JP2009071736A - Image processing apparatus, image processing method, program, and recording medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To achieve color clustering without preparing a color histogram from an input image during image processing. <P>SOLUTION: Color information of each pixel in an acquired color clustering processing target image is compared with color clustering result data stored in a storage apparatus to determine a color of each pixel. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a color clustering technique in image processing.

  A technique called color clustering is frequently used for image compression and image quality improvement. Color clustering is an algorithm for obtaining a set of points that can be treated as a set when each pixel of the image is considered as a point on the color space and all the pixels in the image are projected onto the color space (see FIG. 8). . The color clustering algorithm basically takes a long processing time. For example, the processing can be shortened by limiting the processing range (dividing the image into several parts and dividing the color space) (see FIG. 9). Color clustering on multiple pages was not possible.

Here, from the circumscribed rectangle generated from the input image, a circumscribed rectangle having a size or position within a predetermined range is extracted, a search area is set within a certain range from the extracted circumscribed rectangle, and this search area A technique that can extract a pattern to be extracted with high accuracy even when the boundary of the pattern to be extracted is unclear by extracting a specific pattern from the input image based on the circumscribed rectangle There is.
JP 2005-302056 A

  However, creating a color histogram from an input image has a problem that it takes a long time and consumes memory.

  The present invention has been made in view of such a situation, and an object of the present invention is to realize color clustering without creating a color histogram from an input image during image processing.

  According to the first aspect of the present invention, the color information of each pixel in the acquired color clustering processing target image is compared with the color clustering result data stored in the storage device to determine the color of each pixel. An image processing apparatus is characterized.

  According to a second aspect of the present invention, in the image processing device according to the first aspect, the color clustering result data stored in the storage device is updated based on the color clustering result data of the color clustering target image. It is characterized by.

  According to a third aspect of the present invention, in the image processing device according to the first or second aspect, the difference between the pixel data of the color clustering processing target image and each cluster center which is the stored color clustering result data is obtained. The color cluster center to which the pixel belongs is searched based on the difference, and the value of the pixel is replaced with the value of the color cluster center to which the pixel belongs.

  According to a fourth aspect of the present invention, in the image processing apparatus according to the first or second aspect, the difference between the pixel data of the color clustering processing target image and each cluster center which is the stored color clustering result data is obtained. If all the differences are greater than or equal to the threshold value, the target pixel is registered as the provisional color cluster center. If all the differences are less than or equal to the threshold value, the color cluster center to which the pixel belongs is searched based on the difference, and the pixel value It is characterized by replacing it with a value.

  According to the fifth aspect of the present invention, the color information of each pixel in the acquired color clustering processing target image is compared with the color clustering result data stored in the storage device, and the color of each pixel is determined. This is a featured image processing method.

  According to a sixth aspect of the present invention, in the image processing method according to the fifth aspect, the color clustering result data stored in the storage device is updated based on the color clustering result data of the color clustering target image. It is characterized by that.

  The invention according to claim 7 is the image processing method according to claim 5 or 6, wherein the difference between the pixel data of the color clustering processing target image and each cluster center which is the stored color clustering result data is obtained. The color cluster center to which the pixel belongs is searched based on the difference, and the value of the pixel is replaced with the value of the color cluster center to which the pixel belongs.

  According to an eighth aspect of the present invention, in the image processing method according to the fifth or sixth aspect, the difference between the pixel data of the color clustering processing target image and each cluster center which is the stored color clustering result data is obtained. If all the differences are equal to or greater than the threshold, the target pixel is registered as the provisional color cluster center. If all the differences are equal to or less than the threshold, the color cluster center to which the pixel belongs is searched based on the difference, and the pixel value is assigned to the color cluster center to which the pixel belongs. It is characterized by replacing it with a value.

  The invention according to claim 9 is a program that causes a computer to realize the function according to any one of claims 1 to 4.

  A tenth aspect of the present invention is a computer-readable recording medium storing the program according to the ninth aspect.

  According to the present invention, color clustering can be realized without creating a color histogram from an input image during image processing.

  (Embodiment 1) This embodiment shows an example in which a document to be stored as a digital image is captured as an image using an image processing apparatus, and the number of colors of the image is reduced using color clustering.

  FIG. 1 is a diagram illustrating a configuration of an image processing apparatus according to the present exemplary embodiment. The CPU 1, memory 2, CD-ROM drive 3, hard disk 4, input device 5, display 6, mouse, etc. A recording medium 7 such as a CD-ROM stores a program for realizing the processing functions and processing procedures of the present invention. The document image to be processed is stored in the hard disk 4 or the like through the input device 5, for example. The CPU 1 reads out and executes a program for realizing the processing functions and procedures described above from the recording medium 7, stores the result of the image processing in the hard disk 4, and outputs it to the display 6 as necessary. Examples of the image processing apparatus include a copying machine (multifunction machine), a FAX, a scanner, a digital camera, and the like.

  FIG. 2 is a functional block diagram of the image processing apparatus. Image acquisition unit 21, data storage unit 22, color histogram creation unit 23, color histogram information storage unit 24, color clustering unit 25, color clustering information storage unit 26, pixel information extraction unit 27, and pixel color A determination unit 28, a signal replacement unit 29, and a cumulative color histogram creation unit 30 are provided.

  The image acquisition unit 21 acquires first image data for performing color clustering and second image data for performing signal replacement processing. The data storage unit 22 stores each acquired image data, image data after signal replacement processing, and the like.

  The color histogram creation unit 23 creates a color histogram of the image data and stores the result data in the color histogram information storage unit 24. The color clustering unit 25 performs color clustering based on the color histogram of the image data, and stores the result data including the center (cluster center) of each cluster constituting the color cluster in the color clustering information storage unit 26.

  The pixel information extraction unit 27 extracts color information (or luminance information) in each pixel of the second image data for which cluster center replacement processing is performed. The pixel color determination unit 28 compares the color information (or luminance information) of each pixel in the second image with a plurality of color cluster centers recorded in the color clustering information storage unit 26, and determines the color of each pixel. decide. When there is a color cluster center to which each pixel belongs, the signal replacement unit 29 replaces the signal of the target pixel with the signal of the color cluster center.

  The cumulative color histogram creation unit 30 stores the color histogram of the second image data created by the color histogram creation unit 23 and the first stored in the color histogram information storage unit 24 when the image processing apparatus is in an idle state. The color histograms of the image data are integrated to create a cumulative color histogram, and the data is stored in the color histogram information storage unit 24.

  Next, the operation of the present embodiment will be described in detail with reference to the drawings.

  First, the image acquisition unit 21 of the image processing apparatus acquires first image data for performing color clustering, and records the acquired first image data in the data storage unit 22 (step S301).

  Next, the color histogram creation unit 23 of the image processing device creates a color histogram of the first image data, and stores the result data in the color histogram information storage unit 24 (step S302).

  The color clustering unit 25 of the image processing device performs color clustering based on the color histogram of the first image data created in step S302 described above, and each cluster constituting the color cluster created according to the first image. The result data including the center (cluster center) is stored in the color clustering information storage unit 26 (step S303).

  Next, the image acquisition unit 21 of the image processing apparatus acquires the second image data, and records the acquired second image data in the data storage unit 22 (step S304). The image processing apparatus stores a cumulative color histogram, the number of color clusters, and the color cluster center acquired from previously input image data (see FIG. 4).

  The pixel information extraction unit 27 of the image processing apparatus extracts color information (or luminance information) in each pixel of the second image data (step S305).

  The pixel color determination unit 28 of the image processing apparatus compares the color information (or luminance information) of each pixel in the second image with a plurality of color cluster centers recorded in the color clustering information storage unit 26 (step In step S306, it is determined whether there is a color cluster center to which each pixel belongs, and the color of each pixel is determined according to the determination result (step S307). Note that the plurality of color cluster centers here are color cluster centers as a result of color clustering using past cumulative color histograms relating to past image data including the first image data.

  A method for discriminating the color cluster to which each pixel belongs using a plurality of color cluster centers recorded in the color clustering information storage unit 26 will be described below. Various methods are conceivable as the discrimination method, but here, the sum of the differences between the RGB signals is used. Here, the description will be made on the assumption that each RGB signal is used as the center of the color cluster. However, as a color space, RGB (red, green, blue) and CMY (cyan) are used. ), Magenta, yellow), CMYK (cyan, magenta, yellow, black), CMK (cyan, magenta, black) black)), HSV (= HSB) (hue, saturation value, lightness (value)), HLS (= HSL, HSI) (hue, saturation, luminance) )) May be used.

  Specifically, the difference between the RGB signal values of the pixels of the second image and the RGB signals at the center of each cluster is obtained, and the sum of the absolute values is obtained (step S306-1).

  Thereafter, the color cluster center having the smallest sum is searched for and the target pixel is assigned to the color cluster center (step S306-2) (see FIG. 5).

  When there is a color cluster center to which each pixel belongs, the signal replacement unit 29 of the image processing apparatus replaces the signal of the target pixel with the signal of the color cluster center (step S308). That is, the RGB signals of the target pixel are replaced with RGB signals at the center of the color cluster (see FIG. 6).

  An image in which each pixel is replaced is stored in the data storage unit 22 (step S309).

  When the image processing apparatus is in an idle state or a low processing load, the color histogram creation unit 23 of the image processing apparatus creates a color histogram of the second image data, and the cumulative color histogram creation unit 30 The color histogram of the second image data created by the histogram creation unit 23 and the color histogram of the first image data stored in the color histogram information storage unit 24 are integrated to create a cumulative color histogram. Data is stored in the color histogram information storage unit 24. (Step S310). For example, when the third image is processed, the integrated color histogram of the first image and the second image and the color histogram of the third image are not integrated with the color histogram of the first image. Will be integrated.

  Further, the color clustering unit 25 of the image processing apparatus performs color clustering using the cumulative color histogram created in step S310, finds the center of each cluster (cluster center), and uses the result data as color clustering information. The data is stored in the storage unit 26 (step S311). The number of color clusters and the color cluster center may be updated from the cumulative color histogram.

  The above processing realizes image color reduction processing and the like. The advantage of this method is that, since color clustering is performed after image color reduction processing, no load is generated during image color reduction processing. That is, it is possible to improve the processing speed of color clustering using data used in the past. In addition, when the tendency of the input images is similar, the color reduction processing result is also appropriate.

  According to the present embodiment, the color clustering process can be shortened and a global processing result can be obtained. Specifically, by obtaining the color clustering result from the accumulated color histogram stored in the image processing device, the time required for obtaining the color cluster is shortened, and color clustering that is optimal for documents handled by the image processing device from previous data The result can be determined.

(Example 2)
In the second embodiment, a process of determining the color cluster to which each pixel of the image captured from the image processing apparatus of the first embodiment belongs using the color cluster center stored in the image processing apparatus, that is, the process in step S306 described above. The contents of are partially changed.

  Referring to FIG. 7, the difference between the RGB signals of the pixels of the second image and the RGB signals at the center of each cluster is obtained, and the sum of the absolute values is obtained (step S706-1).

  Then, it is determined whether all the sums obtained above are equal to or greater than the threshold value L (step S706-2). (Step S706-3). On the other hand, if all the values are equal to or smaller than the threshold value L (step S706-2 / No), they are assigned to the color cluster center having the smallest sum (step S706-4).

  Further, when the color cluster center to which the pixel belongs is found, the RGB signals of the pixels of the second image are replaced with the RGB signals of the color cluster center (step S708), and the image in which the pixels are replaced is stored (step S708). After S709), the provisional color cluster center is deleted (step S710).

  The advantage of this method is that even if a value different from the existing color cluster center is suddenly generated, the color can be maintained without being reduced.

It is a figure explaining the structure of Example 1 of this invention. 1 is a functional block diagram of an image processing apparatus according to a first embodiment of the present invention. It is a flowchart which shows operation | movement of Example 1 of this invention. It is a figure explaining the outline of Example 1 of the present invention. FIG. 10 is an explanatory diagram for finding a color cluster center having the smallest sum and assigning a target pixel to the color cluster center; It is explanatory drawing which replaces each RGB signal of an object pixel with each RGB signal of a color cluster center. It is a flowchart which shows operation | movement of Example 2 of this invention. It is a figure explaining the algorithm of a color clustering. It is a figure explaining an example which restrict | limits a process range.

Explanation of symbols

1 CPU
2 Memory 3 CD-ROM drive 4 Hard disk 5 Image processing device 6 Display 7 Recording medium 21 Image acquisition unit 22 Data storage unit 23 Color histogram creation unit 24 Color histogram information storage unit 25 Color clustering unit 26 Color clustering information storage unit 27 Pixel information Extraction unit 28 Pixel color determination unit 29 Signal replacement unit 30 Cumulative color histogram creation unit

Claims (10)

  An image processing apparatus comprising: comparing color information of each pixel in an acquired image for color clustering processing with color clustering result data stored in a storage device to determine a color of each pixel.   The image processing apparatus according to claim 1, further comprising: updating color clustering result data stored in a storage device based on color clustering result data of the processing target image for color clustering.   Find the difference between the pixel data of the color clustering processing target image and each cluster center that is the stored color clustering result data, and based on the difference, find the color cluster center to which the pixel belongs, The image processing apparatus according to claim 1, wherein the value is replaced with a value at the center of the color cluster to which the value belongs.   The difference between the pixel data of the color clustering processing target image and each cluster center which is the stored color clustering result data is obtained, and when all the differences are not less than the threshold, the target pixel is registered as the provisional color cluster center. 3. The image according to claim 1, wherein if all of the values are equal to or less than the threshold value, the color cluster center to which the pixel belongs is searched based on the difference, and the value of the pixel is replaced with the value of the color cluster center to which the pixel belongs. Processing equipment.   An image processing method comprising: comparing color information of each pixel in the acquired image for color clustering processing with color clustering result data stored in a storage device to determine a color of each pixel.   6. The image processing method according to claim 5, further comprising: updating color clustering result data stored in a storage device based on the color clustering result data of the color clustering processing target image.   Find the difference between the pixel data of the color clustering processing target image and each cluster center that is the stored color clustering result data, and based on the difference, find the color cluster center to which the pixel belongs, 7. The image processing method according to claim 5, wherein the value is replaced with a value at the center of the color cluster to which the value belongs.   The difference between the pixel data of the color clustering processing target image and each cluster center which is the stored color clustering result data is obtained, and when all the differences are not less than the threshold, the target pixel is registered as the provisional color cluster center. 7. The image according to claim 5 or 6, wherein if all of the values are equal to or less than the threshold, the color cluster center to which the pixel belongs is searched based on the difference, and the value of the pixel is replaced with the value of the color cluster center to which the pixel belongs. Processing method.   A program for causing a computer to realize the function according to any one of claims 1 to 4.   A computer-readable recording medium storing the program according to claim 9.

Images