in the character area. Next, the character area filling unit 308 (assignment unit) assigns a conversion color determined based on the color distribution of the background color in the character area other than the cut-out characters to a "unit character area" in the character area in the "reduced multi-value image" obtained by "reducing" the multi-value image data, and outputs the converted color. Then, the color conversion unit 311 replaces the color corresponding to the background color in the character area in the image compressed by JPEG compression or the like from the output image by the character area filling unit 308 with the converted color. This makes it possible to suppress color changes in filling in the character background of PDF (high compression), etc. Alternatively, the character region filling unit 308 can create a histogram showing the frequency distribution of each color signal of the background color based on the color distribution, and set the most frequently occurring color in the created histogram as the conversion color.

<Description of Each Process of Data Processing Unit 215>
Next, the processes executed by the data processing unit 215 will be described with reference to the flowcharts of Figures 4(A), (B), and (C). The programs corresponding to the flowcharts are stored in the ROM 210 or storage unit 211 shown in Figure 2 and are executed by the CPU 205. The CPU 205 can transmit and receive data to and from the data processing unit 215 using the Image Bus I/F 212, the system bus 216, and the image bus 217. The process contents will be described with reference to the respective flowcharts of "MMR data generation", "JPEG data generation", and "character color data generation".

<Generation of MMR data>
First, "MMR data generation" will be described with reference to steps S401 to S405 in Fig. 4A. First, in step S401, the CPU 205 generates "gray multi-value image data" from the input data (RGB multi-value image data) read by the scanner unit 201. The gray conversion is as described in the explanation of the operation of the gray conversion unit 301. Next, in step S402, the CPU 205 generates "binary image data" from the "gray multi-value image data" obtained in step S401. The binary image data is as described in the explanation of the operation of the binarization unit 302.

Next, in step S403 (first area determination), the CPU 205 detects a "text area" and a "photographic area" from the "binary image data" generated in step S402. The "text area information (X, Y, W, H)" and "inverted text area information" obtained here are temporarily stored in the RAM 206. The first area determination in step S403 is as described in the explanation of the operation of the first area determination unit 303.

Next, in step S404 (second area determination), the CPU 205 performs character extraction processing on the area determined to be a "character area" in step S403. The "unit character area information (x, y, w, h)" obtained here is temporarily stored in the RAM 206. The second area determination in step S404 is as described in the explanation of the operation of the second area determination unit 304.

Next, in step S405, the CPU 205 inputs the "binary image data" binarized in step S402, and performs MMR compression on the "area (character area on the binary image data)" determined to be a "character area" in step S403. The MMR compression is as described in the explanation of the operation of the MMR compression unit 305. The MMR data is generated in this manner, and temporarily stored in the RAM 206.

<Generation of JPEG data>
Next, "JPEG data generation" will be described with reference to steps S420 to S426 in Fig. 4C. First, in step S420, the CPU 205 performs "reduction" on the input data (RGB multi-value image data) to generate a "reduced multi-value image." The generated "reduced multi-value image" is temporarily stored in the RAM 206. "Reduction" is as described in the explanation of the operation of the reduction unit 307.

Next, in step S421, the CPU 205 extracts the background color in the "text area" by referring to the binary image data generated in step S402 and the "text area information (X, Y, W, H)" obtained in step S403. Specifically, the color values of the area other than the "black part" of the binary image data corresponding to the text in the processing of the text color extraction unit 306 described above are extracted from the "reduced multi-value image 150 (dpi)" shown by reference number "802" in FIG. 8. For example, in the case of the one shown by reference number "802" in FIG. 8, three color values of ", (R, G, B) = (255, 200, 255), (255, 210, 255), (255, 190, 255)" are extracted (see reference number "702" in FIG. 7).

Next, in step S422, the CPU 205 calculates the average value of the background color in the "text area" extracted in step S421. The average value is calculated by averaging the R, G, and B values of the extracted background color, as shown in the following "Formula 2."

Here, Ar, Ag, and Ab are the average values of R, G, and B, respectively, n is the number of extracted background colors, "i" is an integer from "1" to "n", "Ri", "Gi", and "Bi" are the R, G, and B values of the "i"th extracted background color, respectively, and Σ represents the calculation of the sum.

Here, the RGB values of the background color extracted in step S421 and the background color average value calculated in step S422 are associated and temporarily stored in RAM 206. For example, as shown in FIG. 6, "No. (601)", "background color (602)" and "background color average (603)" are associated and stored as table data. This "No. (601)" is assigned in order to each character area. Note that any form of storage may be used as long as the extracted background color and the calculated average value can be associated and referenced.

Next, in step S423, the obtained "unit character area information (x, y, w, h)" is referenced, and the calculated average value of the background color is assigned to the "unit character area" of the reduced multi-value image. In other words, the "unit character area" of the "reduced multi-value image" is filled with the calculated background color, and a filled "filled reduced multi-value image" is generated. Character area filling is as described in the explanation of the operation of character area filling unit 308.

Next, in step S424, the CPU 205 determines whether or not processing of the "all character areas" has been completed. If it is determined that processing of the "all character areas" has not been completed (N), the process returns to step S421. On the other hand, if it is determined in step S424 that processing of the "all character areas" has been completed (Y), the process proceeds to step S425. Then, in step S425, the CPU 205 JPEG-compresses the "filled reduced multi-value image" generated in step S423. The JPEG compression is as described in the explanation of the operation of the JPEG compression unit 309.

Then, in step S426, the CPU 205 performs color conversion on the image JPEG compressed in step S425. In step S423, the CPU 426 refers to the table data stored in the RAM 206 and replaces the RGB values of the background color with the average RGB values of the background color. Note that in this embodiment, color conversion is performed on the entire image, but color conversion may be performed only on a specific area within the image. JPEG data is generated in this manner and temporarily stored in the RAM 206. Note that in this embodiment, the RGB values of the background color after replacement are set to the average value of the background color within the "text area," but this is not limited to this. For example, replacement may be performed with the RGB value of the peak of the color distribution, or with RGB values determined according to the characteristics of the color distribution.

<Generating character color data>
Next, the "character color data generation" will be described with reference to steps S410 to S414 in Fig. 4B. First, in step S410, the CPU 205 searches for the first "character area" by referring to the binary image data and the "character area information (X, Y, W, H)" obtained in step S403. Next, in step S411, the CPU 205 determines whether the "attention area" that is the area of interest is a "character area". If it is determined that the "attention area" is a "character area" (Y), the process proceeds to step S412, whereas if it is determined that the "attention area" is not a "character area" (N), the process proceeds to step S414.

Next, in step S412, the CPU 205 performs the following process by referring to the "character area information (X, Y, W, H)" obtained in step S403 and the "unit character area information (x, y, w, h)" obtained in step S404. That is, the CPU 205 extracts the character color for each "character area" while matching the positions of the "black parts" of the binary image data with the "reduced multi-value image" generated in step S420. The method of extracting the representative color is as described above in the explanation of the operation of the character color extraction unit 306.

Next, in step S413, CPU 205 determines whether or not processing for "all character areas" has been completed. If CPU 205 determines that processing for "all character areas" has been completed (Y), this process ends. On the other hand, if CPU 205 determines that processing for "all character areas" has not been completed (N), the process proceeds to step S414 and searches for the next "character area." As described above, character color data is generated for each character area and temporarily stored in RAM 206.

<Generating PDF (highly compressed)>
As explained in the flowchart of Fig. 4, "MMR data", "JPEG data", and "character color data" are generated. The CPU 205 generates a PDF (high compression) by converting these data into PDF format. By performing the above, color unevenness due to block noise, etc. in the character background color, etc. can be replaced with a single color, and color changes at the time of PDF (high compression) can be suppressed.

Second Embodiment
In the first embodiment, the RGB values of the background color in the "text area" were the subject of color conversion. However, factors that cause color changes in PDF (high compression) include "background correction" in addition to block noise due to compression and resolution conversion. "Background correction" corrects the background area of the image, such as the paper part. Generally, when scanning a printed material, even if white paper is read, the color after scanning will not be completely white, but will be slightly colored and noise will occur.

Therefore, by correcting the background, the background of the paper part, etc. is corrected to "white" and noise is removed at the same time, improving the visibility of the image after scanning and improving the compression effect. Normally, "background correction" replaces all values higher than a bright signal value (for example, RGB values of "240" etc.) with white (RGB value "255"). However, since this replacement is performed on the entire image, bright signal values other than the background are replaced with "255", causing color changes. Therefore, by extracting the "background area" and the value of the background color, and replacing the background color only for the "background area" with white, PDF (high compression) processing is performed with reduced color changes other than the background area even during "background correction". Note that a description of the same processing as in the first embodiment will be omitted.

Figure 9 is a flow chart showing the "JPEG data generation" process in the second embodiment. Steps S900 to S904 in Figure 9 are the same as those in Figure 4 (C) in the first embodiment, so repeated explanations will be omitted. First, in step S905, the CPU 205 detects a "background region" from the image after the "text region filling" process in step S903. For example, the "background region" is determined by extracting the color distribution of the image after the "text region filling" process, and determining the signal value range that is a commonly distributed color and is bright as the "background signal value" and the group of pixels that have this "background signal value" as the "background region." Of course, any method may be used as long as the background region can be detected and the background signal value can be extracted.

Next, in step S906, the CPU 205 determines the conversion signal value of the "background signal" of the detected "background area." Basically, to replace it with "white," if it is RGB, the R, G, and B values are replaced with "255." Note that if you want to replace the background with something other than white, such as if the paper of the printed matter to be scanned is colored paper, the conversion color may be determined to match the color of the paper.

Step S907 (JPEG compression) and step S908 (color conversion) are the same processes as in the first embodiment (FIG. 4(C)), so a duplicated explanation will be omitted. By performing the above-described processes, the "base color" can be replaced with white while the color of the background in the text area and the like is not changed.

As described above, the CPU 205 detects the background area of the multi-value image data (background area detection section), and determines the conversion color of the background from the detected background area (background color determination section). The CPU 205 then replaces the background color in the background area of the image compressed using JPEG compression or the like with the conversion color determined by the background color determination section (background color conversion section). This makes it possible to improve the visibility of the scanned image and the compression effect by correcting the background of the paper part, etc. to "white" and simultaneously removing noise.

<Modification>
(1) The device connected to the MFP 101 via a network may be a mobile terminal or the like in addition to a PC. Examples of the mobile terminal include mobile terminals such as smartphones and tablet terminals. In addition to wired communication such as a communication line, the network may be partially or entirely wireless communication. (2) Furthermore, a plurality of recommended RGB values to be used for color conversion may be stored in advance in a non-volatile manner, and any one of these may be adopted.

<Additional Note>
The disclosure of this embodiment includes the following configuration.
(Configuration 1) An image processing device capable of extracting a rectangular area including characters as a character area, comprising:
a detection unit for detecting a character region from input multi-value image data;
a segmentation unit that segments a character within the character region detected by the detection unit, and obtains a unit character region, which is a region in which a single character exists within the character region;
an allocation unit that allocates a conversion color determined based on a color distribution of a background color in a character area other than the character extracted by the extraction unit to a unit character area in the character area in a reduced multi-value image obtained by reducing the multi-value image data, and outputs the conversion color;
a color conversion unit that replaces a color corresponding to a background color in a character area in an image obtained by compressing an output image from the allocation unit with the conversion color.
(Configuration 2) The image processing device according to configuration 1, wherein the allocating section determines an average value of color signals of a background color within the character region as the conversion color.
(Configuration 3) The image processing device according to configuration 1, wherein the allocating unit creates a histogram indicating a frequency distribution for each color signal of the background color based on the color distribution, and sets a color showing a highest frequency in the created histogram as a conversion color.
(Configuration 4) The apparatus further includes an association unit that associates a background color in a character region with a conversion color and stores the associated color,
4. The image processing device according to claim 1, wherein the color conversion unit replaces a background color in a character area with a conversion color by referring to the contents stored in the association unit.
(Configuration 5) The image processing device according to any one of configurations 1 to 3, further comprising a character color extraction unit that extracts a character color for each of the detected character areas.
(Configuration 6) The character color extraction unit further comprises:
The image processing device according to configuration 5, characterized in that a position of a black portion of the binary image data of the multi-value image data is compared with a color of a position corresponding to the black portion in the reduced multi-value image, and a character color of each character in a character area is extracted.
(Configuration 7) The character color extraction unit further comprises:
7. The image processing apparatus according to configuration 6, wherein an average color of a character having the largest number of black pixels in the binary image data is output as a representative color of the character color.
(Configuration 8) An MMR compression unit that performs MMR compression on a character region detected by the detection unit in the image data obtained by binarizing the multi-value image data;
The image processing device according to any one of configurations 5 to 7, further comprising: a PDF generation unit that synthesizes JPEG data obtained by color conversion of the MMR data compressed by the MR compression unit, the character color extracted by the character color extraction unit, and a background color in a character area in an image obtained by JPEG compressing the output image by the color conversion unit.
(Configuration 9) A background area detection unit for detecting a background area of the multi-value image data;
a background color determination unit that determines a converted background color from the background area detected by the background area detection unit;
8. The image processing apparatus according to claim 1, further comprising a background color conversion unit that replaces a background color in a background area in the compressed image with the conversion color determined by the background color determination unit.

Although the preferred embodiment of the present invention has been described above, the present invention is not limited to the above-mentioned embodiment, and various modifications and variations are possible within the scope of the gist of the present invention. For example, the present invention can also be realized by supplying a program that realizes one or more of the functions of the above-mentioned embodiment to a system or device via a network or recording medium, and having the computer processor of the system or device read and execute the program. The present invention can also be realized by a circuit (e.g., an ASIC) that realizes one or more functions.

101 MFP
102 Client PC
201 Scanner unit 202 Printer unit 205 CPU
206 RAM
207 Operation unit 210 ROM
211 Storage unit 215 Data processing unit 213 RIP unit 301 Gray conversion unit 302 Binarization unit 303 First area determination unit 304 Second area determination unit 305 MMR compression unit 306 Character color extraction unit 307 Reduction unit 308 Character area filling unit 309 JPEG compression unit 310 PDF generation unit 311 Color conversion unit

Claims (11)

An image processing device capable of extracting a rectangular area including characters as a character area,
a detection unit for detecting a character region from input multi-value image data;
a segmentation unit that segments a character within the character region detected by the detection unit, and obtains a unit character region, which is a region in which a single character exists within the character region;
an allocation unit that allocates a conversion color determined based on a color distribution of a background color in a character area other than the character extracted by the extraction unit to a unit character area in the character area in a reduced multi-value image obtained by reducing the multi-value image data, and outputs the conversion color;
a color conversion unit that replaces a color corresponding to a background color in a character area in an image obtained by compressing an output image from the allocation unit with the conversion color. The allocation unit:
2. The image processing apparatus according to claim 1, wherein an average value of the color signals of the background color within the character area is determined as the converted color. The allocation unit:
2. The image processing apparatus according to claim 1, further comprising: a histogram showing a frequency distribution for each color signal of the background color based on the color distribution; and a color showing the highest frequency in the histogram is determined as the conversion color. The method further includes an association unit that associates and stores a background color in the character area with a conversion color,
4. The image processing apparatus according to claim 1, wherein the color conversion unit replaces a background color in a character area with a conversion color by referring to the contents stored in the association unit. The image processing device according to any one of claims 1 to 3, further comprising a character color extraction unit that extracts a character color for each of the detected character areas. The character color extraction unit further
6. The image processing device according to claim 5, further comprising: a processor for processing the image data; a processor for processing the image data; and a processor for processing the image data. The character color extraction unit further
7. The image processing apparatus according to claim 6, wherein an average color of a character having the largest number of black pixels in the binary image data is output as a representative color of the character. an MMR compression unit that performs MMR compression on the character region detected by the detection unit in the image data obtained by binarizing the multi-value image data;
6. The image processing device according to claim 5, further comprising a PDF generation unit that synthesizes the MMR data compressed by the MR compression unit, the character color extracted by the character color extraction unit, and JPEG data color-converted by the color conversion unit into a background color in a character area in an image obtained by JPEG compressing the output image. a background region detection unit for detecting a background region of the multi-value image data;
a background color determination unit that determines a converted background color from the background area detected by the background area detection unit;
4. The image processing apparatus according to claim 1, further comprising a background color conversion unit that replaces a background color in a background area in the compressed image with the conversion color determined by the background color determination unit. A method for controlling an image processing device capable of extracting a rectangular area including characters as a character area, comprising the steps of:
a detection step of detecting a character region from inputted multi-value image data;
a segmentation step of segmenting a character from within the character region detected by the detection step, and obtaining a unit character region, which is a region in which a single character exists within the character region;
an allocating step of allocating converted colors determined based on a color distribution of a background color in a character region other than the characters extracted in the extracting step to unit character regions in the character region in a reduced multi-valued image obtained by reducing the multi-valued image data, and outputting the converted colors;
a color conversion step of replacing a color corresponding to a background color in a character area in an image obtained by compressing the output image of the allocation step with the conversion color. A program for causing a computer to execute a control method for an image processing device capable of extracting a rectangular area including characters as a character area, the program comprising:
The control method includes:
a detection step of detecting a character region from inputted multi-value image data;
a segmentation step of segmenting a character from within the character region detected by the detection step, and obtaining a unit character region, which is a region in which a single character exists within the character region;
an allocating step of allocating a converted color determined based on a color distribution of a background color in a character region other than the characters extracted in the extracting step to a unit character region in the character region in a reduced multi-valued image obtained by reducing the multi-valued image data, and outputting the converted color;
a color conversion step of replacing a color corresponding to a background color in a character area in an image obtained by compressing an output image from the allocation step with the conversion color.