JP2013058966A - Image processing apparatus and image processing program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing apparatus and an image processing program capable of separating the foreground and background in an image more reliably when compared with a case not having this configuration.SOLUTION: When a determination is made in a pixel value determination unit 11 that the pixel value of a target pixel is out of a predetermined range from a first threshold, a foreground/background determination unit 13 determines whether the target pixel belongs to foreground or background by comparing the pixel value of the target pixel with the first threshold. When a determination is made in the pixel value determination unit 11 that the pixel value of a target pixel is in a predetermined range from the first threshold, a determination is made as to whether the target pixel belongs to foreground or background by extracting an average pixel value from the pixel values of pixels in a predetermined range including the target pixel, as a feature value, in a feature value extraction unit 12, and then comparing the pixel value of the target pixel with a second threshold set according to the average pixel value.

Description

  The present invention relates to an image processing apparatus and an image processing program.

  Conventionally, an image is separated into a background (background) and a foreground (characters, patterns, tables, etc.), and various processes are performed on each. For example, as a first process when analyzing and understanding a document image, it is generally known to perform region separation of the document image. As an example, in Patent Document 1, as a first stage, a document image is separated into a pattern and a character, and then binarization processing is performed on each of the pattern areas to realize character extraction.

  In the region separation of the document image, first, it is necessary to separate the foreground and the background of the document image. A process for calculating a fixed threshold value from a document image by using a characteristic that the pixel value of the background of the document image falls within a certain range, and separating the document image into the foreground and the background using the calculated fixed threshold value is widely performed. ing. As a technique for calculating the fixed threshold, there are a technique using a density histogram described in Patent Document 2 and a technique for extracting a background color by clustering described in Patent Document 3. In Patent Document 4, the foreground and the background are separated from the color image using the histogram of each color component.

  FIG. 11 is an explanatory diagram of an example of foreground / background separation processing when a fixed threshold is used. In the process using the fixed threshold value, the pixel value (for example, brightness) of each pixel of the document image is compared with the fixed threshold value, and if it is brighter than the fixed threshold value, the background is determined, and if it is dark, the foreground is determined. In FIG. 11, the fixed threshold value is indicated by a one-dot chain line. With this fixed threshold, the light character shown in FIG. 11B is correctly determined as the foreground. However, if there are noise components such as show-through and paper stains shown in FIG. 11A, these components may be determined as a foreground by a fixed threshold value. Further, for example, as in the example illustrated in FIG. 11C, when there is a region having a color lighter than a density assumed in advance, the background may be determined to be brighter than the fixed threshold. For example, if the threshold value is changed so that the noise component shown in FIG. 11A is not determined as the foreground, the light character shown in FIG. 11B is determined as the background. Conversely, even if the threshold value is changed so that the light character shown in FIG. 11C is determined as the foreground, the noise component shown in FIG. 11A is still determined as the foreground. In this way, erroneous determination occurs even if the fixed threshold is changed.

  In a document image read by an image reading device or the like, there is a possibility that these regions have pixel values that cross a fixed threshold within the region, and as a result, a part of the region is erroneously determined and erroneously extracted. Such an influence of the erroneously extracted region also affects subsequent processing. A misjudgment may occur in common with foreground and background separation techniques using a fixed threshold regardless of the method of calculating the fixed threshold.

  On the other hand, as a technique for binarizing an image, several techniques have been proposed in addition to a method using a single fixed threshold value. For example, in Patent Document 5, after correcting the pixel value of the target pixel according to the average value of the pixels in a predetermined range surrounding the target pixel and the pixel value of the target pixel, the threshold value is set and binarized. Further, in Patent Document 6, depending on whether or not a target pixel is an edge pixel, a first threshold based on a threshold value of an adjacent pixel and a density average value of pixels surrounding the target pixel, and a density of a pixel surrounding the target pixel The second threshold based on the value is switched and used for binarization. Further, in Patent Document 7, if the difference between the average value of the density of 3 × 3 pixels including the target pixel and the density of the target pixel is within a certain range, it is binarized by a fixed threshold value and exceeds the range having the density difference. In this case, binarization is performed by the average value of the density of 3 × 3 pixels.

  Each of these binarization techniques is a technique for switching the threshold according to the situation of pixels within a certain range from the target pixel. For this reason, for example, when a character is present on a pattern present in the image, binarization is performed with the pattern portion as the background and the character as the foreground. Therefore, the background (background) and other foregrounds (characters, pictures, tables, etc.) may not be separated.

Japanese Patent Laid-Open No. 2005-184402 JP-A-5-63968 JP 2001-222717 A JP 2010-187270 A JP 2001-298615 A JP 2005-311992 A JP-A-6-276392

  An object of the present invention is to provide an image processing apparatus and an image processing program capable of reliably separating a foreground and a background in an image as compared with a case where the present configuration is not provided.

  The invention according to claim 1 of the present application includes a pixel value determination unit that determines whether or not a pixel value of a target pixel is within a predetermined range from a predetermined first threshold value, and includes a target pixel in advance. A feature amount extracting unit that extracts a feature amount from a predetermined range of pixels, a pixel value of a target pixel, the first threshold value, a determination result by the pixel value determining unit, and a feature amount extracted by the feature amount extracting unit. An image processing apparatus comprising foreground / background determination means for determining whether a pixel of interest belongs to a foreground or a background based on the background.

  The invention according to claim 2 of the present application is such that the foreground / background determination means in the invention according to claim 1 is within a range in which the pixel value of the pixel of interest is determined in advance by the pixel value determination means from the first threshold value. And determining whether the pixel of interest belongs to the foreground or the background according to the pixel value of the pixel of interest and the feature amount extracted by the feature amount extraction unit. .

  In the invention according to claim 3 of the present application, the feature amount extraction means in the invention of claim 2 extracts an average pixel value as a feature amount from pixel values of pixels in a predetermined range including the target pixel. The foreground / background determination unit sets a second threshold value according to the average pixel value extracted by the feature amount extraction unit, compares the second threshold value with the pixel value of the target pixel, and determines that the target pixel is the foreground. It is an image processing apparatus characterized by determining whether it belongs to the background or the background.

  According to a fourth aspect of the present invention, the foreground / background determination unit according to any one of the first to third aspects of the present invention is configured such that the pixel value of the target pixel is the first pixel value determination unit. A pixel value of a target pixel is compared with the first threshold value to determine whether the target pixel belongs to the foreground or the background when it is determined that the target pixel is out of a predetermined range from the threshold value The image processing apparatus.

  In the invention according to claim 5 of the present application, regarding the luminance or brightness of the target pixel, whether the target pixel belongs to the foreground or the background by the configuration of the image processing device according to any one of claims 1 to 4. 5. The first determination means for determining the color component other than the luminance or brightness component of the target pixel for the target pixel determined as the background by the first determination means. According to the configuration of the image processing apparatus, the image processing apparatus includes a second determination unit that determines whether the target pixel belongs to the foreground or the background.

  In the invention according to claim 6 of the present application, the pixel of interest is brought to the foreground by the configuration of the image processing device according to any one of claims 1 to 4 for each of a plurality of color components constituting the value of the pixel of interest. A plurality of color component determination means for determining whether the target pixel belongs to the background, and a comprehensive determination means for determining whether the pixel of interest belongs to the foreground or the background based on the determination results of the plurality of color component determination means An image processing apparatus characterized by this.

  The invention described in claim 7 of the present application is an image processing program for causing a computer to execute the function of the image processing apparatus described in any one of claims 1 to 6.

  According to the first aspect of the present invention, there is an effect that the foreground and the background can be reliably determined as compared with the case where only the fixed threshold is used.

  According to the second aspect of the present invention, foreground and background can be more reliably determined when there is a possibility of erroneous determination at the first threshold than when the present configuration is not provided. .

  According to the third aspect of the present invention, noise components such as show-through and paper stains and foreground components such as light characters can be reliably determined as compared with the case where the present configuration is not provided.

  According to the invention described in claim 4 of the present application, compared with the case where the present configuration is not provided, the areas that are clearly determined to be the foreground and the background are excluded from the determination using the feature amount, thereby speeding up the processing. Can do.

  According to the fifth aspect of the present invention, the foreground and the background can be reliably determined as compared with the case where only the fixed threshold is used for the color image.

  According to the sixth aspect of the present invention, it is possible to reliably determine the foreground and the background as compared with the case where only the fixed threshold is used for the color image.

  According to the invention of claim 7 of the present application, the effect of the invention of any one of claims 1 to 6 can be obtained.

It is a block diagram which shows one Embodiment of this invention. It is a flowchart which shows an example of the operation | movement in one Embodiment of this invention. It is explanatory drawing of an example of the foreground / background separation process in the embodiment of the present invention. It is explanatory drawing of an example of the determination process of the foreground and the background with respect to the specific image in one Embodiment of this invention. It is a block diagram which shows the modification of one Embodiment of this invention. It is a flowchart which shows an example of the operation | movement in the modification of one Embodiment of this invention. It is a block diagram of the 1st application example of one Embodiment of this invention. It is a flowchart which shows an example of operation | movement of the 2nd determination part 32 in the 1st application example of one Embodiment of this invention. It is a block diagram of the 2nd application example of one Embodiment of this invention. FIG. 15 is an explanatory diagram of an example of a computer program, a storage medium storing the computer program, and a computer when the functions described in the embodiment of the present invention or its modifications and application examples are realized by a computer program. It is explanatory drawing of an example of a foreground and background separation process at the time of using a fixed threshold value.

  FIG. 1 is a configuration diagram showing an embodiment of the present invention. In the figure, 11 is a pixel value determination unit, 12 is a feature amount extraction unit, and 13 is a foreground / background determination unit. Here, it is assumed that an image to be processed is given, and each pixel of the image is set as a target pixel, and processing for determining whether the target pixel is a foreground or a background is performed. Of course, the foreground or the background may be determined using a plurality of pixels as a target pixel.

  The first threshold value is a fixed threshold value conventionally used for discriminating between the foreground and the background, and is given in advance. For example, it is only necessary to set a threshold value at which an assumed light character color is determined as the foreground. For example, the first threshold value may be generated from a density histogram using an image to be processed. For example, the techniques of Patent Document 1 and Patent Document 2 may be used. Of course, the first threshold may be set by other various techniques, or may be set manually. When a document image is a target, it is assumed that a certain background area exists, and foreground and background are basically determined based on a first threshold which is a fixed threshold. When assumed, the determination using the feature amount is performed.

  The pixel value determination unit 11 determines whether or not the pixel value of the target pixel is within a predetermined range from the given first threshold value. This range is set as a range that may be erroneously determined when the foreground and background are determined using the first threshold that is a fixed threshold. The range may be set in advance. For example, the range may be set as a relative range from the first threshold.

  The feature amount extraction unit 12 extracts a feature amount from pixels in a predetermined range including the target pixel. As the feature quantity to be extracted, an average pixel value from the pixel values of pixels in a predetermined range including the target pixel may be used as the feature quantity. Of course, in addition to this, for example, other feature amounts such as a first-order differentiation and a second-order differentiation that can obtain the difference between the surrounding pixels and the target pixel may be used.

  The foreground / background determination unit 13 includes the target pixel in the foreground based on the pixel value of the target pixel, the first threshold value, the determination result by the pixel value determination unit 11, and the feature amount extracted by the feature amount extraction unit 12. Or belonging to the background. As a specific example, when the pixel value determining unit 11 determines that the pixel value of the target pixel is outside the predetermined range from the first threshold, the pixel value of the target pixel and the first threshold are set. In comparison, it is determined whether the pixel of interest belongs to the foreground or the background. When the pixel value determination unit 11 determines that the pixel value of the target pixel is within a predetermined range from the first threshold, the pixel value of the target pixel and the feature amount extracted by the feature amount extraction unit 12 Thus, it is determined whether the target pixel belongs to the foreground or the background. When the feature amount extracted by the feature amount extraction unit 12 is an average pixel value, a second threshold value is set according to the average pixel value, and the second threshold value is compared with the pixel value of the target pixel. It may be determined whether the pixel of interest belongs to the foreground or the background. Even when the feature quantity extracted by the feature quantity extraction unit 12 is a feature quantity other than that, the determination according to the feature quantity may be performed.

  FIG. 2 is a flowchart showing an example of the operation in the embodiment of the present invention. In S21, the pixel of interest in the image is set, the pixel value is acquired and passed to each unit. In S22, the feature amount extraction unit 12 extracts feature amounts from pixels in a predetermined range including the target pixel. Here, an average pixel value is extracted as a feature amount. The process of S22 may be performed in parallel with the process of the next S23 or after the process of S23.

  In S <b> 23, the pixel value determination unit 11 determines whether or not the pixel value of the pixel of interest is within a predetermined range from the given first threshold value. If it is outside the predetermined range, it is assumed that there is no problem even if the foreground and the background are determined using the first threshold which is a fixed threshold. In S24, the foreground / background determination unit 13 determines the target pixel. The pixel value is compared with the first threshold value to determine whether the target pixel belongs to the foreground or the background. For example, a pixel darker than the first threshold may be determined as the foreground, and a pixel brighter than the first threshold may be determined as the background.

  On the other hand, if it is determined in S23 that the pixel value of the pixel of interest is within a predetermined range from the first threshold, the foreground and background determination using the first threshold that is a fixed threshold is incorrect. In step S25, the foreground / background determination unit 13 sets a second threshold value in accordance with the average pixel value extracted by the feature amount extraction unit 12, and the second threshold value and the pixel value of the target pixel are determined. To determine whether the target pixel belongs to the foreground or the background. The second threshold value may be an average pixel value itself or a value obtained by adding or subtracting a predetermined value to the average pixel value in consideration of a change amount due to a noise component or the like. When the pixel value of the target pixel is within a predetermined range from the first threshold, if the target pixel has different characteristics from the surrounding pixels, it is determined as the foreground. If it has unchanging characteristics, it is determined as a background. Specifically, if the pixel is darker than the second threshold, the target pixel is determined to be the foreground as having a different characteristic from the surroundings, and if the pixel is brighter than the second threshold, the target The pixel may be determined as the background as having the same characteristics as the surroundings.

  In S26, it is determined whether or not each pixel of the given image has been processed as a target pixel. If any unprocessed pixels remain, the process returns to S21 to change the target pixel and repeat the process. If there are no unprocessed pixels, the process is terminated.

  FIG. 3 is an explanatory diagram showing an example of foreground / background separation processing according to the embodiment of the present invention. In FIG. 3, the change of the pixel value in an image is shown by the curve. Further, the range of ± α from the first threshold indicated by the alternate long and short dash line is set as a predetermined range, and the inside / outside determination of this range is performed by the pixel value determination unit 11.

  FIG. 3A shows the background portion. In this case, since it is outside the range of ± α from the first threshold value, the foreground and background are determined by comparison with the first threshold value. In this example, it is brighter than the first threshold value, so it is determined as the background.

  FIG. 3B shows a dark character portion, which is also outside the range of ± α from the first threshold value, and therefore the foreground and background are determined by comparison with the first threshold value. In this example, it is darker than the first threshold, so it is determined as the foreground.

  3C, 3D, and 3E show cases where the pixel value of the target pixel is within a range of ± α from the first threshold value. In this case, if the foreground and background are determined using the first threshold that is a fixed threshold, the foreground may be erroneously determined. In such a case, if the pixel of interest has a characteristic different from the surrounding pixels, it is determined as the foreground, and if it has a characteristic that is not different from the surrounding pixels, it is determined as the background. Here, an average pixel value is obtained as a feature amount from pixel values of pixels in a predetermined range including the target pixel, that is, pixels including the periphery, and the average pixel value is shifted to the dark side (to the average pixel value). As a second threshold value (by adding a predetermined value), the second threshold value is used to determine whether the foreground or the background. This second threshold value is indicated by a broken line.

  FIG. 3C shows a portion where noise components such as show-through and paper stains exist, and is within a range of ± α from the first threshold. Therefore, the foreground and the background are determined by comparing the target pixel with the second threshold value. Noise components such as show-through and paper stains are often blurred and spread, and the difference in characteristics between the pixel of interest and surrounding pixels is not as clear as in the case of characters. Therefore, the pixel value of the target pixel becomes a value similar to the average pixel value, and is determined to be brighter than the second threshold value obtained by darkening the average pixel value, that is, the background.

  FIG. 3D shows a case of a light character, which is within a range of ± α from the first threshold value, and therefore, the foreground and the background are determined by comparing the target pixel with the second threshold value. . In light characters, the character line is lighter than the surroundings, but the color difference is clear and distinctive. Therefore, the pixel value of the target pixel is determined to be darker than the second threshold value obtained from the average pixel value, and is determined to be the foreground.

  In the example shown in FIG. 3E, the case of a lighter character than the example shown in FIG. In this example, in the foreground / background determination using the first threshold that is a fixed threshold, the pixel value of the target pixel is brighter than the first threshold, so that the background is determined. However, in this example, since it is within the range of ± α from the first threshold value, the pixel of interest is compared with the second threshold value instead of comparing with the first threshold value. Make a decision. Also in the example shown in FIG. 3E, the pixel value of the target pixel is determined to be darker than the second threshold value obtained from the average pixel value, and it is determined to be the foreground.

  As in this example, in this embodiment, a portion where there is a noise component that is not determined to be the background by the fixed threshold shown in FIG. 11 is determined as the background, and light characters that are not determined to be the foreground by the fixed threshold are also included. It is determined as the foreground.

  FIG. 4 is an explanatory diagram showing an example of foreground / background determination processing for a specific image according to the embodiment of the present invention. In the specific example of the image shown in FIG. 4A, a title character is drawn in a colored rectangle in the area a. In addition, a pie chart is drawn in the area b, a table is drawn in the area c, a photograph is drawn in the area d, and light characters are drawn in the area e. Furthermore, there are three noise regions p in this image. The foreground / background determination result is shown in FIG. 4B, and the black area is the portion determined to be the foreground.

  For area a, the color of the rectangle is outside the predetermined range from the first threshold, and the title character drawn in the rectangle is also outside the predetermined range from the first threshold. . Therefore, the region a is determined as the foreground as a whole. Note that if a determination is made using the second threshold value obtained from the average pixel value used in the conventional binarization technique without using the first threshold value, a portion determined to be the background is generated in the rectangle. Different judgment results are obtained.

  The region b and the region d are assumed to be outside the predetermined range from the first threshold as a whole, and are determined to be the foreground. For the area c, the pixels such as the ruled line of the table and the line segment of the character are determined as the foreground as they are, and are determined as shown in FIG. In this case, the entire area c is not determined as the foreground.

  For a light character drawn in the area e, for example, if the density is outside the range predetermined from the first threshold, it is determined as the foreground. Even if it is within a predetermined range from the first threshold, the foreground is determined by determining using the second threshold obtained according to the average pixel value of surrounding pixels.

  As for the noise region p, if it is a light noise component that is outside the predetermined range from the first threshold, it is determined as the background. Even if it is within a predetermined range from the first threshold value, the pixels of the noise component are spread out in the surroundings, and the background is determined by using the second threshold value obtained according to the average pixel value. It is determined.

  In this way, in the case of the image shown in FIG. 4A, the determination shown in white (background) and black (foreground) in FIG. 4B is performed.

  FIG. 5 is a configuration diagram showing a modification of the embodiment of the present invention. In this modification, the determination result of the pixel value determination unit 11 is also passed to the feature amount calculation unit 12. The feature amount calculation unit 12 extracts a feature amount from pixels in a predetermined range including the target pixel. This process is performed by the pixel value determination unit 11 in which the pixel value of the target pixel is determined from the first threshold value. This is performed when it is determined to be within the range. Others are as described above. In this configuration, the feature amount extraction processing is reduced as compared with the case where the feature amount calculation unit 12 extracts feature amounts for all the target pixels.

  FIG. 6 is a flowchart showing an example of the operation in the modification of the embodiment of the present invention. Compared with FIG. 2, the process of S22 is different in that the process moves to a process in the case where it is determined in S23 that the pixel value of the target pixel is within a predetermined range from the first threshold.

  In S21, a target pixel in the image is set, and in S23, the pixel value determination unit 11 determines whether or not the pixel value of the target pixel is within a predetermined range from a given first threshold value. . If it is outside the predetermined range, in S24, the foreground / background determination unit 13 compares the pixel value of the target pixel with the first threshold value to determine whether the target pixel belongs to the foreground or the background. . The flow of this process is as described in FIG.

  On the other hand, when it is determined in S23 that the pixel value of the target pixel is within a predetermined range from the first threshold, in S22, the feature amount extraction unit 12 determines a predetermined value including the target pixel. A feature amount is extracted from a range of pixels. For example, an average pixel value of pixels in a predetermined range including the target pixel is extracted as a feature amount. In step S25, the foreground / background determination unit 13 sets a second threshold value according to the average pixel value extracted by the feature amount extraction unit 12, compares the second threshold value with the pixel value of the target pixel, and determines that the target pixel is the foreground. Or belonging to the background. The contents of the processing performed in S22 and S25 are as described with reference to FIG.

  In S26, it is determined whether or not each pixel of the given image has been processed as a target pixel. If any unprocessed pixels remain, the process returns to S21 to change the target pixel and repeat the process. If there are no unprocessed pixels, the process is terminated.

  In the case of a color image, each pixel value includes a plurality of color component values. For example, in the case of a YCbCr color space, a pixel value is composed of a Y component, a Cb component, and a Cr component. For example, in the LAB color space, a pixel value is composed of an L * component, an a * component, and a b * component. In other color spaces, pixel values are constituted by a plurality of components. For a color image, for example, the above-described determination may be performed on the luminance or lightness component. Further, the determination may be made using components other than the luminance or lightness components, and an example thereof is shown as an application example.

  FIG. 7 is a configuration diagram of a first application example of one embodiment of the present invention. In the figure, 31 is a first determination unit, and 32 is a second determination unit. Here, as an example, the color space of the image is assumed to be a YCbCr color space, and the pixel value is assumed to be composed of a Y component, a Cb component, and a Cr component. This first application example shows an example in which one embodiment of the present invention or a modification thereof is applied to the configuration described in Patent Document 4.

  The first determination unit 31 uses the Y component, which is a luminance component of the pixel, to determine the foreground and the background by the method described in the above embodiment or its modification.

The second determination unit 32 uses the Cb component and the Cr component, which are color difference components of the pixels, to further determine the foreground and the background for the pixels determined to be the background by the first determination unit 31. This determination is basically performed by the method described in the above-described embodiment or its modification. At this time, the pixel value determination unit 11 may determine whether each of the Cb component and the Cr component is within a predetermined range from the given first threshold value. Alternatively, a color difference such as √ (Cb ² + Cr ² ) is calculated from the Cb component and the Cr component, and it is determined whether or not the color difference is within a predetermined range from the given first threshold value. Also good.

  FIG. 8 is a flowchart illustrating an example of the operation of the second determination unit 32 in the first application example of the embodiment of the invention. In S41, the pixel of interest in the image is set, the pixel value is acquired and passed to each unit.

  In S42, it is determined whether or not the Cb component of the target pixel is within a predetermined range from the first threshold corresponding to the Cb component. In S43, it is determined whether or not the Cr component of the target pixel is within a predetermined range from the first threshold corresponding to the Cr component. Note that this determination may be made first.

  When it is determined in S42 that the Cb component is outside the predetermined range from the corresponding first threshold value, or in S43, it is determined that the Cr component is outside the predetermined range from the corresponding first threshold value. In this case, it is assumed that there is no problem even if the foreground and the background are determined using the first threshold that is the fixed threshold, and the foreground / background determination unit 13 converts the Cb component and the Cb component of the target pixel in S44. The corresponding first threshold value, the Cr component and the first threshold value corresponding to the Cr component are compared to determine whether the pixel of interest belongs to the foreground or the background. For example, if the Cb component is larger than the first threshold value corresponding to the Cb component, it is determined as the foreground regardless of the Cr component, and the Cr component is larger than the first threshold value corresponding to the Cr component. For example, the foreground may be determined regardless of the Cb component. Moreover, what is necessary is just to determine as a background, when Cb component is smaller than the 1st threshold value corresponding to Cb component, and Cr component is smaller than the 1st threshold value corresponding to Cr component.

  On the other hand, in S42, it is determined that the Cb component of the pixel of interest is within a predetermined range from the first threshold corresponding to the Cb component, and the Cr component of the pixel of interest corresponds to the Cr component in S43. If it is determined that it is within a predetermined range from the threshold value, it may be erroneously determined in the foreground and background determination using the first threshold value which is a fixed threshold value. A feature amount is extracted from pixels in a predetermined range including. This process corresponds to the process of the feature amount extraction unit 12 described above. For example, the average Cb component and Cr component values may be extracted as feature amounts. Of course, other feature amounts may be extracted. According to the flow of the process described with reference to FIG.

  In S46, the second threshold value of each component is set according to the feature values extracted in S45, for example, the average Cb component value and Cr component value, and the second threshold value corresponding to the Cb component and Cb component, Cr component and Cr component is set. The second threshold value corresponding to the component is compared to determine whether the target pixel belongs to the foreground or the background. This process corresponds to the process of the foreground / background determination unit 13 described above.

  In S47, it is determined whether or not each pixel of the given image has been processed as a target pixel. If any unprocessed pixels remain, the process returns to S41 to change the target pixel and repeat the process. If there are no unprocessed pixels, the process is terminated.

  In the first application example, the first determination unit 31 determines the foreground and the background using the Y component that is a luminance component. Even in a foreground region having a different saturation from the background, the background may be determined as the background, and the second determination unit 32 determines using the color difference component, and in such a case also separates the background and the foreground. doing. The first application example is not limited to the YCbCr color space, and may be a case where various color spaces including lightness components such as a LAB color space are used. For example, in the RGB color space, the CMY color space, the CMYK color space, etc., the luminance or brightness component is calculated, the foreground and background are determined by the first determination unit 31, and the second determination is performed for the pixels determined as the background. The unit 32 may determine the foreground and the background using each color component.

  FIG. 9 is a configuration diagram of a second application example of the embodiment of the invention. In the figure, 51 is an R determination section, 52 is a G determination section, 53 is a B determination section, and 54 is a comprehensive determination section. This second application example shows a case where determination is performed for each color component, and as an example, a configuration in the case of an RGB color space is shown.

  The R determination unit 51 uses the R component of the pixel to determine the foreground and the background by the method described in the above embodiment or its modification. The G determination unit 52 determines the foreground and the background using the G component of the pixel by the method described in the above-described embodiment or its modification. Further, the B determination unit 53 uses the B component of the pixel to determine the foreground and the background by the method described in the above embodiment or its modification. The R determination unit 51, the G determination unit 52, and the B determination unit 53 use different color components, but the process determines the foreground and the background by the method described in the above-described embodiment or its modification. .

  The overall determination unit 54 determines whether it belongs to the foreground or the background by combining the determination results of the R determination unit 51, the G determination unit 52, and the B determination unit 53. As a specific example, if any of the R determination unit 51, the G determination unit 52, and the B determination unit 53 determines the foreground, the foreground is used, and the R determination unit 51, the G determination unit 52, and the B determination unit 53 all have the background. If it is determined, the background may be used.

  In this example, the case of the RGB color space is shown, but not limited to this, the CMY color space, the CMYK color space, the YCbCr color space and the LAB color space used in the first application example, and various other colors. In the space, a determination for each color component and a determination that combines the determination results may be performed.

  FIG. 10 is an explanatory diagram of an example of a computer program, a storage medium storing the computer program, and an example of a computer when the functions described in the embodiment of the present invention or its modifications and application examples are realized by a computer program. . In the figure, 61 is a program, 62 is a computer, 71 is a magneto-optical disk, 72 is an optical disk, 73 is a magnetic disk, 74 is a memory, 81 is a CPU, 82 is an internal memory, 83 is a reading unit, 84 is a hard disk, and 85 is An interface 86 is a communication unit.

  You may implement | achieve the function of each part demonstrated by one Embodiment of the above-mentioned embodiment of this invention or its modification example, and an application example entirely or partially with the program 61 which makes a computer perform. In that case, the program 61 and data used by the program may be stored in a storage medium read by the computer. A storage medium causes a state of change in energy such as magnetism, light, electricity, etc. to the reading unit 83 provided in the hardware resource of a computer according to the description content of the program, and a signal corresponding thereto. In this format, the description content of the program is transmitted to the reading unit 83. For example, a magneto-optical disk 71, an optical disk 72 (including CD, DVD, etc.), a magnetic disk 73, a memory 74 (including IC card, memory card, flash memory, etc.) and the like. Of course, these storage media are not limited to portable types.

  The program 61 is stored in these storage media. For example, the program 61 is read from the computer by mounting these storage media in the reading unit 83 or the interface 85 of the computer 62, and the internal memory 82 or the hard disk 84 (magnetic disk or silicon). And the functions described in the above-described embodiment of the present invention, its modified examples, and applied examples are realized in whole or in part. Alternatively, the program 61 is transferred to the computer 62 via the communication path, and the computer 62 receives the program 61 by the communication unit 86 and stores it in the internal memory 82 or the hard disk 84, and the program 81 is executed by the CPU 81. May be.

  In addition, the computer 62 may be connected to various devices via an interface 85. Of course, it may be partially configured by hardware, or all may be configured by hardware. Or you may comprise as a program which included all or one part of the function demonstrated by one Embodiment of this invention with the other structure. For example, the program may be integrated with another program that uses the foreground and background determination results described in the embodiment of the present invention or its modifications and application examples. In the case of applying to other purposes, the program may be integrated with the program for that purpose.

  DESCRIPTION OF SYMBOLS 11 ... Pixel value determination part, 12 ... Feature-value extraction part, 13 ... Foreground / background determination part, 31 ... 1st determination part, 32 ... 2nd determination part, 51 ... R determination part, 52 ... G determination part, 53 ... B Determining unit, 54 ... Comprehensive determining unit, 61 ... Program, 62 ... Computer, 71 ... Magneto-optical disk, 72 ... Optical disk, 73 ... Magnetic disk, 74 ... Memory, 81 ... CPU, 82 ... Internal memory, 83 ... Reading part, 84: hard disk, 85: interface, 86: communication unit.

Claims (7)

  A pixel value determination unit that determines whether or not the pixel value of the target pixel is within a predetermined range from a predetermined first threshold, and a feature amount from a pixel in a predetermined range including the target pixel Feature amount extraction means to extract, pixel value of target pixel, first threshold, determination result by pixel value determination means, and feature amount extracted by feature amount extraction means, whether target pixel belongs to foreground or background An image processing apparatus comprising foreground / background determination means for determining whether the image belongs to   The foreground / background determination unit is configured to extract the pixel value of the target pixel and the feature amount when the pixel value determination unit determines that the pixel value of the target pixel is within a predetermined range from the first threshold value. The image processing apparatus according to claim 1, wherein it is determined whether the target pixel belongs to the foreground or the background according to the feature amount extracted by the means.   The feature amount extraction unit extracts an average pixel value as a feature amount from pixel values of pixels in a predetermined range including the target pixel, and the foreground / background determination unit extracts the average value extracted by the feature amount extraction unit. The second threshold value is set in accordance with the pixel value, and the second threshold value is compared with the pixel value of the target pixel to determine whether the target pixel belongs to the foreground or the background. The image processing apparatus described.   The foreground / background determination unit determines the pixel value of the target pixel and the first value when the pixel value determination unit determines that the pixel value of the target pixel is out of a predetermined range from the first threshold. 4. The image processing apparatus according to claim 1, wherein the threshold value is compared to determine whether the pixel of interest belongs to the foreground or the background. 5.   A first determination unit that determines whether the target pixel belongs to the foreground or the background according to the configuration of the image processing device according to any one of claims 1 to 4 with respect to luminance or brightness of the target pixel; 5. The image processing apparatus according to claim 1, wherein color components other than luminance or brightness components of the target pixel are determined for the target pixel determined to be the background by the first determination unit. An image processing apparatus comprising: a second determination unit configured to determine whether a target pixel belongs to a foreground or a background.   5. A plurality of color components constituting a value of a target pixel, each of which determines whether the target pixel belongs to the foreground or the background by the configuration of the image processing apparatus according to any one of claims 1 to 4. An image processing apparatus comprising: color component determination means; and comprehensive determination means for determining whether the target pixel belongs to a foreground or a background based on determination results of the plurality of color component determination means.   An image processing program for causing a computer to execute the function of the image processing apparatus according to any one of claims 1 to 6.

Images