JP2012114869A - Image processing device, image processing method, image processing program and program recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing device, an image processing method, an image processing program and a program recording medium for determining whether the color of the ground of an image is a chromatic color or an achromatic color.SOLUTION: An image processing device comprises: an image input unit for inputting an image; a minimum value filter processing unit for generating a minimum value filter image of saturation from the image; a count unit for counting the number of pixels with a predetermined saturation value or less from the minimum value filter image; and a determination output unit for determining that the color of the ground of the image is an achromatic color when the ratio of the number of pixels with the predetermined saturation value or less in the minimum value filter image to the number of all pixels of the image is equal to or greater than a predetermined threshold value, and the color of the ground of the image is a chromatic color when it is smaller than the predetermined threshold value.

Description

  The present invention relates to an image processing technique, and relates to an image processing apparatus, an image processing method, an image processing program, and a program recording medium that determine whether a background of an image is a chromatic color or an achromatic color.

  In recent years, with the widespread use of digital cameras, scenes in which documents such as paper and business cards, posters at exhibitions, whiteboards and blackboards used in meetings, etc. are imaged and used as digital image data are spreading. These digital image data are output as noise when output to a screen or paper due to digital noise at the time of imaging, dust or dirt originally attached to the subject itself, and the image becomes dirty. It was. In particular, in an image such as a document, a whiteboard, or a blackboard, an area that occupies most of an image other than characters and figures (hereinafter, an area that occupies an image other than letters and figures is referred to as a base) is uniform. In many cases, the achromatic colors are almost the same, and noise is further emphasized, resulting in poor visibility.

  Therefore, chromatic images such as paintings and posters with a non-uniform background are automatically distinguished from achromatic images with the same uniform background such as documents, whiteboards, and blackboards. In order to perform an appropriate image correction and obtain an image with good visibility, a technique for determining whether the background of the image is a chromatic color or an achromatic color is required.

  In Patent Document 1, it is determined whether the read data is a color pixel or a black-and-white pixel in units of pixels, and if the number of consecutive pixels determined to be color pixels and the number of appearances exceed a predetermined number, An apparatus for discriminating is disclosed.

  In Patent Document 2, edge detection is performed on a line drawing in an image to remove the line drawing by removing an edge portion, and the count of pixels that are determined not to be edges and that are determined to be chromatic colors. An apparatus is disclosed that compares a number with a threshold and discriminates a document as a monochrome image when the count number is small.

JP-A-4-282968 (published October 8, 1992) JP 2009-5398 A (published January 8, 2009)

  However, the devices disclosed in Patent Documents 1 and 2 each determine whether an image is a color image or a monochrome image, and do not have a function of determining the background. In addition, since the device described in Patent Document 1 is a device that determines whether the entire image is a chromatic color or an achromatic color, if characters or drawings are drawn in a chromatic color even if the background is an achromatic color, May not be correctly determined as an achromatic color. Furthermore, in the apparatus described in Patent Document 2, in the case of an image that is chromatic and has characters or figures drawn with thick lines, or an image that has a large proportion of line drawings in the entire image, Since the center portion is not judged by the edge, the number of pixels that are chromatic colors in particular increases, and it has not been possible to appropriately determine whether the background is a chromatic color or an achromatic color.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide an image processing apparatus, an image processing method, an image processing program, and a program recording for appropriately determining whether the background of an image is a chromatic color or an achromatic color. The purpose is to provide a medium.

  An image processing apparatus according to the present invention includes an image input unit that inputs an image, a minimum value filter processing unit that generates a minimum saturation filter image from the image, and a predetermined saturation value or less from the minimum value filter image. When the ratio of the number of pixels equal to or less than a predetermined saturation value of the minimum value filter image to the total number of pixels of the image is equal to or greater than a predetermined threshold, there is no background of the image. When the chromatic color is less than a predetermined threshold, a determination output unit that determines that the background of the image is a chromatic color is provided. The image processing apparatus according to the present invention is characterized in that the image processing apparatus includes a shift estimation unit that estimates and corrects the shift of the predetermined saturation value.

  The image processing apparatus according to the present invention is characterized in that the image processing apparatus includes a reduced image generation unit that generates a reduced image from the image. Further, the image processing apparatus according to the present invention is characterized in that the minimum value filter processing unit varies the size of the generated minimum value filter image.

  Further, the image processing apparatus according to the present invention is characterized in that the minimum value filter processing unit extracts pixels from the image at regular intervals, and performs minimum value filter processing on the pixels.

  The image processing method according to the present invention includes an image input step of inputting an image, a minimum value filter processing step of generating a minimum value filter image of saturation from the image, and a predetermined saturation from the minimum value filter image. A counting step for counting the number of pixels equal to or less than a value; and a ratio of the number of pixels equal to or less than a predetermined saturation value of the minimum value filter image to the total number of pixels of the image is equal to or greater than a predetermined threshold value, And a determination output step of determining that the background of the image is a chromatic color when the color is less than a predetermined threshold value.

  An image processing program according to the present invention includes an image input step for inputting an image, a minimum value filter processing step for generating a saturation minimum value filter image from the image, and a predetermined saturation from the minimum value filter image. A counting step for counting the number of pixels equal to or less than a value; and a ratio of the number of pixels equal to or less than a predetermined saturation value of the minimum value filter image to the total number of pixels of the image is equal to or greater than a predetermined threshold value, Is less than a predetermined threshold, the computer is caused to execute a determination output step for determining that the background of the image is a chromatic color.

  In addition, a program recording medium according to the present invention records the above-described image processing program.

  According to the present invention, even if an image in which characters and figures are drawn in chromatic colors or an image in which characters and figures are written in chromatic colors and thick lines, the background is chromatic or achromatic. Can be correctly determined.

It is a block diagram which shows the structure of the image processing apparatus which concerns on this invention. It is a flowchart which shows the process of the image processing apparatus which concerns on the 1st Embodiment of this invention. It is a figure which shows the specific example of a minimum value filter process. It is a figure which shows an example of the histogram obtained from the image whose background is a chromatic color. It is a figure which shows an example of the histogram obtained from the image whose background is an achromatic color. It is a figure which shows the example of the minimum value filter process. It is a figure which shows the example of the minimum value filter process. It is a figure which shows an example of the histogram obtained from the image in which the background is achromatic and the white balance is lost. It is a block diagram which shows the structure of the image processing apparatus which concerns on the 2nd Embodiment of this invention. It is a flowchart which shows the process of the image processing apparatus which concerns on the 2nd Embodiment of this invention. It is a figure for demonstrating the example of the process which estimates the shift | offset | difference of the saturation value by the white balance breaking. It is a figure which shows an example of the histogram obtained from the image where the background is achromatic and the white balance is lost when the saturation value C is changed. It is a block diagram which shows the structure of the image processing apparatus which concerns on the 3rd Embodiment of this invention. It is a flowchart which shows the process of the image processing apparatus which concerns on the 3rd Embodiment of this invention.

  An embodiment of an image processing apparatus 1 according to the present invention will be described below with reference to the drawings. In the drawings of the present invention, the same reference numerals represent the same or corresponding parts.

  FIG. 1 is a block diagram showing a basic configuration of the image processing apparatus 1. The image processing apparatus 1 according to the present invention includes an image input unit 11, a minimum value filter processing unit 12, a counting unit 13, a determination output unit 14, a memory unit 15, and a function control unit 16.

  The image input unit 11 includes a CCD camera, a scanner, and the like, and receives an image such as a document or a poster. As long as the input image to be input is a color image, for example, an image captured by any imaging means such as an image captured by a digital camera or an image read by a scanner may be used.

  The minimum value filter processing unit 12 performs minimum value filter processing on the saturation component of the input image to generate a minimum value filter image of the saturation component. Although details will be described later, the minimum value filtering process is a process of replacing the pixel value of the target pixel of the input image with the minimum value of the pixel value of the pixel in the predetermined area. The counting unit 13 counts a ratio P of the number of pixels equal to or less than the saturation value C in the total number of pixels of the generated minimum value filter image. The saturation value C may be set as appropriate. By dividing the total number of pixels of the minimum value filter image from the number of pixels having the saturation value C or less of the saturation value filter image, the ratio P of the number of pixels having the saturation value C or less to the total number of pixels is calculated. The

  The determination output unit 14 determines whether the ratio P counted by the counting unit 13 is equal to or greater than the threshold th or less than the threshold th. If the ratio P is equal to or greater than the threshold th, the background of the input image is determined to be an achromatic color. If it is less than that, the background of the input image is determined to be a chromatic color, and the determination result is output. The memory unit 15 includes, for example, a ROM (Read Only Memory) and a RAM (Random Access Memory), and stores programs and work data related to the operation of the image processing apparatus 1. The function control unit 16 is realized by a CPU (Central Processing Unit) and a program executed on the CPU, and controls the function of each component of the image processing apparatus 1.

(First embodiment)
The processing procedure of the image processing apparatus 1 will be described with reference to a flowchart showing the overall processing flow of the present embodiment shown in FIG.

  First, the image processing apparatus 1 is activated, and the image input unit 11 converts the paper surface into electronic data such as a CCD camera or a scanner and inputs it (step S101). Next, the minimum value filter processing unit 12 creates a minimum value filtered image of the saturation component from the input image data input in step S101 (step S102). Here, the saturation component minimum value filtering process will be described. The saturation component minimum value filtering process is a process of performing a minimum value filtering process on the saturation component of the input image to obtain a saturation component minimum value filter image. In the minimum value filtering process, the pixel value of the target pixel of the input image is replaced with the minimum value of the pixel values of the pixels in the region and output.

  FIG. 3 shows a specific example of the minimum value filtering process. In the minimum value filter processing shown in FIG. 3, the pixel of interest A and eight pixels surrounding the pixel of interest A, that is, a 3 × 3 pixel region are employed as pixels in the vicinity region. For example, in FIG. 3A, the target pixel A in the input image is normally the center of a 3 × 3 pixel region (45). The saturation of the eight pixels surrounding the target pixel A is (129), (54), (12), (7), (56), (15), (128), (186). Of the 3 × 3 pixel area including one pixel and (45) corresponding to the target pixel A, the minimum value of the saturation is (7). Accordingly, the target pixel B in the output image shown in FIG. 3B is (7), and when the minimum value filtering process is performed on the target pixel A in the input image, (7) is output as the target pixel B. The Next, as shown in FIG. 3C, when the target pixel A is set to (7), and the minimum value filtering process is performed on the eight pixels surrounding the target pixel A and the target pixel A, FIG. ), The target pixel B in the output image is (4). When the minimum value filtering process is performed on (7) corresponding to the target pixel A in the input image, (4) is the target pixel B. Is output. In this way, by performing the minimum value filtering process on the entire input image, it is possible to reduce the influence of a non-background part such as a character or a figure that affects the accuracy of the background determination result of the image.

  Next, the counting unit 13 divides the total number of pixels in the minimum value filter image from the number of pixels in which the saturation value of the minimum value filter image is C or less, and the number of pixels having the saturation value C or less in the total number of pixels. Is counted (step S103). As for the counting process of the ratio P, for example, the image may be simply scanned by raster scanning or may be counted after creating a histogram.

  Next, the determination output unit 14 compares the threshold P with the ratio P of the number of pixels equal to or less than the saturation value C in the total number of pixels counted in step S103, and determines the background of the image (step S104). .

  Here, a process of determining the background of the image by comparing the ratio P of the number of pixels equal to or less than the saturation value C in the total number of pixels and the threshold th with reference to FIGS. 4 and 5.

  FIG. 4 is a diagram illustrating an example of a histogram of an image whose background is a chromatic color, where the vertical axis indicates the number of pixels and the horizontal axis indicates a saturation value. In an image in which the background is a chromatic color, since the background that occupies most of the entire image is a chromatic color, pixels are distributed over a wide range of saturation values, and the number of pixels below the saturation value C in the total number of pixels ( The ratio P of the hatched portion in FIG.

  FIG. 5 is a diagram showing an example of a histogram of an image in which the background is achromatic. The vertical axis indicates the number of pixels and the horizontal axis indicates the saturation value. An image with an achromatic background is an achromatic color that occupies most of the entire image, so most of the pixels are distributed in a range where the saturation value is less than or equal to the saturation value C, and the saturation that occupies the total number of pixels. The ratio P of the number of pixels equal to or smaller than the value C (shaded portion in FIG. 5) is increased.

  Therefore, by setting an appropriate value for the threshold th that determines whether the background of the image is an achromatic color or a chromatic color, the ratio P is compared with the threshold th to determine whether the background of the image is an achromatic color or a chromatic color. can do. If the ratio P is greater than or equal to the threshold th, the background of the image is determined to be achromatic (step S105), and the background determination result is output (step S107). If the ratio P is less than the threshold th, the background of the image is determined to be a chromatic color (step S106), and the background determination result is output (step S107). By such a procedure, it is possible to determine whether the background of the input image is a chromatic color or an achromatic color.

  In step S102, by creating a saturation minimum value filter image, it is possible to reduce the influence of non-background parts such as characters and figures that affect the accuracy of the background determination result of the image, as shown in FIG. In the numerical value indicated by the histogram, a remarkable increase or decrease appears with respect to the change in saturation, and the accuracy of the background determination can be increased.

  In FIG. 3, as a specific example of the minimum value filtering process in step S102, the pixel of interest is described using 3 × 3 pixels including the pixel of interest and eight pixels surrounding the pixel of interest as pixels in the vicinity region. It is possible to appropriately change the size of the minimum value filter in consideration of the size of the input image, the characters included in the image, and the thickness of the line in the figure.

  Further, the minimum value filtering process may be performed on pixels extracted from the input image at regular intervals. FIG. 6 shows a procedure for performing minimum value filtering on pixels extracted from the input image at regular intervals. In FIG. 6A, pixels are extracted from the input image by skipping one pixel. Here, the pixel to which the black circle is given is the extracted pixel. Next, in FIG. 6B, when the minimum value filtering process is performed within the 3 × 3 region of the extracted pixel centered on the target pixel A, the target pixel B is output in FIG. The In this way, by performing the minimum value filtering process on the pixels extracted every other pixel, the number of target pixels becomes 1/2 each of the vertical and horizontal pixels compared to before extraction, and the total number of pixels is 1/4. Therefore, the number of times of filter processing is also reduced to ¼, and the number of pixels of the minimum value filter image with respect to the number of pixels of the input image is ¼. As a result, by extracting pixels from the input image at regular intervals, the minimum value filtering process that takes the processing load and processing time of the image processing apparatus 1, and the number of pixels equal to or less than the saturation value C in the total number of pixels. The burden of the counting process of the ratio P can be reduced.

  Further, as shown in FIG. 7, a method may be used in which pixels of the input image are filtered by skipping the target pixel A at regular intervals. In FIG. 7A, the target pixel A is a pixel with a black circle. For this pixel of interest A, a minimum value filtering process is performed on the 3 × 3 pixel region, and then, as shown in FIG. 7B, from the position of the pixel of interest A in FIG. A pixel obtained by skipping one pixel in the right lateral direction is set as a target pixel A, and minimum value filtering is performed. Next, similarly, as shown in FIG. 7C, the minimum value filtering process is performed with the pixel that has been skipped by one pixel from the position of the target pixel A in FIG. Note that, as shown in FIG. 7C, the same processing is performed assuming that there is a pixel at the end of the input image. Even in such a method, the processing load can be reduced as described above.

(Second Embodiment)
Next, a second embodiment of the present invention will be described with reference to the drawings. In the present embodiment, processing when the white balance of an input image is lost will be described. When the white balance is lost, a color that looks natural to the naked eye appears as an unnatural color when taken with a camera. A method for correcting this and determining the background correctly will be described below.

  FIG. 8 is a diagram showing an example of a histogram of an image in which the background is an achromatic color and the white balance is lost. The vertical axis indicates the number of pixels, and the horizontal axis indicates the saturation value. Since the background is an achromatic image, most pixels are normally distributed in a range with a small saturation value as shown in FIG. However, since the white balance is lost, the achromatic portion becomes a thin chromatic color such as light cyan or magenta. Therefore, compared with the histogram of FIG. The histogram is shifted to the right side in FIG. Therefore, since the ratio P of the number of pixels equal to or less than the saturation value C (the hatched portion in FIG. 8) in the total number of pixels becomes small, there arises a problem that it is impossible to accurately determine whether the background of the image is a chromatic color or an achromatic color. On the other hand, it is possible to correctly determine the background by performing processing for the saturation value shift G due to the loss of white balance.

  With reference to FIG. 9 and FIG. 10, the processing procedure of the image processing apparatus 1a according to the present embodiment will be described. FIG. 9 is a block diagram showing the configuration of the image processing apparatus 1a in the present embodiment. FIG. 10 is a flowchart showing the overall processing flow of this embodiment. In this embodiment, in addition to the processing procedure of FIG. 2 shown in the first embodiment, the processing of step S203 and step S204 is added. Hereinafter, the process of step S203 and step S204 will be described in detail.

  The deviation estimation unit 17 estimates the saturation value deviation G due to the loss of the white balance based on the saturation value minimum value filtered image created in step S202 (step S203).

  Here, a process of estimating the magnitude G of the saturation value shift G will be described with reference to FIG. FIG. 11 is an enlarged view of a small saturation value range from the histogram of FIG. 8, where the vertical axis indicates the number of pixels and the horizontal axis indicates the saturation value. Saturation value deviation G due to the loss of white balance is calculated by counting the number of pixels for each saturation value from saturation value 0, and is the smallest of the saturation values having a count value exceeding a predetermined number of pixels. Estimate to saturation value.

  Next, a saturation value obtained by adding the saturation value shift G estimated in step S203 to the predetermined saturation value C is set as a new saturation value C ′ (step S204).

  FIG. 12 is a diagram showing an example in which the saturation value C is changed to the saturation value C ′ based on the processing procedure of FIG. 10 with respect to the histogram of FIG. The horizontal axis indicates the saturation value. In consideration of the saturation value shift G due to the loss of the white balance, the saturation value C is shifted to the right on the horizontal axis in FIG. The ratio P of the number of pixels (shaded area in FIG. 12) increases, and it is determined that the background is achromatic. As described above, even when the white balance is lost, it is possible to accurately determine whether the background of the image is a chromatic color or an achromatic color by calculating and using the deviation G of the saturation value.

(Third embodiment)
Next, a third embodiment of the present invention will be described with reference to the drawings. The present embodiment is different from the above-described embodiment in that a reduced image is created from an input image and a minimum value filter image is created based on the created reduced image.

  With reference to FIG. 13 and FIG. 14, the processing procedure of the image treatment apparatus 1b according to the present embodiment will be described. FIG. 13 is a block diagram showing a configuration of the image processing apparatus 1b in the present embodiment. FIG. 14 is a flowchart showing the overall processing flow of the present embodiment. In this embodiment, in addition to the processing procedure of FIG. 2 shown in the first embodiment, the process of step S302 is added. Hereinafter, the process of step S302 will be described in detail.

  The reduced image generation unit 18 creates a reduced image from the input image input in step S301 (step S302). As a process for generating a reduced image, for example, there are the following methods.

  In order to generate a reduced image in which N × N pixels of the input image stored in the memory unit 15 are one pixel, the input image is divided into a matrix for each N × N pixel, and is within each divided range. An average value is calculated for each saturation of pixel values, and one pixel is assigned according to the saturation value of the average value. Generally, N is about 4 to 16, and it is desirable to select a value corresponding to the input image. Since noise can be reduced by creating a reduced image, it is possible to more accurately determine whether the background of the image is a chromatic color or an achromatic color. Further, it is possible to reduce the burden in each step of the minimum value filtering process that takes the processing load and processing time of the image processing apparatus 1b and the counting process of the ratio P of the number of pixels equal to or less than the saturation value C in the total number of pixels. it can.

  As described above, according to the present invention, even if an image in which characters and figures are drawn in chromatic colors or an image in which characters and figures are written in chromatic colors and thick lines, the background is provided. It becomes possible to correctly determine whether the color is achromatic or achromatic.

  The image processing apparatus, the image processing method, the image processing program, and the program recording medium according to the present invention can be suitably used for, for example, a digital camera, a printer, a mobile phone, and the like.

DESCRIPTION OF SYMBOLS 1, 1a, 1b Image processing apparatus 11 Image input part 12 Minimum value filter process part 13 Count part 14 Judgment output part 15 Memory part 16 Function control part 17 Deviation estimation part 18 Reduced image generation part

Claims (8)

An image input unit for inputting an image;
A minimum value filter processing unit for generating a minimum value filter image of saturation from the image;
A counting unit that counts the number of pixels that are equal to or less than a predetermined saturation value from the minimum value filter image; and a ratio of the number of pixels that is equal to or less than the predetermined saturation value of the minimum value filter image to the total number of pixels of the image is a predetermined threshold value In the case described above, an image processing apparatus comprising: a determination output unit that determines that the background of the image is achromatic and has a chromatic color when the background of the image is less than a predetermined threshold.   The image processing apparatus according to claim 1, further comprising a shift estimation unit that estimates and corrects a shift of the predetermined saturation value.   The image processing apparatus according to claim 1, wherein the image processing apparatus includes a reduced image generation unit that generates a reduced image from the image.   The image processing apparatus according to claim 1, wherein the minimum value filter processing unit changes a size of a generated minimum value filter image.   The minimum value filter processing unit extracts pixels from the image at regular intervals, and performs minimum value filter processing on the pixels. Image processing apparatus. An image input step for inputting an image;
A minimum value filter processing step for generating a minimum value filter image of saturation from the image;
A counting step of counting the number of pixels below a predetermined saturation value from the minimum value filter image;
If the ratio of the number of pixels below the predetermined saturation value of the minimum value filter image occupying the total number of pixels of the image is equal to or greater than a predetermined threshold, the background of the image is achromatic, if less than the predetermined threshold, An image processing method comprising: a determination output step for determining that the background of the image is a chromatic color. An image input step for inputting an image;
A minimum value filter processing step for generating a minimum value filter image of saturation from the image;
A counting step of counting the number of pixels below a predetermined saturation value from the minimum value filter image;
If the ratio of the number of pixels below the predetermined saturation value of the minimum value filter image occupying the total number of pixels of the image is equal to or greater than a predetermined threshold, the background of the image is achromatic, if less than the predetermined threshold, An image processing program for causing a computer to execute a determination output step for determining that the background of the image is a chromatic color.   A computer-readable recording medium on which the image processing program according to claim 7 is recorded.