JP4438834B2 - Image color determination apparatus, image color determination method, and program - Google Patents

Description

  The present invention relates to an image color determination apparatus, an image color determination method, and a program for determining a ground color of a target image, and more particularly to an improvement for performing good image color determination even when a document has a ground color.

  Conventionally, an image processing apparatus and an image analysis apparatus that execute predetermined processing based on the background color and background color of an original are known (for example, Patent Documents 1 and 2). Patent Document 1 discloses a technique for replacing colors of portions other than the background color of the document based on a result of comparing the colors of the background color portion of the document with the colors of other portions.

  Further, in Patent Document 2, for example, the approximate position of the document edge is detected from the input image by Hough conversion or the like, and for each of the detected RGB values of the document edge, the background color candidate and the background color candidate are detected from the peak value of the histogram. A technique for estimating a background color candidate and obtaining a final document edge from the estimated background color candidate and background color candidate is disclosed.

JP 09-149276 A JP 2004-096435 A

  Here, an image read from an original (hereinafter, also simply referred to as “original image”) is converted from an RGB color space to a Lab (lightness L and chromaticity a, b) color space, and the chromaticity is ground color. In the case where a range that is substantially the same as (chromatic color) and a region that is an achromatic color are determined as an achromatic region, a method for selecting a processing mode and image processing by this processing mode are examined.

  For example, in the selection of the processing mode, when it is determined that each pixel of the document image belongs to the achromatic region and does not belong to the chromatic region, the monochrome mode is selected as the processing mode of the document image. Then, not only the pixels corresponding to the ground color but also the ground color removal processing for replacing the pixels having the same chromaticity as the ground color and having a lightness higher than the lightness of the ground color with white, and ground color removal When the recording process for recording the subsequent document image on the recording paper is executed as an image process in the monochrome mode, the following problems have occurred.

  That is, pixels having lightness higher than the ground color are also replaced with the same color (white) as the ground color. As a result, there has been a problem that the aesthetics of the original image differ from the aesthetics of the recorded image recorded on the recording paper.

  Even when the ground color removal processing is performed on the pixels included in the predetermined chromaticity range and brightness range centered on the ground color, the aesthetics before and after the ground color removal (that is, regardless of the color mode and the monochrome mode) There has been a problem that the aesthetics of the original image and the recorded image are different.

  Further, there is a demand for a document image having pixels having the same chromaticity as the ground color and having a lightness higher than the lightness of the ground color as a color and selecting the color mode as the processing mode.

  Therefore, an object of the present invention is to provide an image color determination apparatus, an image color determination method, and a program that can execute a color determination process favorably on an image having a ground color.

An image color determination device according to the subject of the present invention provides a target image for each color region formed on the two-dimensional color plane in a two-dimensional color plane defined by a hue parameter and a saturation parameter. The number of constituent units of the image data is counted, and the presence / absence of the ground color of the target image is determined using the counting result. If the target image has a ground color, A color area corresponding to the ground color is determined, and based on a three-dimensional distribution of structural units included in the color area corresponding to the ground color, a brightness range of the ground color as a three-dimensional ground color area A ground color determination unit that determines a chromaticity range, and a ground color position obtained based on the ground color region using the determination result of the ground color determination unit is an origin on the two-dimensional color plane. To translate the image data related to the target image Accordingly, characterized in that it comprises a target image determining unit that performs color determination before Symbol target image.

According to the first to fifth aspects of the invention, color determination can be performed in consideration of not only the chromaticity of the ground color but also the brightness of the ground color. That is, it is possible to perform color determination after distinguishing a color having the same chromaticity as the ground color and having brightness different from the ground color from the ground color.

In particular, according to the first aspect of the present invention, the configuration of the image color determination device is simplified. Moreover, according to the invention of Claim 2 , it can respond to various user needs. For example, the user determines that the background color of the color is monochrome and, when there is a color brighter than the background color, if the user wants to determine the color as a color, the user determines the color using the ground color determination result. If the background color of the color is determined to be a color, the color determination can be performed without using the background color determination result.

In particular, according to the first aspect of the present invention, the ground color can be easily and accurately determined using a two-dimensional color plane.

In particular, according to the invention described in claim 3 , the influence of false color can be eliminated.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

<1. Configuration of Image Color Determination Device>
FIG. 1 is a diagram illustrating an example of a configuration of an image color determination apparatus 1 according to the present embodiment. Here, the image color determination apparatus 1 is a scanner, a printer, a copier, a facsimile, or a complex machine that combines these functions. Further, the image color determination apparatus 1 can detect the background color of the document read by the scanner unit 41, for example.

  As shown in FIG. 1, the image color determination apparatus 1 mainly includes a modem 22, a scanner unit 41, a recording unit 51, a ground color determination unit 70, and a target image determination unit 80. Here, “ground color” refers to a background color existing in the document.

  The modem 22 converts digital data into an audio signal for transmission, or converts an audio signal transmitted from the outside of the image color determination device 1 and received by the image color determination device 1 into digital data. The NCU 21 is a device required when the image color determination apparatus 1 is connected to the public switched telephone network, and performs outgoing / incoming calls and dial control. Further, the communication unit 25 is a LAN interface that performs data communication with an information processing apparatus (not shown) connected via a network.

  The CODEC 31 is used for lossless compression processing of images transmitted by facsimile communication. For example, the CODEC 31 encodes an image read from a document by the scanner unit 41 and binarized by the image processing circuit 61. The encoded image is stored in the image memory 16. The CODEC 31 decodes facsimile data (binary data) transmitted from another image color determination device. The decoded binary data is supplied to the recording unit 51, and a recording process is executed.

  As encoding by CODEC 31, when binary data is encoded, MH (Modified Huffman), MR (Modified Read), MMR (Modified MR), and JBIG (Joint Bi-level Image experts Group). Either method is adopted. The CODEC 31 also encodes multi-value data. Multi-level data is encoded by, for example, the JPEG method.

  The scanner unit 41 is a reading unit that reads an image from a document. The image data (read image data) read by the scanner unit 41 is compressed by, for example, the JPEG method by the CODEC 31 and stored in the image memory 16.

  That is, the scanner unit 41 reads an image drawn on a document by a CCD line sensor 41a, which will be described later, and generates image data related to the image. The scanner unit 41 reads an image drawn on a document by an ADF (Automatic Document Feeder) method or an FBS (Flat Bed Scanner) method. The ADF method is a method of reading a document by feeding it one by one from a document bundle obtained by bundling a plurality of documents, and the FBS method is a method of reading a document placed on a contact glass. The ADF method includes a method of reading a moving document with a stationary reading optical system (sheet-through method) and a method of reading a stationary document with a reading optical system that moves. When the latter method is adopted, the scanner unit Reference numeral 41 repeatedly executes a reading procedure in which an original is stopped on a contact glass, the stationary original is read by a moving reading optical system, and the original that has been read is discharged. The read image data may be subjected to an averaging process.

  The CCD line sensor 41a repeats reading at a constant cycle while the reading optical system that guides light from the document to the CCD line sensor 41a scans the document, whereby image data expressed in RGB color space, That is, image data having color component data of R (red), G (green), and B (blue) is generated. The scanning speed of the reading optical system is set based on the resolution in the sub-scanning direction of the image data generated by the CCD line sensor 41a. Specifically, the higher the resolution in the sub-scanning direction, the slower the scanning speed of the reading optical system, and the lower the resolution in the sub-scanning direction, the higher the scanning speed of the reading optical system. The “scanning speed” is a relative moving speed between the original and the reading optical system.

  The recording unit 51 is an image forming unit that records a toner image based on an electrostatic latent image on a recording sheet by an electrophotographic method. For example, the recording unit 51 forms a toner image based on the read image on a photosensitive drum (not shown), and transfers the toner image to a recording sheet. As described above, the recording unit 51 is used as a processing unit that performs a recording process on an image to be recorded.

  The image processing circuit 61 is a processing unit that performs predetermined image processing on image data. The image processing circuit 61 performs, for example, a process of performing gamma correction on the image data, a process of converting the resolution of the image data, and a color space of the image data from the first color space to the second color space (for example, RGB color). Processing for converting from space to Lab (color space of lightness L and chromaticity a, b) is executed. Note that these image processes may be realized in software by the MPU 11 based on the program 13 a stored in the ROM 13.

  The display unit 63 is configured by a so-called liquid crystal display, and has a function as a “touch panel” that can specify a position on the screen by touching the screen with a finger or a dedicated pen. Therefore, the user of the image color determination apparatus 1 (hereinafter simply referred to as “user”) performs an instruction using the “touch panel” function of the display unit 63 based on the content displayed on the display unit 63. The image color determination apparatus 1 can be caused to execute a predetermined operation. Thus, the display unit 63 is also used as an input unit.

  The operation unit 64 is an input unit configured by a so-called keypad. The user can cause the image color determination apparatus 1 to execute a predetermined operation by performing an input operation based on the display content of the display unit 63.

  The ground color determination unit 70 is based on an image obtained by reading a document by the scanner unit 41 or an image stored in the image memory 16 (hereinafter also referred to as “image data related to the target image”). Determine. That is, the ground color determination unit 70 detects (determines) the presence or absence of the ground color of the target image. When the target image is an image having a ground color (ground color original), the ground color (on the two-dimensional color plane) Is detected (determined).

  The target image determination unit 80 executes the color determination of the target image based on the image data related to the target image using the determination result of the ground color determination unit 70. In addition, the target image determination unit 80 selects a processing mode to be executed by a processing unit such as the recording unit 51 or the image processing circuit 61 based on the result of the color determination. For example, the target image determination unit 80 can select a color mode that enables execution of color processing and a monochrome mode that enables execution of monochrome processing as processing modes. That is, the target image determination unit 80 selects the color mode as the processing mode when there is a chromaticity similar to the ground color set by the ground color determination unit 70 and there is a lighter color than the ground color. To do.

  The detailed configurations of the ground color determination unit 70 and the target image determination unit 80 will be described later. The ground color determination unit 70 counts the number of structural units (unit image data) of image data for each color region on a two-dimensional plane, and determines the presence or absence of a ground color using the count result. That is, the ground color determination unit 70 counts the constituent units for each color region formed on the two-dimensional color plane, and determines whether there is a ground color depending on whether the distribution of the constituent units is biased to a specific color region. Determine. If it exists in a specific color area, it is determined that there is a ground color corresponding to the color area. When there is no bias toward a specific color area, that is, when the same color is scattered over a plurality of color areas, it is determined that there is no ground color. In addition, even when there is a bias in the achromatic region, it is determined that the ground color does not exist. The target image determination unit 80 counts the number of structural units (unit image data) of image data for each color region on the two-dimensional color plane, and performs color determination using the count result. That is, the target image determination unit 80 counts the constituent units for each color region formed on the two-dimensional color plane, and performs color determination of the target image using the count result. Whether the target image is a color image based on the distribution of structural units in the achromatic region and the chromatic region, assuming that the color region is an achromatic region (region near the origin) and a chromatic region (region excluding the achromatic region) Whether the image is a monochrome image can be determined. For example, if there are many structural units in the achromatic area and few structural units in the chromatic area, the image is determined to be a monochrome image, and if there are many structural units in the chromatic area, color It can be determined that it is an image.

  As described above, the ground color determination unit 70 and the target image determination unit 80 execute different processes, but pay attention to the fact that they are common in that they perform the counting process for each color region. Then, by using a common circuit, the counting by the ground color determination unit 70 and the counting by the target image determination unit 80 are executed. Thereby, the structure of the image color determination apparatus 1 can be simplified.

  A RAM (Random Access Memory) 12 and an image memory 16 are readable and writable volatile memories (storage units). A ROM (Read Only Memory) 13 is a read only memory. An MPU (Micro Processing Unit) 11 executes control according to a program 13 a stored in the ROM 13. In addition, each of the MPU 11, the ROM 13, the recording unit 51, and the like is electrically connected via the signal line 15. Therefore, the MPU 11 can execute, for example, a recording process by the recording unit 51 at a predetermined timing.

  The CODEC 31 functioning as a compression / decompression unit acquires image data, generates compressed image data, and stores the compressed image data in the image memory 16.

  Note that the processing modes of the image color determination apparatus 1 include the “monochrome mode” and the “color mode” as described above.

  When scanning in the “color mode”, the color space conversion unit 61b converts the multi-gradation RGB image data input from the resolution conversion unit 61a into multi-gradation YCbCr image data, and the CODEC 31 The gradation YCbCr image data is compressed by the JPEG method or the like and stored in the image memory 16.

  When copying in the “color mode”, the color space conversion unit 61b converts the multi-gradation RGB image data input from the resolution conversion unit 61a into multi-gradation Lab image data (L * a * b * color). Image data expressed in space), and the multi-tone Lab image data is further converted into multi-tone CMYK image data (image data expressed in the CMYK color space). The multi-tone CMYK image data is binarized into two-tone CMYK image data, and the recording unit 51 acquires the two-tone CMYK image data and forms “Y”, “M”, “C”, and “K” image formations. A color image is formed on a recording medium using a necessary engine (four-color image forming engine). Here, “use necessary one of“ Y ”,“ M ”,“ C ”, and“ K ”image forming engines (four color image forming engines) is necessary to express colors included in an image. This means that it is only necessary to operate a simple image forming engine. For example, if the color classification of the color included in the image is only “Y”, only the image forming engine of “Y” may be operated. If the color classification of the color included in the image is “R”, Only the “Y” and “M” image forming engines need be operated.

  When scanning in the “monochrome mode” when the original is a gray original (monochrome photo original or the like), the color space conversion unit 61b converts the multi-gradation RGB image data input from the resolution conversion unit 61a into multiple gradations. Then, the CODEC 31 compresses the luminance component Y of the multi-gradation YCbCr image data by the JPEG method and stores the compressed YCbCr image data in the image memory 16.

  Alternatively, the color space conversion unit 61b converts the multi-gradation RGB image data input from the resolution conversion unit 61a into multi-gradation YCbCr image data. The luminance component Y of the multi-gradation YCbCr image data is binarized into the two-gradation luminance component Y by a systematic dither method or the like, and the CODEC 31 compresses the two-gradation luminance component Y by the JBIG method or the like to generate an image memory. 16 accumulate.

  On the other hand, when the document is a monochrome document (such as a monochrome character document), the color space conversion unit 61b converts the multi-gradation RGB image data input from the resolution conversion unit 61a into multi-gradation YCbCr image data. To do. The luminance component Y of the multi-gradation YCbCr image data is binarized into a two-gradation luminance component Y by a simple binarization method or the like, and the CODEC 31 compresses the two-gradation luminance component Y by a JBIG method or the like. Accumulate in memory 16.

  When copying in the “monochrome mode” when the original is a gray original, the color space conversion unit 61b converts the multi-gradation RGB image data input from the resolution conversion unit 61a into multi-gradation YCbCr image data. To do. The luminance component Y of the multi-gradation YCbCr image data is binarized into a two-gradation luminance component Y by a systematic dither method, an error diffusion method, or the like, and the recording unit 51 acquires the two-gradation luminance component Y, A monochrome image is formed on the recording medium using the “K” image forming engine.

  On the other hand, when the document is a monochrome document, the color space conversion unit 61b converts the multi-gradation RGB image data input from the resolution conversion unit 61a into multi-gradation YCbCr image data. The luminance component Y of the multi-gradation YCbCr image data is binarized into the two-gradation luminance component Y by an error diffusion method or a simple binarization method, and the recording unit 51 acquires the two-gradation luminance component Y. , “K” is used to form a monochrome image on the recording medium.

  It should be noted that binarization when copying a gray document is preferably performed with more emphasis on reproducibility of gradation than when copying a black and white document, and binarization when copying a black and white document is performed. Therefore, it is desirable to emphasize character reproducibility rather than copying gray originals.

<2. Configuration of ground color determination unit and target image determination unit>
Here, the configurations of the ground color determination unit 70 and the target image determination unit 80 will be described. As shown in FIG. 1, the ground color determination function by the ground color determination unit 70 includes a first unit image data determination unit 65a, a first unit image data counting unit 66a, a first block determination unit 67a, and a first block. This is realized by the counting unit 68a and the ground color setting unit 71.

  As shown in FIG. 1, the color determination function of the target image determination unit 80 includes a first unit image data determination unit 65b, a second unit image data determination unit 82, a first unit image data counting unit 66b, Second unit image data counting unit 83, first block determining unit 67b, second block determining unit 84, first block counting unit 68b, second block counting unit 85, target image converting unit 72, mode And the selection unit 81.

  Further, the first unit image data determining unit 65b, the first unit image data counting unit 66b, the first block determining unit 67b, and the first block counting unit 68b of the target image determining unit 80 are respectively associated with the corresponding ground color determining units 70. Functions similar to those of the first unit image data determination unit 65a, the first unit image data counting unit 66a, the first block determination unit 67a, and the first block counting unit 68a are realized.

  Here, the first unit image data determination unit 65 (65a, 65b), the first unit image data counting unit 66 (66a, 66b), the first block determination unit 67 (67a, 67b), the first of the present embodiment. Each of the block counting unit 68 (68a, 68b), the second unit image data determining unit 82, the second unit image data counting unit 83, the second block determining unit 84, and the second block counting unit 85 constitutes a target image. The counting process and the determination process are executed based on unit image data (constituent unit) acquired based on the target image instead of the pixel data to be processed.

  The unit image data is a pixel group obtained by dividing the target image into adjacent n (n is a natural number) pixels, and the value of each unit image data is an average value of the values of each pixel included in the corresponding pixel group. Is used.

  For example, each unit image data is constituted by a pixel group divided into vertical (sub-scanning direction) 2 × horizontal (main scanning direction) 2 (that is, n = “4”) pixels. When the color space of the target image is RGB, the R value, G value, and B value of four pixels included in the corresponding pixel group are averaged as the values (R, G, B) of each unit image data. Things are used. Thus, in the present embodiment, the process of generating unit image data is a process of averaging a predetermined number of adjacent pixel data, and is realized by the averaging unit 61c.

  In the present embodiment, the first unit image data determination unit 65, the first unit image data counting unit 66, the first block determination unit 67, the first block counting unit 68, the second unit image data determination unit 82, The processing by the two-unit image data counting unit 83, the second block determining unit 84, and the second block counting unit 85 is executed on image data that has been affine transformed on the color plane H that is a two-dimensional color plane.

  Therefore, in the following, first, the color plane H will be described, and then the first unit image data determination unit 65 and the second unit image data determination unit 82 which are constituent elements of the ground color determination unit 70 and the target image determination unit 80. , First unit image data counting unit 66, second unit image data counting unit 83, first block determining unit 67, second block determining unit 84, first block counting unit 68, second block counting unit 85, ground color setting The unit 71, the target image conversion unit 72, and the mode selection unit 81 will be described.

<2.1. Color plane configuration>
FIG. 2 is a diagram illustrating an example of the color plane H. As illustrated in FIG. FIG. 3 is a diagram illustrating a method for acquiring the color plane H. The color plane H is obtained by performing affine transformation such as rotation transformation and scaling transformation on a chromaticity diagram of a general Lab color system. Therefore, the color plane H has the same properties as the Lab chromaticity diagram with respect to hue and saturation. That is, each color is arranged radially around the origin O. Further, the hue is indicated by the direction from the origin O, and the saturation is indicated by the distance from the origin O.

  As shown in FIG. 2, the color plane H is constituted by an orthogonal coordinate system having a horizontal axis a3 and a vertical axis b3, and each point of the color plane H is represented by a coordinate position (a3, b3). The Therefore, the coordinate position (a3, b3) is used as a parameter relating to hue and saturation (hereinafter also referred to as “color parameter”). Where the unit image data is located on the color plane H is determined by the chromaticity values (a, b) of the unit image data.

  Here, it is known that the origin O of the color plane H and the vicinity thereof are approximately achromatic. In the present embodiment, a saturation boundary line D0 indicating a boundary between a chromatic color and an achromatic color is set near the origin O of the color plane H. Thereby, the inside of the saturation boundary line D0 becomes an achromatic region mA indicating an achromatic color. Further, the outside of the saturation boundary line D0 is a chromatic region indicating a chromatic color.

  In the chromatic region outside the saturation boundary line D0, six hue boundary lines D1 to D6 indicating the hue boundary are set. Each of the hue boundary lines D1 to D6 is a half straight line extending radially from the origin side. The chromatic area is divided into a plurality (six in this embodiment) of color areas mC, mM, mY, mR, mG, and mB by the hue boundary lines D1 to D6 and the saturation boundary line D0. .

  That is, as shown in FIG. 2, the hue boundary line D1 is a boundary between the color region mR and the color region mY. Further, the hue boundary line D2 is a boundary between the color area mY and the color area mG. Further, the hue boundary line D3 is a boundary between the color region mG and the color region mC. Further, the hue boundary line D4 is a boundary between the color region mC and the color region mB. Further, the hue boundary line D5 is a boundary between the color region mB and the color region mM. Furthermore, the hue boundary line D6 is a boundary between the color region mM and the color region mR.

  These color regions mC, mM, mY, mR, mG, and mY are six determination colors C (cyan), M (magenta), Y (yellow), R (red), G (green), and B ( Blue).

  The ground color determination and the color determination of the present embodiment are executed based on the position of the coordinate position (a3, b3) of each unit image data on the color plane H. For example, when the coordinate position (a3, b3) of the unit image data exists in the achromatic region mA, it is determined that the color of the unit image data is an achromatic color. On the other hand, when the coordinate position (a3, b3) of the unit image data is present in any of the color regions mC, mM, mY, mR, mG, mB, the color of the unit image data has a hue and a color corresponding to the color region. It is determined that the chromatic colors C, M, Y, R, G, and B have saturation.

  Here, the color plane H is acquired by the following procedure. First, hue boundary lines D1 to D6 are set for the color plane H0 of the Lab color system (see the left side of FIG. 3). This color plane H0 corresponds to a chromaticity diagram of a general Lab color system, and has an orthogonal two-dimensional coordinate system in which the horizontal axis is a and the vertical axis is b. Further, the hue boundary lines D1 to D6 may be set based on human senses. Furthermore, the hue boundary line pair that is approximately symmetric with respect to the origin O is set to be a straight line, but is not limited thereto. However, when the pair of hue boundary lines sandwiching the origin O is a straight line as in the present embodiment, the calculation cost of the determination process can be reduced.

  Subsequently, the entire hue is rotationally converted around the origin O so that the hue boundary lines D1 and D4 close to the vertical axis b of the color plane H0 coincide with the vertical axis b. Thereby, the color plane H1 is obtained (see the center of FIG. 3). The color plane H1 has an orthogonal two-dimensional coordinate system in which the horizontal axis is a2 and the vertical axis is b2.

  Subsequently, the hue boundaries D2, D3, D5, and D6 are inclined at 45 degrees with respect to the horizontal axis a2 and the vertical axis b2, respectively, at different rates in the vertical axis direction and the horizontal axis direction with respect to the entire hue. Conversion to enlarge or reduce is executed. Thereby, a color plane H2 is obtained (see the right side of FIG. 3). The color plane H2 has an orthogonal two-dimensional coordinate system in which the horizontal axis is a3 and the vertical axis is b3. Then, by setting a saturation boundary line D0 for this color plane H2, the color plane H in FIG. 2 is obtained.

  Note that the color determination of the unit image data uses the value of the unit image data as expressed in Lab, and the coordinate position (a, b) indicated by the chromaticity corresponds to the chromaticity diagram of the Lab color system. This may be done by determining at which position of the color plane H0 to be present. In this case, it is not necessary to obtain a color parameter in addition to chromaticity.

  However, in general, in the color plane H0, the slopes of the hue boundary lines D1 to D6 may become irrational numbers, and the determination of the color of the unit image data requires an irrational number calculation such as an inverse trigonometric function, and the determination efficiency is improved. It may get worse. Therefore, the first unit image data determining unit 65, the second unit image data determining unit 82, the first unit image data counting unit 66, the second unit image data counting unit 83, the first block determining unit 67, and the second block determining unit. 84, the first block counting unit 68 and the second block counting unit 85 perform at least affine transformation such as rotational transformation on the image data on the color plane H0, and the gradients of the hue boundary lines D1 to D6 are rational numbers. The color plane H is used.

<2.2. Components of Ground Color Determination Unit and Target Image Determination Unit>
Here, the components of the ground color determination unit 70 and the target image determination unit 80 will be described. The first unit image data determination unit 65 (65a, 65b) executes a determination process on the image data converted from the RGB color space to the Lab color space by the image processing circuit 61. That is, the first unit image data determination unit 65 determines that the unit image data acquired based on the image data has a color region mC, mM, mY, mR, mG, corresponding to a specific chromatic color in the color plane H. Whether it belongs to mB or achromatic region mA is determined.

  The first unit image data counting unit 66 (66a, 66b) divides the image data into a plurality of blocks composed of unit image data of, for example, 30 × 30 (total 900). In addition, the first unit image data counting unit 66, based on the determination result of the first unit image data determination unit 65, each of the plurality of color regions mC, mM, mY, mR, mG, mB and the achromatic region mA. The number of unit image data belonging to is counted for each block.

  For example, the first unit image data counting unit 66 counts the number of 900 unit image data included in the block that belongs to the color area mC, and stores the count result in the RAM 12. Similarly, the first unit image data counting unit 66 counts the number of unit image data belonging to each of the color regions mM, mY, mR, mG, mB, and the achromatic region mA, and stores each count result in the RAM 12. Store.

  The first block determination unit 67 (67a, 67b) determines the color included in each block based on the counting result of the first unit image data counting unit 66. That is, for each block, the block determination unit 67 compares each count value of each color region mC, mM, mY, mR, mG, mB and the achromatic region mA with the reference value for existence confirmation, and includes the block. Determine the color to be displayed.

  For example, for a certain block, when only the color regions mC and mM have count values exceeding the existence confirmation reference value, it is determined that the block includes cyan C and magenta M. In addition, in a certain block, when only the achromatic region mA has a count value exceeding the existence confirmation reference value, it is determined that only the achromatic color exists in the block.

  Note that the threshold value (presence confirmation threshold value) used for the presence confirmation of the unit image data may be a different value for each color region mC, mM, mY, mR, mG, mB, and the achromatic region mA. The same value may be used.

  The first block counting unit 68 (68a, 68b), based on the determination result of each block executed by the first block determination unit 67, each color region mC, mM, mY, mR, mG, mB and achromatic region mA. A block count value is calculated every time. That is, the first block counting unit 68 adds the count values (block count values) of the color region and the achromatic region corresponding to the color determined to be included in each block by the first block determination unit 67. For example, when it is determined by the first block determination unit 67 that only a cyan C and magenta M are included in a certain block, the first block counting unit 68 performs the color regions mC and mM as the counting process related to the block. Add the count value. When the first block determination unit 67 determines that a certain block includes only an achromatic color, the first block counting unit 68 adds the count value of the achromatic region mA as a counting process related to the block. To do. Then, the first block counting unit 68 stores the result (block count value) counted for each of the plurality of color regions mC, mM, mY, mR, mG, mB and the achromatic region mA in the RAM 12.

  The ground color setting unit 71 processes the image data by the first unit image data determining unit 65a, the first unit image data counting unit 66a, the first block determining unit 67a, and the first block counting unit 68a. Is executed, the background color of the image data is set based on the counting result of the first block counting unit 68a.

  For example, when one or two of the block count values counted for each of the plurality of color regions mC, mM, mY, mR, mG, mB and the achromatic region mA is equal to or greater than the ground color reference value, the ground color setting unit 71 determines that a ground color exists in a color region that is equal to or greater than the ground color reference value, and selects the corresponding color region.

  That is, the ground color setting unit 71 determines the presence / absence of a ground color based on the block count value and ground color reference value of each color region (each color region in the two-dimensional color plane) and determines that there is a ground color. In this case, the color area corresponding to the ground color is determined.

  If there is a color region in which the block count value exceeds the ground color reference value, the ground color setting unit 71 determines that there is a ground color. Then, it is determined that the color area is a color area corresponding to the ground color. It should be noted that a plurality of ground color reference values are provided, not only when one color region exceeds the first ground color reference value, but also when two color regions exceed the second ground color reference value. It is determined that there is a ground color.

  The ground color setting unit 71 then selects a ground color based on the barycentric positions of the plurality of unit image data included in the selected color region for the color region selected based on the counting result of the first block counting unit 68. A position is set, and a three-dimensional ground color region P1 is calculated based on a three-dimensional distribution of a plurality of unit image data included in the selected color region.

  For example, in the setting of the three-dimensional ground color region P1, first, the maximum value and the minimum value of the L value, the a3 value, and the b3 value are obtained for a plurality of unit image data included in the selected color region. That is, the lightness range and chromaticity range of the ground color are obtained based on the three-dimensional distribution of the unit image data in the color region. Then, the barycentric position based on the unit image data included in the calculated brightness range and chromaticity range (that is, the ground color region P1 (see FIG. 6)) is determined as the ground color position. Therefore, detection of an erroneous ground color can be further prevented.

  The ground color position calculation method is not limited to this. For example, the calculation may be performed based on the maximum value and the minimum value of the unit image data included in the selected color region (for example, the average value of the maximum value and the minimum value). Further, the maximum value and the minimum value of the L value, the a3 value, and the b3 value obtained when setting the ground color are stored in the RAM 12.

  The target image conversion unit 72 performs affine transformation on the image data (unit image data) on the color plane H based on the ground color set by the ground color setting unit 71. Specifically, in the target image conversion unit 72, the ground color position (center of gravity position) calculated by the ground color setting unit 71 includes a plurality of color regions mC, mM, mY, mR, mG, mB, and an achromatic region. The image data to be determined is translated so as to be the origin of the color plane H constituted by mA.

  The second unit image data determination unit 82 performs a determination process on the image data that has been subjected to affine transformation by the target image conversion unit 72. That is, the second unit image data determination unit 82 includes unit image data acquired based on the image data after the affine transformation included in the chromaticity range of the ground color set by the ground color setting unit 71 and is bright. Is determined to be equal to or greater than the brightness threshold set by the ground color setting unit 71. Note that the maximum value of the L value obtained by the ground color setting unit 71 may be used as the brightness threshold.

  The second unit image data counting unit 83 executes a counting process based on the determination result of the second unit image data determining unit 82. That is, the second unit image data counting unit 83 is included in the chromaticity range of the ground color set by the ground color setting unit 71, and the brightness is equal to or higher than the brightness threshold set by the ground color setting unit 71. The number of unit image data (hereinafter also referred to as “background color upper image data”) is counted.

  The second block determination unit 84 determines whether or not the background color upper image data is included in each block based on the counting result of the second unit image data counting unit 83. For example, when the count result by the second unit image data counting unit 83 is equal to or greater than a predetermined threshold, the second block determination unit 84 determines that the background color upper image data is included in the block of interest. On the other hand, when the counting result by the second unit image data counting unit 83 is smaller than the predetermined threshold, the second block determination unit 84 determines that the background color upper image data is not included in the target block.

  The second block counting unit 85 counts the number of blocks determined to include the ground color upper image data based on the determination result of each block by the second block determining unit 84. Then, the second block counting unit 85 stores the count result (block count value) in the RAM 12.

  The mode selection unit 81 performs color determination based on image data on which the affine transformation with the ground color position as the origin of the color plane H is performed on the color plane H, and selects a processing mode based on the color determination result Execute. That is, the first unit image data determination unit 65b, the second unit image data determination unit 82, the first unit image data counting unit 66b, and the image data subjected to affine transformation by the target image conversion unit 72, When processing by the second unit image data counting unit 83, the first block determining unit 67b, the second block determining unit 84, the first block counting unit 68b, and the second block counting unit 85 is executed, The mode selection unit 81 selects a processing mode based on the count results of the first block counting unit 68b and the second block counting unit 85.

  For example, when the block count value of the color region mC is equal to or greater than a predetermined threshold, the mode selection unit 81 determines that the target image is a color image, and selects “color mode” as the processing mode. Similarly, when any of the block count values of the color regions mM, mY, mR, mG, and mB is equal to or greater than a predetermined threshold value, the mode selection unit 81 determines that the target image is a color image, and “ Select “Color Mode”.

  Further, when the count result (block count value) by the second block counting unit 85 is equal to or higher than the brightness reference value, the mode selection unit 81 selects “color mode” as the processing mode. For this reason, the processing unit such as the recording unit 51 or the image processing circuit 61 performs good processing on the portion of the image data that is included in the background color range and is equal to or greater than the brightness threshold without losing aesthetics. Can be executed.

  On the other hand, if the block count values of the color regions mC, mM, mY, mR, mG, and mB are all smaller than the predetermined threshold, the mode selection unit 81 determines that the target image is a monochrome image, and “monochrome mode”. Select. When the count result by the second block counting unit 85 is smaller than the brightness reference value, the mode selection unit 81 selects the monochrome mode.

  Then, the processing units such as the recording unit 51 and the image processing circuit 61 perform predetermined processing on the image data related to the target image based on the processing mode (color mode or monochrome mode) selected by the mode selection unit 81. Apply.

  For example, when the monochrome mode is selected by the mode selection unit 81, the recording unit 51 records the image data converted to monochrome on a recording sheet. When the color mode is selected by the mode selection unit 81, the CODEC 31 performs JPEG compression using the image data as color data.

  The threshold value used for color determination may be a different value for each color region mC, mM, mY, mR, mG, mB, and achromatic region mA, or the same value. .

<3. Ground color determination and color determination procedure>
FIG. 4 is a flowchart for explaining the ground color determination procedure. FIG. 5 is a flowchart for explaining the color determination procedure. Hereinafter, the ground color determination procedure will be described, and then the color determination procedure will be described.

<3.1. Ground color judgment procedure>
Here, the ground color determination procedure of the image data will be described. As shown in FIG. 4, in step S101, prior to the background color determination, an image of predetermined lines (for example, several tens to several hundred lines) from the top of the document (hereinafter also referred to as “original document image”) in the image data. Pre-determination processing is executed.

  For example, as the pre-determination process in step S101, the image processing circuit 61 executes a process (averaging process) for generating unit image data from image data in the RGB color space. The image processing circuit 61 performs gamma correction processing on the generated unit image data. Further, the image processing circuit 61 executes processing for converting the color space of the unit image data subjected to the gamma correction processing from RGB to Lab (lightness L and chromaticity a, b).

  Note that pixel data that has not been subjected to averaging processing may be used as unit image data. However, if an average value for each pixel set is used as unit image data instead of pixel data, a false color caused by a minute mechanical error such as a line sensor in the scanner unit 41 or a false color caused by a scaling factor of a copy function Can be corrected, and the accuracy of the subsequent determination can be improved.

  Therefore, if the unit data is composed of the average values Ravg, Gavg, and Bavg of the color component data for each pixel set, the R, G, and B reading lines are provided with a line interval of several μm in the CCD line sensor 41a. Therefore, the influence of the false color due to the presence of the image can be eliminated, and the determination in the unit image data determination unit 65 can be performed with high accuracy.

  The false color described above often becomes a problem when performing variable-size copying. This is because if the line interval of the R, G, B reading lines of the CCD line sensor 41a corresponds to an integral multiple of the reading interval in the sub-scanning direction, the reading line preceding the reading timing in the subsequent reading line. Although it is possible to prevent false colors by delaying more, even if the “integer multiple” relationship is established when performing the same size copy, the “integer multiple” relationship is established when performing the variable size copy. This is because there is no limit.

  For example, let us consider a case where the resolution in the sub-scanning direction at the time of copying at the same magnification is 600 dpi and the line interval corresponds to four times the reading interval. In this case, when performing the same-size copy, the false timing can be prevented by delaying the read timing in the subsequent read line by four cycles from the preceding read line, but when performing the 70% reduced copy. Since the resolution in the sub-scanning direction is 420 dpi and the line interval corresponds to 2.8 times the reading interval, even if the reading timing in the subsequent reading line is delayed by three cycles from the preceding reading line, The reading position is shifted by 0.2 times the reading interval from the subsequent reading line, which causes false color.

  In addition, a mechanical shift in the formation position of the R, G, B reading lines of the CCD line sensor 41a may cause false colors.

  In addition, the size of the pixel set should be determined according to the line interval, scaling factor, and the like, and the above-mentioned “vertical 2 × 2 horizontal 4 pixels” is merely an example. Absent.

  For example, in order to eliminate the influence of false colors when performing variable magnification copying, it is effective to increase the number of pixels in the sub-scanning direction of the pixel set as the “reading position shift” increases. Since the “positional deviation” is determined by the scaling factor, a table describing the relationship between the scaling factor and the number of pixels in the sub-scanning direction of the pixel set is prepared in the ROM 13 in advance, and the specified scaling factor is referred to the table. It is desirable to determine the number of pixels in the sub-scanning direction of the pixel set according to the magnification.

  Note that the averaging unit 61c averages image data that has not been subjected to resolution conversion by the resolution conversion unit 61a. Therefore, there is no particular problem even if the number of pixels in the main scanning direction of the pixel set is constant.

  Next, for each block divided by a predetermined number (for example, 30 × 30 = 900) of the first unit image data counting unit 66a, the first unit image data determination unit 65a includes the unit image data included in the block of interest. Belongs to a plurality of color regions mC, mM, mY, mR, mG, mB and achromatic region mA (S102). Subsequently, the first unit image data counting unit 66a applies to each of the color regions mC, mM, mY, mR, mG, mB, and the achromatic region mA based on the determination result of the first unit image data determination unit 65a. The number of unit image data to which it belongs is counted (S103). The unit image data determination process and counting process in steps S102 and S103 are executed for all unit image data in the block of interest (S104).

  Subsequently, when the determination process and the counting process are completed for all the unit image data in the target block, the first block determination unit 67a performs the target block determination process based on the counting result of the first unit image data counting unit 66a. Execute. That is, the first block determination unit 67a performs a determination process for the color included in the block of interest (S105).

  Subsequently, the first block counting unit 68a adds the block count values of the necessary color regions mC, mM, mY, mR, mG, mB and the achromatic region mA based on the determination result of the first block determining unit 67a. (S106).

  The block determination processing and counting processing in steps S105 and S106 are executed for all blocks in the document head image (S107).

  Subsequently, the ground color setting unit 71 is based on the block count value for each of the plurality of color regions mC, mM, mY, mR, mG, mB and the achromatic region mA counted by the first block counting unit 68a. A ground color is set, and the center of gravity position (ground color position) based on the ground color region P1 is calculated (S108). Then, based on the ground color set in step S108, an affine transformation formula in the color plane H is calculated (S109), and the ground color determination process ends.

Thus, in the ground color determination Priority determination of the image data in the present embodiment, it is possible to perform the determination and counting for each block. Therefore, it is possible to prevent an erroneous ground color from being detected by accumulating erroneous chromatic / achromatic determinations and performing color determination of image data based on the erroneous ground color. In addition, the counter for counting can be made small.

  In the present embodiment, the ground color can be set based on the unit image data included in the selected color region. Therefore, detection of an erroneous ground color can be further prevented.

<3.2. Color judgment procedure>
Here, the color determination of the target image will be described. As shown in FIG. 5, in step S201, a pre-determination process is performed on image data that has been affine transformed based on the affine transformation equation calculated in the ground color judgment process prior to color judgment.

  For example, as pre-determination processing in step S201, the image processing circuit 61 executes processing for generating unit image data from image data in the RGB color space, as in step S101 in FIG. The image processing circuit 61 performs gamma correction processing on the generated unit image data. Further, the image processing circuit 61 executes processing for converting the color space of the unit image data subjected to the gamma correction processing from RGB to Lab (lightness L and chromaticity a, b).

  As the unit image data, the pixel data of the image data may be used as it is without performing such an averaging process. However, preferably, as described above, not the pixel data itself but an average value of the values of each pixel included in the pixel group is used as the unit image data.

  Next, the first unit image data determination unit 65b determines whether the unit image data included in the block of interest belongs to a plurality of color regions mC, mM, mY, mR, mG, mB, or an achromatic region mA. (S202). In step S202, the second unit image data determination unit 82 determines whether the background color upper image data is included in the block of interest.

  Subsequently, the first unit image data counting unit 66b applies to each of the color regions mC, mM, mY, mR, mG, mB and the achromatic region mA based on the determination result of the first unit image data determination unit 65b. The number of unit image data to which it belongs is counted (S203). Further, the second unit image data counting unit 83 counts the number of background color upper image data based on the counting result of the second unit image data determining unit 82 (S203).

  The unit image data determination and counting process in steps S202 and S203 is executed until completion for all unit image data in the block of interest (S204).

  FIG. 6 is a diagram for explaining the ground color position calculated by the ground color setting unit 71. FIG. 7 is a diagram for explaining a method for enlarging the achromatic region when the image data is subjected to affine transformation based on the ground color set by the ground color setting unit 71. 8 and 9 are diagrams for explaining the ground color region P1.

Here, since the target image conversion unit 72 performs affine conversion (conversion of contents for moving the ground color position to the origin), the achromatic region is expanded in a direction corresponding to the movement, as shown in FIG. That is, the achromatic region is enlarged so that unit image data corresponding to the achromatic color (unit image data existing near the origin before affine transformation) does not deviate from the achromatic region. As shown in FIG. 6, the achromatic region mA before the enlargement process has a square shape and a point-symmetric shape with respect to the origin of the color plane H, and the length of one side of the achromatic region mA. Is 2W. When the ground color position (centroid position) on the color region mG where a3 = −Δa and b3 = Δb is moved to the origin of the color plane H by affine transformation, the region mA to be determined as an achromatic color is In the color area mB symmetrical to the color area mG across the origin of the plane H, it is necessary to enlarge Δa in the positive direction of the a3 axis and Δb in the negative direction of the b axis, respectively (see FIG. 7). That is, the region to be determined as an achromatic color is enlarged to the color region side symmetrical to the ground color region with the origin on the color plane H sandwiched between the ground color positions set by the ground color setting unit 71. The

  As described above, when the determination process is executed on the image data affine-transformed on the color plane H by the target image conversion unit 72, the first unit image data determination unit 65 sandwiches the origin of the color plane H across the ground. An area to be determined as an achromatic color is enlarged to a color area symmetrical to the color area. That is, in the determination process after affine transformation, the achromatic region mA is a region including the ground color region and the achromatic region. Then, the first unit image data determination unit 65 executes determination processing for each unit image data by using the enlarged achromatic region mA and the plurality of color regions mC, mM, mY, mR, mG, and mB. This makes it possible to select a processing mode for the image data read from a document having a ground color, such as colored paper, with the ground color as achromatic.

  Further, when the target image conversion unit 72 performs affine transformation based on the ground color, the ground color region P1 has a lightness range of Lmin ≦ L ≦ Lmax and a chromaticity range as shown in FIGS. , Amin ≦ a3 ≦ amax and bmin ≦ b3 ≦ bmax. Then, the second unit image data determination unit 82 executes determination processing for each unit image data based on the chromaticity range and the maximum value Lmax of the brightness range used as the brightness threshold.

  Note that the enlargement process of the achromatic region mA based on the ground color may be executed based on the setting of the user via the display unit 63 or the operation unit 64. For example, when it is desired to determine a manuscript with a color ground color (characters and the like are black) as monochrome, the user sets ON the enlargement process of the achromatic region mA based on the ground color. On the other hand, when it is desired to determine a document with a colored ground color (characters are black characters) as a color, it is set to OFF. As described above, the target image determination unit 80 may determine that the ground color portion of the document is an achromatic color based on the setting of the user via the display unit 63 and the operation unit 64.

  Subsequently, when the determination process and the counting process are completed for all the unit image data in the target block, the first block determination unit 67b, based on the counting result of the first unit image data counting unit 66b, Is determined to be included in each of the plurality of color regions mC, mM, mY, mR, mG, mB, and the achromatic region mA (S205). Further, the second block determination unit 84 determines the color included in the block of interest based on the counting result of the second unit image data counting unit 83 (S205).

  Subsequently, based on the determination result of the first block determination unit 67b, the first block counting unit 68b is configured to block the block for each of the plurality of color regions mC, mM, mY, mR, mG, mB, and the achromatic region mA. The block count value of the area determined to be included in the area is added (S206). Further, the second block counting unit 85 counts the number of blocks determined to include the ground color upper image data (S206).

  The block determination processing and counting processing in steps S205 and S206 are executed until completion for all the blocks in the image data (S207).

  When the processing mode is selected by the mode selection unit 81, the color determination process ends (S208 to S210). That is, when any of the plurality of color regions mC, mM, mY, mR, mG, mB and the block count value of the background color upper image data is equal to or greater than a predetermined threshold value or a brightness reference value, the mode selection unit In 81, the color mode is selected as the operation mode of the processing unit (S209). On the other hand, when any block count value is smaller than the predetermined threshold value or the brightness reference value (S208), the mode selection unit 81 selects the monochrome mode as the operation mode (S210).

<4. Modification>
(1) In the present embodiment, the ground color determination unit 70 and the target image determination unit 80 include a first unit image data determination unit 65 (65a, 65b) and a first unit image data count unit 66 (66a, 66b), respectively. The first block determination unit 67 (67a, 67b) and the first block counting unit 68 (68a, 68b) are described as being included, but the configuration of the image color determination device 1 is not limited to this.

  For example, as shown in FIG. 10, the ground color determination unit 70 and the target image determination unit 80 share the same functions with a first unit image data determination unit 65, a first unit image data counting unit 66, and a first block determination unit. 67 and the first block counting unit 68.

  (2) In the present embodiment, the ground color determination unit 70 and the target image determination unit 80 have been described as being implemented in a circuit (hardware) manner, but are not limited thereto. For example, the functions of the ground color determination unit 70 and the target image determination unit 80 may be realized by the MPU 11 based on the program 13 a stored in the ROM 13.

  (3) In the present embodiment, the ground color determination unit 70 performs ground color determination based on the document head image. However, the present invention is not limited to this. The background color determination may be performed using, for example, image data at the rear end of the image data or a part of the front end of the right end.

  In this embodiment, since the ground color determination is performed based on the leading image of the document, there is no storage means that can simultaneously store image data related to the document image for one page, and from the document leading image toward the document trailing edge image. The present invention can also be applied to an image color determination apparatus that performs sequential processing. That is, even if the capacity of the storage means is less than or equal to one page, the ground color determination result based on the document head image can be reflected in the subsequent color determination processing of the document image data. In the case where a storage unit capable of storing one page of document image data at the same time is provided, ground color determination can be performed based on not only the document leading image but also the document trailing edge image or document right edge image.

  (4) In the present embodiment, the region to be determined as achromatic (achromatic region) is expanded in the direction in which the image data is moved by affine transformation as shown in FIG. Is not necessarily limited to this. That is, when the ground color position on the two-dimensional color plane H is determined in the ground color determination, the achromatic region mA can be enlarged in a direction including the ground color position. The direction and size for enlarging the achromatic region mA are determined according to the positional relationship between the ground color position on the color plane H and the origin. In this case, affine transformation for the image data for color determination is not necessary.

  (5) Furthermore, in the present embodiment, the first unit image data determination unit 65 has been described as performing ground color determination and color determination on image data expressed in Lab. However, the present invention is not limited to this. It is not something. YCbCr, YIQ, Luv, etc., a color system that has parameters relating to brightness (for example, lightness L in Lab color space or luminance Y in YCbCr color space) and parameters relating to hue and saturation (color difference, chromaticity) Color systems other than Lab can also be used. In this embodiment, chromaticity is also used as a concept including all of color difference, hue, and saturation.

It is a figure which shows an example of a structure of the image color determination apparatus in embodiment of this invention. Used for processing by the first and second unit image data determination unit, the first and second unit image data counting unit, the first and second block determination unit, and the first and second block counting unit. It is a figure which shows an example of a color plane. It is a figure for demonstrating the method of acquiring a color plane. It is a flowchart for demonstrating the procedure of ground color determination. It is a flowchart for demonstrating the procedure of a color determination. It is a figure for demonstrating the ground color position calculated by the ground color setting part. It is a figure for demonstrating the expansion method of an achromatic area | region when affine transformation is performed to the target image based on the ground color set by the ground color setting part. It is a figure for demonstrating a ground color area | region. It is a figure for demonstrating a ground color area | region. It is a figure which shows the other example of a structure of the image color determination apparatus in embodiment of this invention.

Explanation of symbols

1,100 Image color determination device 21 NCU
22 MODEM 25 COMMUNICATION UNIT 31 CODEC
Reference Signs List 41 Scanner Unit 51 Recording Unit 61 Image Processing Circuit 63 Display Unit 64 Operation Unit 65 First Unit Image Data Determination Unit 66 First Unit Image Data Counting Unit 67 First Block Determination Unit 68 First Block Counting Unit 70 Ground Color Determination Unit 71 Ground color setting unit 72 Target image converting unit 80 Target image determining unit 81 Mode selecting unit 82 Second unit image data determining unit 83 Second unit image data counting unit 84 Second block determining unit 85 Second block counting unit

Claims (5)

In the two-dimensional color plane defined by the hue-related parameter and the saturation-related parameter, for each color region formed on the two-dimensional color plane, the number of constituent units of image data related to the target image is counted. While determining the presence or absence of the ground color of the target image using the counting result, if the target image has a ground color, determine a color region corresponding to the ground color among a plurality of color regions, A ground color determination unit that determines a lightness range and a chromaticity range of the ground color as a three-dimensional ground color region based on a three-dimensional distribution of structural units included in a color region corresponding to the ground color; ,
Using the determination result of the ground color determination unit, the image data related to the target image is translated so that the ground color position obtained based on the ground color region is the origin on the two-dimensional color plane. Accordingly, the target image determining unit that performs color determination before Symbol target image,
An image color determination apparatus comprising: The image color determination apparatus according to claim 1,
An image color determination apparatus, wherein a user can set whether or not to use a determination result of the ground color determination unit in color determination by the target image determination unit. The image color determination device according to claim 1 or 2,
The target image determination unit averages image data for each pixel set including a plurality of pixels, and performs color determination of the image data using the averaged pixel data as a constituent unit. In the two-dimensional color plane defined by the hue-related parameter and the saturation-related parameter, for each color region formed on the two-dimensional color plane, the number of constituent units of image data related to the target image is counted. While determining the presence or absence of the ground color of the target image using the counting result, if the target image has a ground color, determine a color region corresponding to the ground color among a plurality of color regions, A ground color determination step of determining a lightness range and a chromaticity range of the ground color as a three-dimensional ground color region based on a three-dimensional distribution of structural units included in the color region corresponding to the ground color; ,
Using the determination result of the background color determination step, the image data related to the target image is translated so that the background color position obtained based on the background color region is the origin on the two-dimensional color plane. Accordingly, the target image determining step of performing color determination before Symbol target image,
An image color determination method comprising: In the two-dimensional color plane defined by the hue-related parameter and the saturation-related parameter, for each color region formed on the two-dimensional color plane, the number of image data constituent units related to the target image is counted. While determining the presence or absence of the ground color of the target image using the counting result, if the target image has a ground color, determine a color region corresponding to the ground color among a plurality of color regions, A ground color determination process for determining a lightness range and a chromaticity range of the ground color as a three-dimensional ground color region based on a three-dimensional distribution of structural units included in a color region corresponding to the ground color; ,
Using the determination result of the ground color determination process, the image data related to the target image is translated so that the ground color position obtained based on the ground color region is the origin on the two-dimensional color plane. Accordingly, the target image determining process that performs color determination before Symbol target image,
A program for image color determination to be realized by a computer.

Images