JP2008059406A - Image processing apparatus and image processing method - Google Patents

Abstract

An image processing apparatus for removing noise from an image signal is obtained.
An edge extraction unit that extracts edge information about the presence / absence of an edge around a pixel of interest in an input image signal, and whether the color signal of the pixel of interest is within a specific color region or not The noise removal selection unit that selects a noise removal process to be applied to the target pixel using the determination result and the edge information extracted by the edge extraction unit, and the noise removal process selected by the noise removal selection unit And a noise removal processing unit applied to the target pixel.
[Selection] Figure 1

Description

The present invention relates to an image processing apparatus and an image processing method for removing noise from an image signal.

Conventionally, as an image processing method for removing noise from an image signal, Patent Document 1 discloses an example in which noise is removed by filter processing that applies different coefficients to colors in a specific color region and outside a specific color region of an image signal. It is shown.
Patent Document 2 discloses an example in which noise is removed by detecting a specific color region of an image signal based on scene information.

Japanese Patent Laying-Open No. 2004-172987 (page 11, FIG. 15) Japanese Patent Laid-Open No. 2002-101422 (page 5, FIG. 2)

  In the noise removal filter processing, surrounding pixel values are reflected. In the conventional image processing apparatus and image processing method, only whether or not the pixel has a color signal value of a specific color region is used as a criterion for selecting a filter coefficient for noise removal. For this reason, in the portion where the color signal value of the peripheral pixel of the pixel changes greatly, even if it is not desired to influence the peripheral pixel having a color signal value significantly different from the pixel to which the filter is applied, the filter coefficient is Since the determination is based only on the color signal of the pixel, filter processing that is greatly affected by the value of the color signal of the surrounding pixel may be performed, and the image near the boundary between the specific color area and the outside of the specific color area may be blurred. There was a problem that there was.

  The present invention has been made to solve the above-described problems, and appropriate noise removal is performed in consideration of edge information corresponding to the magnitude of the change in the value of the color signal of the pixel and the surrounding pixels. It is an object of the present invention to realize noise removal that suppresses blurring of an image near the boundary between a specific color region and the outside of the specific color region.

  The image processing apparatus and the image processing method according to the present invention include an edge extraction unit that extracts edge information about the presence or absence of an edge around a target pixel of an input image signal, and a color signal of the target pixel based on the specific color region information. A noise removal selection unit that determines whether the image is within the specific color region or outside the specific color region, and uses the determination result and edge information extracted by the edge extraction unit to select a noise removal process to be applied to the target pixel; and a noise removal selection And a noise removal processing unit that applies the noise removal processing selected by the unit to the target pixel.

  According to the present invention, since the appropriate noise removal is performed in consideration of the edge information corresponding to the magnitude of the change in the color signal value of the pixel and the surrounding pixels, the specific color area and the specific color area Noise removal that suppresses blurring of the image near the boundary with the outside can be realized.

  Embodiments of an image processing apparatus and an image processing method according to the present invention will be described. In the following embodiments, edge processing is used to remove noise from an image signal, thereby suppressing blurring of the image, storing the image without blurring the edge, and performing image processing that effectively removes noise from the image signal. Do.

Embodiment 1 FIG.
In Embodiment 1 of the present invention, when a color signal of a noise removal target pixel (hereinafter referred to as a target pixel) of an input image signal is included in a specific color region, a specific color having a different intensity based on edge information An image processing apparatus 10a that switches between noise removal and normal noise removal will be described.

  FIG. 1 is a block diagram showing an example of the configuration of an image processing apparatus 10a according to Embodiment 1 of the present invention. The outline of the function of each block will be described based on this block diagram.

  In FIG. 1, the edge extraction unit 11 a extracts edge information related to the presence or absence of an edge within the edge extraction range for the target pixel of the input image signal. The noise removal selection unit 12a determines whether the color signal of the pixel of interest is within the specific color region or outside the specific color region based on the specific color region information, and uses the edge information extracted by the edge extraction unit to determine the pixel of interest. Select the noise removal process to apply. The noise removal processing unit 13a applies the noise removal processing selected by the noise removal selection unit to the target pixel.

  FIG. 2 is a block diagram illustrating an example of the configuration of the noise removal selection unit 12a of the image processing apparatus 10a according to Embodiment 1 of the present invention. The noise removal selection unit 12a includes the specific color region determination unit 21a and the noise removal selection unit 12a. The noise removal determination means 22 is comprised.

  In FIG. 2, the specific color area determination unit 21a determines whether the color signal of the target pixel of the input image signal is within the specific color area or outside the specific color area based on the specific color area information. The noise removal determination unit 22 removes the specific color noise applied to the target pixel whose color signal is in the specific color region and the target pixel outside the specific color region with respect to the target pixel determined by the specific color region determination unit 21a. Any one of the normal noise removal to be applied to is selected and used as determination information. However, the noise removal determination unit 22 uses the edge information extracted by the edge extraction unit 11a together with the determination whether the target pixel output from the specific color region determination unit 21a is within the specific color region or outside the specific color region. Select whether to remove color noise or normal noise. The noise removal processing unit 13a applies noise removal processing of any one of specific color noise removal and normal noise removal from the determination information selected by the noise removal selection unit 12a to remove noise from the target pixel. .

  Next, the operation of each unit of the image processing apparatus 10a will be described with reference to the flowchart of FIG. Here, it is assumed that the input image signal and the output image signal are composed of luminance Y, color difference Cb, and Cr color signals (color space components).

  First, the edge extraction unit 11a in FIG. 1 receives an input image signal, and within a set range (referred to as an edge extraction range) including peripheral pixels of horizontal m pixels, vertical n pixels, and a total of m × n pixels around the pixel of interest. The presence or absence of an edge is extracted as edge information (step ST101). That is, the edge information is information indicating the presence / absence of an edge around the pixel of interest (edge extraction range). Here, m and n are integers, and 0 <m <horizontal pixel number and 0 <n <vertical pixel number are satisfied with respect to the image size. Note that the shape of the edge extraction range may not be a rectangle.

  As an example of a method for obtaining edge information, a method for extracting an edge based on a distribution of luminance Y values of an input image signal will be described. The edge extraction unit 11a obtains a histogram of the signal distribution of the Y value within the edge extraction range in the peripheral region of the target pixel, and is distributed in a biased manner where the Y value is large (closest to the maximum value) and small (close to the minimum value). If the intermediate value is small and bipolar, it is determined that an edge exists in the peripheral area of the target pixel (the probability of including an edge is high), and in other cases, the edge does not exist (the probability of including an edge is low). ) To obtain edge information.

  As a more specific example, the first threshold value T1 and the second threshold value T2 are used as internal dividing points of respective predetermined ratios calculated from the maximum value and the minimum value of the Y value signal distribution in the peripheral pixels within the edge extraction range. By setting (<T1), the number of pixels equal to or larger than the first threshold T1 where the Y value is large is equal to or larger than the predetermined number and the number of pixels equal to or smaller than the second threshold T2 where the Y value is small. Is greater than or equal to the predetermined number, it is determined that an edge exists in the peripheral region of the target pixel. The two predetermined numbers do not have to be the same value, and can be set as a predetermined number of pixels, respectively, if the total number of pixels is determined as a rectangular area having an edge extraction range of m × n pixels.

  Similarly, the edge information may be obtained based on the distribution of the color difference Cb value and the Cr value of the input image signal. In this case, the edge extraction unit 11a may obtain the signal distribution of the Cb value and the Cr value within the edge extraction range of the peripheral region of the target pixel, determine whether the edge exists or not, and may use the edge information. Further, the edge information of the Y value and the edge information of the Cb value and the Cr value may be weighted to comprehensively determine whether or not an edge exists in the peripheral region of the target pixel, and may be used as the edge information. In short, the edge information is information indicating whether or not a pixel having a large change in the color signal exists, and may be obtained from any color signal.

  Next, the specific color region determination unit 21a of the noise removal selection unit 12a in FIG. 1 determines whether the color signal of the target pixel of the input image signal is within the specific color region based on the specific color region information or outside the specific color region. It is determined whether or not there is (step ST102). Here, the specific color area is an area of a specific color signal that falls within a range having an upper limit value and a lower limit value respectively set as Y value, Cb value, and Cr value as specific color area information. It may be set in advance whether the color signal on each boundary value is handled as inside or outside the specific color region.

  When the specific color area determination unit 21a determines that the color signal of the target pixel is outside the specific color area (No in step ST102), the noise removal determination unit 22 of the noise removal selection unit 12a For a normal pixel of interest outside, selection of normal noise removal is used as determination information of the noise removal selection unit 12a, and as a result, the process proceeds to step ST104.

  On the other hand, when the specific color region determination unit 21a determines that the color signal of the target pixel is within the specific color region (Yes in step ST102), the noise removal determination unit 22 of the noise removal selection unit 12a performs the edge extraction unit. With reference to the edge information input from 11a, it is determined whether or not an edge exists in the peripheral region of the target pixel (step ST103).

  Here, when the noise removal determination unit 22 determines that an edge exists in the peripheral region of the target pixel (Yes in step ST103), even if the color signal is within the specific color region, it is the same as outside the specific color region. The normal noise removal selection is set as the determination information of the noise removal selection unit 12a, and the process proceeds to step ST104 as a result. On the other hand, if it is determined that there is no edge in the peripheral region of the target pixel (No in step ST103), the selection of specific color noise removal is set as the determination information of the noise removal selection unit 12a, and the process proceeds to step ST105 as a result. .

  In the noise removal selection unit 12a, the specific color region determination unit 21a performs the specific color region determination in step ST102, and the noise removal determination unit 22 further adds a pixel around the target pixel in step ST103 for the target pixel in the specific color region. By determining whether or not there is an edge, even if it is within a specific color area, if there is an edge around the pixel of interest, normal noise processing is applied that can be stored with reduced blurring of the edge, and the edge is present. When not, it is possible to apply specific color noise removal, which has a large noise removal effect.

  And the noise removal process part 13a in FIG. 1 applies noise removal based on the determination information input from the noise removal selection part 12a with respect to an input image signal, and outputs it as an output image signal. More specifically, when step ST102 in which normal noise removal is selected is No, or when step ST103 is Yes, normal noise removal is applied to the input image signal (step ST104). When step ST103 in which specific color noise removal is selected is No, specific color noise removal is applied to the input image signal (step ST105).

  Here, noise removal is performed with a low-pass filter having a total area of a × b pixels of horizontal a pixels and vertical b pixels. Here, a and b are integers, and 0 <a <horizontal pixel number and 0 <b <vertical pixel number are satisfied with respect to the image size. The area size of the a × b pixel is not necessarily matched with the size of the edge extraction range of the m × n pixel of the edge extraction unit 11a.

  The noise removal unit 13a changes the low-pass filter coefficient applied to the noise removal of the specific color region and the low-pass filter coefficient applied to the normal noise removal, thereby changing the low-pass filter having a different noise removal effect (intensity). And an appropriate low-pass filter based on the determination information input from the noise removal selection unit 12a is selected to remove noise from the input image signal. Note that the low-pass filter for removing noise may be applied only to the Y value, but may be applied to the Cb value and the Cr value. Further, by combining both, the Y value, Cb value, Cr It may be applied to all of the values.

  As described above, in the image processing apparatus according to the first embodiment of the present invention, the specific color region determination is performed, and the target pixel in the specific color region is further determined by determining the presence or absence of an edge around the target pixel. Even if the edge exists around the target pixel even in the color area, the same normal noise processing as that for the pixel outside the specific color area is applied to prevent the edge from blurring. When it does not exist, it is possible to achieve highly effective noise removal by applying specific color noise removal that has a large noise removal effect.

  In addition, the edge information extracted by the edge extraction unit 11a has been described with the binary presence / absence of the edge in the peripheral region of the pixel of interest, that is, whether the edge exists or not, but the probability of edge-likeness in each pixel For example, a value calculated using a Sobel filter which is a linear first-order differential filter may be used as edge information. As described above, when the edge information is expressed as a probability, a predetermined threshold (input signal is not shown) is provided. For example, an edge is present if the threshold is equal to or greater than the threshold, and an edge is present if the threshold is less than the threshold. What is necessary is just to determine that it does not exist. Note that the edge extraction unit of the edge extraction unit 11a is not limited to the Sobel filter, and, for example, a Prewitt filter, a Roberts filter of another primary differential filter, a Laplacian filter of a secondary differential filter, or the like may be applied.

  Here, the edge information is described as a case where the edge extraction unit 11a performs binarization determination of the presence / absence of an edge based on a predetermined threshold with respect to the probability of edge-likeness. It is also possible to perform the binarization determination of the presence / absence of an edge based on the same predetermined threshold value as it is with the probability of likelihood.

  In the first embodiment of the present invention, step ST101 for extracting edge information in the flowchart of FIG. 3 is performed before step ST102 for determining a specific color area, but since edge information is not used in step ST102. The edge information extraction step may be performed thereafter, in short, may be performed by step ST103 in which the extracted edge information is used. For example, the arrangement may be replaced immediately before step ST103.

  As described above, the image processing apparatus according to the first embodiment of the present invention switches between specific color noise removal and normal noise removal having different intensities based on edge information for the target pixel included in the specific color region. Since it is applied, the blurring of the edge is suppressed in the area where the edge exists around the target pixel, the resolution is maintained, and the effect of performing appropriate noise removal for the specific color area where the edge does not exist is obtained. Can do.

Embodiment 2. FIG.
The second embodiment of the present invention is configured as an image processing device 10b including a noise removal selection unit 12b obtained by changing the internal configuration of the noise removal selection unit 12a of the image processing device 10a according to the first embodiment of the present invention. .

  In the second embodiment of the present invention, compared to the first embodiment of the present invention, statistical values calculated from the color signals of the target pixel and the peripheral pixels based on the edge information, for example, the average value and the mode value of the color signal. When the calculated statistical value is included in the specific color area, switching between specific color noise removal and normal noise removal with different intensities based on edge information is performed. The image processing apparatus to be applied will be described.

  An image processing apparatus 10b according to Embodiment 2 of the present invention has the same configuration as that of FIG. 1 shown in Embodiment 1 of the present invention, except that the noise removal selection unit 12a is replaced with a noise removal selection unit 12b. It has become. It should be noted that the edge extraction unit 11a and the noise removal processing unit 13a having the same reference numerals have the same or equivalent functions as the block in the block diagram of FIG. The description is omitted.

  The noise removal selection unit 12b according to the second embodiment of the present invention selects noise removal processing to be applied to the target pixel based on the edge information extracted by the edge extraction unit 11a and the specific color area information.

  FIG. 4 is a block diagram showing an example of the configuration of the noise removal selection unit 12b of the image processing apparatus 10b according to Embodiment 2 of the present invention. The noise removal selection unit 12b includes a specific color area determination unit 21b. The noise removal determination means 22 is comprised. It should be noted that there is no change from the block in the block diagram of FIG. 2 according to the first embodiment of the present invention, and the noise removal determination means 22 denoted by the same reference sign performs the same or equivalent function.

  In FIG. 4, the specific color area determination unit 21b sets a setting range (referred to as a specific color determination range) for the pixel of interest based on the edge information extracted by the edge extraction unit 11a, Whether the statistical value of the color signal of the pixel of interest is within the specific color area or outside the specific color area is determined based on the specific color area information. The specific color determination range does not necessarily need to match the size of the edge extraction range of the edge extraction unit 11a and the size of the low-pass filter of the noise removal processing unit 13a. The noise removal determination unit 22 removes the specific color noise applied to the target pixel whose color signal is in the specific color region and the target pixel outside the specific color region with respect to the target pixel determined by the specific color region determination unit 21b. Any one of the normal noise removal to be applied to is selected and used as determination information. However, the noise removal determination unit 22 determines whether the statistical value in the specific color determination range output by the specific color region determination unit 21b is within the specific color region or outside the specific color region, and the edge information extracted by the edge extraction unit 11a. Use to select whether to remove specific color noise or normal noise. The noise removal processing unit 13a applies noise removal processing of any one of specific color noise removal and normal noise removal from the determination information selected by the noise removal selection unit 12b to remove noise from the target pixel. .

  Next, the operation of each part of the image processing apparatus 10b will be described with reference to the flowchart of FIG. In the flowchart of FIG. 3, a part that performs an operation different from each part of the image processing apparatus 10a according to the first embodiment of the present invention is step ST102 that shows processing related to the noise removal selection unit 12b. Description of the same or equivalent operation in other parts is omitted.

  The specific color area determination unit 21b of the noise removal selection unit 12b in FIGS. 1 and 4 includes a target pixel of the input image signal and peripheral pixels in the specific color determination range set based on the edge information extracted by the edge extraction unit 11a. The statistical value of the color signal is calculated, and it is determined whether the statistical value is within the specific color area or outside the specific color area based on the specific color area information (step ST102).

  As described above, the noise removal selection unit 12b appropriately selects noise removal based on the determination by the specific color region determination unit 21b and the determination by the noise removal determination unit 22 based on the edge information. In other words, by determining the presence or absence of an edge around the target pixel for the target pixel in the specific color area determined with respect to the statistical value of the color signal with the peripheral pixels in the specific color determination range, Even when there is an edge around the pixel of interest, determination information for selecting the same normal noise processing as that for the pixel outside the specific color region is output, and when there is no edge, the effect of noise removal is large By outputting the determination information for selecting the specific color noise removal, the noise removal process of the noise removal unit 13a is highly effective noise while preserving the edges around the pixel of interest having the color signal in the specific color region so as not to blur. Realize removal.

  Further, the specific color region determination unit 21b can change the specific color determination range of the color signal of the peripheral pixel applied to the calculation of the statistical value based on the edge information by referring to the edge information. . For example, if the edge is not in the vicinity, the specific color determination range is widened, and if the edge is in the periphery, the specific color determination range is narrowed to suppress the influence of the edge on the statistical value of the color signal.

  In the noise removal selection unit 12a of the image processing apparatus 10a according to the first embodiment of the present invention described above, the specific color area determination unit 21a determines whether the color signal of the target pixel is within the specific color area. It was determined whether it was outside the specific color area. However, in the noise removal selection unit 12b of the image processing device 10b according to the second embodiment of the present invention, the specific color region determination unit 21b includes the target pixel and the surrounding pixels of the specific color determination range set based on the edge information. As described above, for example, it is determined whether the color signal is within the specific color region or outside the specific color region with respect to the average value or the mode value of the color signal. Alternatively, the statistical value of the color signal may be a weighted average value obtained by changing the weight according to the distance from the target pixel. By using such a statistical value, the influence of a large noise is suppressed locally.

  As described above, the image processing apparatus according to Embodiment 2 of the present invention switches between specific color noise removal and normal noise removal having different intensities based on edge information for the pixel of interest included in the specific color region. Therefore, the same effects as those of the image processing apparatus according to Embodiment 1 of the present invention can be obtained.

  In addition, the image processing apparatus according to Embodiment 2 of the present invention sets a specific color determination range of peripheral pixels used for calculating a statistical value based on edge information for a target pixel included in the specific color region. Since it is determined whether the statistical value calculated from the color signal of the target pixel and the surrounding pixels is within the specific color region or outside the specific color region, the influence of local noise is suppressed. The effect of improving the determination accuracy of the specific color region can be obtained as compared with determination by itself.

Embodiment 3 FIG.
In the third embodiment of the present invention, the specific color noise removal and the normal noise removal having different intensities are switched based on the edge information as compared with the first embodiment of the present invention. The image processing apparatus 10c that switches the removal will be described.

  FIG. 5 is a block diagram showing an example of an image processing apparatus 10c according to Embodiment 3 of the present invention. The edge extraction unit 11a according to Embodiment 2 of the present invention is replaced with an edge extraction unit 11b, and noise removal selection is performed. The unit 12b is replaced with a noise removal selection unit 12c. The outline of the function of each block will be described based on this block diagram. It should be noted that there is no change from the block in the block diagram of FIG. 1 according to the second embodiment of the present invention, and that the noise removal processing unit 13a assigned the same reference sign performs the same or equivalent function, and the description thereof is omitted. .

  The blocks of the edge extraction unit 11b and noise removal selection unit 12c in FIG. 5 will be described in detail with reference to FIGS. 6 and 7 showing the internal configuration in more detail.

  First, FIG. 6 is a block diagram showing an example of the configuration of the edge extraction unit 11b of the image processing apparatus 10c according to the third embodiment of the present invention, and includes an edge extraction unit 31 and an edge information storage unit 32.

  The edge extraction unit 31 of the edge extraction unit 11b in FIG. 6 extracts edge information indicating whether or not an edge exists in the edge extraction range of peripheral pixels with respect to the target pixel of the input image signal. The edge information storage unit 32 stores the edge information extracted by the edge extraction unit 31 as peripheral pixel information, and outputs edge information for the peripheral pixel, that is, peripheral pixel information together with the edge information for the target pixel.

  Next, FIG. 7 is a block diagram showing an example of the configuration of the noise removal selection unit 12c of the image processing apparatus 10c according to Embodiment 3 of the present invention. The noise removal selection unit 12c is a specific color region determination unit 21c. And noise removal determination means 22. It should be noted that there is no difference from the block in the block diagram of FIG. 4 according to the second embodiment of the present invention, and the noise removal determination means 22 denoted by the same reference sign performs the same or equivalent function.

  The specific color region determination means 21c of the noise removal selection unit 12c in FIG. 7 includes the edge information extracted by the edge extraction unit 11a for the target pixel of the input image signal and the surroundings within the specific color determination range stored by the edge extraction unit 11a. Whether the statistical value of the color signal of the pixel based on the pixel information is within the specific color area or outside the specific color area is determined based on the specific color area information. The noise removal determination unit 22 removes the specific color noise applied to the target pixel whose color signal is in the specific color region and the target pixel outside the specific color region with respect to the target pixel determined by the specific color region determination unit 21c. Any one of the normal noise removal to be applied to is selected and used as determination information. However, the noise removal determination unit 22 determines whether the statistical value in the specific color determination range output by the specific color region determination unit 21b is within the specific color region or outside the specific color region, and the edge information extracted by the edge extraction unit 11a. Use to select whether to remove specific color noise or normal noise. The noise removal processing unit 13a applies noise removal processing of any one of specific color noise removal and normal noise removal from the determination information selected by the noise removal selection unit 12b to remove noise from the target pixel. .

  Next, the operation of each unit of the image processing apparatus 10c will be described with reference to the flowchart of FIG. In the flowchart of FIG. 8, portions that perform operations different from those of the respective units of the image processing apparatus 10a according to the first embodiment of the present invention are step ST201 (addition) indicating processing related to the edge extraction unit 11b, and a noise removal selection unit 12c. It is ST102 which shows the process which concerns on. The explanation of the operation of other identical or equivalent parts is omitted.

  First, the edge extraction unit 31 of the edge extraction unit 11b in FIGS. 5 and 6 receives an input image signal, and uses edge information as to whether or not an edge exists in an edge extraction range including peripheral pixels of the pixel of interest. Extract (step ST101).

  Next, the edge information storage unit 32 of the edge extraction unit 11b stores the edge information extracted by the edge extraction unit 31 as peripheral pixel information. As a result, the edge information storage means 32 that stores the edge information extracted in the past outputs the stored edge information as the peripheral pixel information, which can be referred to (step ST201).

  Further, the specific color area determination unit 21c of the noise removal selection unit 12c in FIGS. 5 and 7 sets a specific color determination range of peripheral pixels to be referred to based on the edge information extracted by the edge extraction unit 31 of the edge extraction unit 11b. The statistical value of the color signal is calculated for only the peripheral pixels in which no edge exists with reference to the target pixel and the peripheral pixel information within the specific color determination range stored in the edge information storage unit 32 of the edge extraction unit 11b. Then, it is determined whether it is within the specific color area or outside the specific color area (step ST102).

  The specific color area determination unit 21c of the noise removal selection unit 12c is configured to detect the target pixel and the peripheral pixels in the same manner as the specific color area determination unit 21b of the noise removal selection unit 12b of the image processing apparatus 10b according to Embodiment 2 of the present invention. In order to set the specific color determination range of the peripheral pixel to be referred to based on the edge information extracted by the edge extraction unit 11a and the peripheral pixel information stored by the edge extraction unit 11a. Rather than setting the peripheral pixel range based only on the edge information extracted by the extraction unit 11a, only the peripheral pixels in which no edge exists from the peripheral pixel information can be set as the statistical value calculation target. And the influence of the presence of an edge can be suppressed in the subsequent noise removal selection.

  Here, the specific color determination range is set based on edge information and peripheral pixel information, such as the size and shape, and the arrangement with respect to the target pixel. Also good. In calculating the color signal statistic value, the influence of the presence of an edge can be determined even if a pixel having an edge in the vicinity is included in the specific color determination range or excluded from the target pixel by referring to the surrounding pixel information. Therefore, regardless of the edge information, the size, shape, and arrangement of the specific color determination range are fixed without being sequentially changed, that is, the specific color region determination means 21c used for the setting is fixed. A configuration that does not require input edge information is also included in the modification of the third embodiment.

  As described above, in the image processing apparatus according to the third embodiment of the present invention, as in the third embodiment of the present invention, the statistical value of the color signal calculated by the specific color region determining means 21c of the noise removal selecting unit 12c is used. By using it, the influence of local noise can be suppressed. Further, the specific color area determination unit 21b of the noise removal selection unit 12c refers to the edge information, and sets the specific color determination range of the color signal of the peripheral pixels applied to the calculation of the statistical value based on the edge information. Furthermore, even within the set specific color determination range, by referring to the peripheral pixel information, the peripheral pixels to be applied to the calculation of the statistical value are limited to the pixels where the edge does not exist, and the influence of the presence of the edge is limited. Can be suppressed.

  In the image processing apparatus according to Embodiment 3 of the present invention, the edge information extracted by the edge extraction unit 31 of the edge extraction unit 11b will be described in the processing procedure in which the edge information storage unit 32 sequentially stores the information as peripheral pixel information. Went. The peripheral pixel information output from the edge information storage means 32 is set by storing the edge information, but the peripheral pixel information for the pixel that is not extracted from the edge information that is located behind the target pixel is in an indefinite state. It is in. Therefore, a pixel in an indefinite state located behind the target pixel is a target pixel for calculating a statistical value based on color signals of peripheral pixels within the specific color determination range in the specific color region determination unit 21c of the noise removal selection unit 12c. To remove from. For this reason, the edge information indicating the presence or absence of a peripheral edge for each pixel stored in the edge information storage means 32 may be prepared as a value indicating an indefinite state as an initial value, or a target pixel for statistical value calculation A value indicating that there is an edge may be set so as to be removed from the range. In order to reduce the number of pixels located behind the target pixel in the specific color determination range, for example, the target pixel may be arranged below the center of the specific color determination range.

  Further, as a modification of the third embodiment of the present invention, it is necessary as peripheral pixel information for the target pixel so that the edge extraction unit 11b can output peripheral pixel information (edge information) within the specific color determination range for the target pixel. A minimum buffer memory (not shown) for extracting the edge information is prepared, and the edge information output as peripheral pixel information is pre-extracted by pre-scanning the image signal stored in the buffer memory. Then, it may be stored in the edge information storage means 32 to eliminate the output of the peripheral pixel information in an indefinite state. In this buffer memory, the specific color region determination unit 21c of the noise removal selection unit 12c sets a specific color determination range based on the edge information, so that peripheral pixel information (edge information) of peripheral pixels that can be at least the specific color determination range It is assumed that the peripheral pixels necessary for extracting can be accumulated. Further, when a configuration including a frame memory (not shown) for accumulating image signals can be adopted, the image signal is pre-scanned, and the edge information of the entire image frame is extracted in advance and stored in the edge information storage means 32. You may adopt the procedure to do.

  As described above, the image processing apparatus according to Embodiment 3 of the present invention switches between specific color noise removal and normal noise removal with different intensities based on edge information for the pixel of interest included in the specific color region. Therefore, the same effect as that of the image processing apparatus according to Embodiment 1 of the present invention can be obtained.

  In addition, the image processing apparatus according to Embodiment 3 of the present invention sets a specific color determination range of peripheral pixels used for calculating a statistical value based on edge information for the target pixel included in the specific color region. Since it is determined whether the statistical value calculated from the color signal of the target pixel and the surrounding pixel is within the specific color region or outside the specific color region, the influence of local noise is suppressed. The same effects as those of the image processing apparatus according to the second embodiment can be obtained.

  Furthermore, the image processing apparatus according to Embodiment 3 of the present invention uses the peripheral pixel information as a statistical value calculation target only for peripheral pixels that do not have an edge in the specific color determination range, and the colors of the target pixel and the peripheral pixels. Since it is determined whether the statistical value calculated from the signal is within the specific color area or outside the specific color area and the influence of local noise is suppressed, the color signal including peripheral pixels where edges exist It is possible to obtain an effect of improving the determination accuracy of the specific color region rather than calculating and determining the statistic value.

Embodiment 4 FIG.
In the fourth embodiment of the present invention, an image processing apparatus 10d that first performs a color difference reduction process on the color difference components of an input image signal will be described with respect to the third embodiment of the present invention.

  FIG. 9 is a block diagram showing an example of an image processing apparatus 10d according to Embodiment 4 of the present invention, and is configured by adding a color difference reduction processing unit 14 to the block diagram of Embodiment 3 of the present invention. . The outline of the function of each block will be described based on this block diagram. Note that the edge extraction unit 11b, the noise removal selection unit 12c, and the noise removal processing unit 13a having the same reference numerals are the same as or equivalent to those in the block diagram of FIG. 5 according to the third embodiment of the present invention. The description of the function is omitted. Although the color difference reduction processing unit 14 is shown as being shared outside the edge extraction unit 11b, the noise removal selection unit 12c, and the noise removal processing unit 13a, the edge extraction unit 11b and the noise removal selection unit 12c are shown. Each of the noise removal processing units 13a may include a color difference reduction processing unit 14 independently. Alternatively, one of the edge extraction unit 11b, the noise removal selection unit 12c, and the noise removal processing unit 13a may include the color difference reduction processing unit 14 independently, and the remaining two may be used while being shared.

  The color difference reduction processing unit 14 in FIG. 9 performs color difference reduction processing on the input image signal as a pre-process in the edge extraction unit 11b, the noise removal selection unit 12c, and the noise removal processing unit 13a. Here, by applying the color difference reduction process, there is no change in the color signal (color space component) of the image signal composed of the components of luminance Y, color differences Cb, and Cr, and the luminance Y remains unconverted and the components of the color differences Cb and Cr. The input image signal subjected to the color difference reduction processing is subjected to processing corresponding to Embodiment 3 of the present invention to remove noise.

  By performing the color difference reduction process as a noise removal pre-process, the noise of the color difference signal is suppressed, and the effect of noise removal on the subsequent color difference signal can be improved. Such preprocessing is particularly effective in suppressing noise that tends to appear in a high-sensitivity image. An example of color difference reduction processing (color difference reduction method) performed by the color difference reduction processing unit 14 will be described. Here, it is assumed that the color difference signal Cb value and the Cr value take positive and negative values such as −128 to +127, for example.

  The case where the color difference reduction process applied to the color difference signal Cb value and the Cr value is applied to the color difference signal Cb value will be described. First, the average value Ave of the color difference signal Cb values in the local region of horizontal x pixels, vertical y pixels, and total x × y pixels is calculated. Here, x and y are integers, and 0 <x <horizontal pixel number and 0 <y <vertical pixel number are satisfied with respect to the image size. Next, for the Cb value of the target pixel, the absolute value and the sign signal indicating positive / negative are stored separately, and the average value Ave is subtracted from the absolute value of the Cb value. Further, the absolute value obtained by subtracting the average value Ave is multiplied by a coefficient. The value of this coefficient is a value from 0.0 to 1.0, and the absolute value close to the average value Ave, that is, the absolute value close to 0 by subtracting the average value Ave, decreases the coefficient value, The coefficient value is increased as the absolute value is farther from Ave, and is smoothly changed according to the distance from the average value Ave. Finally, the average value Ave is added to the absolute value of the Cb value multiplied by the coefficient, and the stored code signal is added. The color difference signal Cr value may be processed similarly. As described above, after the color difference reduction process is applied to each pixel for the color difference signals Cb and Cr of the input image signal, the subsequent noise removal process is performed.

  Next, the operation of each unit of the image processing apparatus 10d will be described with reference to the flowchart of FIG. In the flowchart of FIG. 10, a part that performs an operation different from each part of the image processing apparatus 10c according to the third embodiment of the present invention is step ST200 (addition) indicating processing related to the color difference reduction processing unit 14. The explanation of the operation of other identical or equivalent parts is omitted.

  The color difference reduction processing unit 14 in FIG. 9 performs color difference reduction processing on the color difference Cb value and Cr value of the input image signal (step ST200). The processing of steps ST101, ST201, ST102 to ST105 is performed on the image signal composed of the color difference Cb value and the Cr value with reduced color difference while the luminance Y value remains unchanged.

  As described above, in the image processing apparatus according to the fourth embodiment of the present invention, the color difference reduction process is applied as the noise removal pre-processing so as to suppress the noise of the color difference signal. Yes.

  As described above, the image processing apparatus according to the fourth embodiment of the present invention switches between specific color noise removal and normal noise removal with different intensities based on edge information for the pixel of interest included in the specific color region. Therefore, the same effect as that of the image processing apparatus according to Embodiment 1 of the present invention can be obtained.

  In addition, the image processing apparatus according to Embodiment 4 of the present invention sets a specific color determination range of peripheral pixels used for calculating a statistical value based on edge information for a target pixel included in the specific color region. Since it is determined whether the statistical value calculated from the color signal of the target pixel and the surrounding pixel is within the specific color region or outside the specific color region, the influence of local noise is suppressed. The same effects as those of the image processing apparatus according to the second embodiment can be obtained.

  In addition, the image processing apparatus according to Embodiment 4 of the present invention uses the peripheral pixel information as a statistical value calculation target only for the peripheral pixels that do not have an edge in the specific color determination range, and the colors of the target pixel and the peripheral pixels. Since it is determined whether the statistical value calculated from the signal is within the specific color region or outside the specific color region and the influence of local noise is suppressed, the image processing according to the third embodiment of the present invention The same effect as the device can be obtained.

  In addition, the image processing apparatus according to the fourth embodiment of the present invention performs the processing of the image processing apparatus according to the third embodiment of the present invention after applying the color difference reduction process to the color difference signal of the input image signal. Since this is done, it is possible to obtain a more accurate noise removal effect based on the color difference signal in which noise is suppressed.

Embodiment 5. FIG.
In Embodiment 5 of the present invention, a color space conversion unit 15 is provided, and an image signal input to the noise removal selection unit 12a of the image processing apparatus 10a according to Embodiment 1 of the present invention is first subjected to color space conversion. An image processing apparatus 10e including a noise removal selection unit 12d that performs processing after color space conversion by the unit 15 will be described.

  FIG. 11 is a block diagram showing an example of an image processing apparatus 10e according to the fifth embodiment of the present invention, and a color preceding the noise removal selection unit 12a in the block diagram (FIG. 1) of the first embodiment of the present invention. The space conversion unit 15 is added. The outline of the function of each block will be described based on this block diagram. It should be noted that the edge extraction unit 11a and the noise removal processing unit 13a having the same reference numerals have the same or equivalent functions as the block in the block diagram of FIG. The description is omitted. Although the color space conversion unit 15 is shown as being outside the noise removal selection unit 12d, the noise removal selection unit 12d may be configured to include the color space conversion unit 15.

  The color space conversion unit 15 in FIG. 11 applies color space conversion to the input image signal. The noise removal selection unit 12d is a noise to be applied to the target pixel based on the edge information and specific color region information extracted by the edge extraction unit 11a from the image signal (color information) color space converted by the color space conversion unit 15. Select the removal process.

  The noise removal selection unit 12d has the same internal configuration as the noise removal selection unit 12a in FIG. However, the specific color area determination unit 21a determines the color signal of the target pixel of the image signal color space converted by the color space conversion unit 15 based on the specific color area information of the color space converted by the color space conversion unit 15. , Whether it is in the specific color area or outside the specific color area.

  Next, the operation of each unit of the image processing apparatus 10e will be described with reference to the flowchart of FIG. In the flowchart of FIG. 12, the portions that perform different operations from the respective units of the image processing apparatus 10a according to the first embodiment of the present invention are step ST301 (addition) indicating processing related to the color space conversion unit 15, and a noise removal selection unit. It is step ST102 which shows the process which concerns on the 12d specific color area | region determination means 21a. The explanation of the operation of other identical or equivalent parts is omitted.

  First, the color space conversion unit 15 in FIG. 11 performs color space conversion on the input image signal (step ST301).

  Also, the edge extraction unit 11a in FIG. 11 receives the input image signal and extracts as edge information whether an edge exists or does not exist within the edge extraction range of the peripheral region of the target pixel (step ST101).

  Next, the specific color region determination unit 21a of the noise removal selection unit 12d in FIG. 11 converts the color signal of the target pixel of the image signal color space converted by the color space conversion unit 15 into the color space conversion by the color space conversion unit 15. It is determined whether the color is within the specific color area or outside the specific color area based on the specific color area information of the color space (step ST102). Thereafter, the processing from step ST103 to ST105 is performed as in the first embodiment of the present invention.

  Thus, in the image processing apparatus according to Embodiment 5 of the present invention, color space conversion is applied, based on color information converted into hue, saturation, and brightness, which are parameters closer to human senses. The specific color area can be determined according to the visual characteristics, and noise can be removed with high accuracy.

  In the fifth embodiment of the present invention, step ST301 for performing color space conversion on the input image signal in the flowchart of FIG. 12 is performed before step ST101 for extracting edge information. Since the information is not used, the edge information extraction step may be performed after that. In short, it may be performed up to step ST102 for determining the specific color region. For example, the arrangement may be replaced immediately before step ST102.

  Further, step ST101 for extracting edge information in the flowchart of FIG. 12 is performed before step ST102 for determining a specific color area, as in the first embodiment of the present invention. In step ST102, edge information is used. Therefore, the step of extracting edge information may be performed after that, and in short, may be performed up to step ST103 in which the extracted edge information is used. For example, the arrangement may be replaced immediately before step ST103.

  As described above, the image processing apparatus according to Embodiment 5 of the present invention switches between specific color noise removal and normal noise removal having different intensities based on edge information for the target pixel included in the specific color region. Therefore, the same effect as that of the image processing apparatus according to Embodiment 1 of the present invention can be obtained.

  In addition, since the image processing apparatus according to Embodiment 5 of the present invention performs noise removal after applying color space conversion to the input image signal to the noise removal selection unit, it uses parameters closer to human senses. A specific color region can be determined according to visual characteristics based on color information converted into a certain hue, saturation, and brightness, and a highly accurate noise removal effect can be obtained.

Embodiment 6 FIG.
In Embodiment 6 of the present invention, a color space conversion unit 15 is provided, and an image signal input to the edge extraction unit 11a of the image processing apparatus 10a according to Embodiment 1 of the present invention is first converted to a color space conversion unit. An image processing apparatus 10f including an edge extraction unit 11c that performs processing after color space conversion at 15 will be described.

  FIG. 13 is a block diagram showing an example of an image processing apparatus 10f according to Embodiment 5 of the present invention. A color space is provided before the edge extraction unit 11a in the block diagram (FIG. 1) of Embodiment 1 of the present invention. The conversion unit 15 is added. The outline of the function of each block will be described based on this block diagram. It should be noted that the noise removal selection unit 12a and the noise removal processing unit 13a assigned the same reference numerals have the same or equivalent functions without any change from the block in the block diagram of FIG. 1 of the first embodiment of the present invention. The description is omitted. Although the color space conversion unit 15 is shown as being outside the edge extraction unit 11c, the edge extraction unit 11c may be configured to include the color space conversion unit 15.

  The color space conversion unit 15 in FIG. 13 applies color space conversion to the input image signal. The edge extraction unit 11c extracts edge information indicating whether or not an edge exists in the edge extraction range of peripheral pixels with respect to the target pixel of the image signal subjected to color space conversion by the color space conversion unit 15.

  Next, the operation of each part of the image processing apparatus 10f will be described with reference to the flowchart of FIG. In the flowchart of FIG. 12, the parts that perform different operations from the respective parts of the image processing apparatus 10a according to the first embodiment of the present invention are step ST301 (addition) indicating processing related to the color space conversion unit 15, and the edge extraction unit 11c. This is step ST101 showing the processing related to. The explanation of the operation of other identical or equivalent parts is omitted.

  First, the color space conversion unit 15 in FIG. 13 performs color space conversion on the input image signal (step ST301).

  Further, the edge extraction unit 11a in FIG. 13 receives the image signal that has been color space converted by the color space conversion unit 15 as input, and uses the edge information as to whether or not an edge exists within the edge extraction range of the peripheral region of the pixel of interest. Extract (step ST101). Thereafter, the processing of steps ST102 to ST105 is performed as in the first embodiment of the present invention.

  As described above, in the image processing apparatus according to the sixth embodiment of the present invention, color space conversion is applied, based on color information converted into hue, saturation, and brightness, which are parameters closer to human senses. Edges can be extracted according to visual characteristics, and noise can be removed with high accuracy.

  As described above, the image processing apparatus according to Embodiment 6 of the present invention switches between specific color noise removal and normal noise removal having different intensities based on edge information for the pixel of interest included in the specific color region. Therefore, the same effect as that of the image processing apparatus according to Embodiment 1 of the present invention can be obtained.

  In addition, since the image processing apparatus according to Embodiment 6 of the present invention performs noise removal after applying color space conversion to the input image signal to the edge extraction unit, it is a parameter closer to a human sense. Edges can be extracted according to visual characteristics based on color information converted into hue, saturation, and brightness, and a highly accurate noise removal effect can be obtained.

Embodiment 7 FIG.
In Embodiment 7 of the present invention, a color space conversion unit 15 is provided, and image signals input to the edge extraction unit 11a and the noise removal selection unit 12a of the image processing apparatus 10a according to Embodiment 1 of the present invention are First, an image processing apparatus 10g including an edge extraction unit 11c and a noise removal selection unit 12d that are processed after color space conversion by the color space conversion unit 15 will be described.

  FIG. 14 is a block diagram showing an example of an image processing apparatus 10g according to Embodiment 7 of the present invention, and an edge extraction unit 11a and a noise removal selection unit in the block diagram (FIG. 1) of Embodiment 1 of the present invention. The color space conversion unit 15 is added to the previous stage of 12a. The outline of the function of each block based on this block diagram is shown in the block diagram of FIG. 1 of the first embodiment of the present invention, FIG. 11 of the fifth embodiment of the present invention, and FIG. 13 of the sixth embodiment of the present invention. The edge extraction unit 11c, the noise removal selection unit 12d, the noise removal processing unit 13a, and the color space conversion unit 15 that have the same reference numerals and are not changed are the same or equivalent functions and will not be described. . Note that the color space conversion unit 15 is shared and shown as being outside the edge extraction unit 11c and the noise removal selection unit 12d, but each of the edge extraction unit 11c and the noise removal selection unit 12d is independently colored. A configuration including the space conversion unit 15 may be used.

  Next, the operation of each unit of the image processing apparatus 10g will be described with reference to the flowchart of FIG. In the flowchart of FIG. 12, the parts that perform different operations from the respective parts of the image processing apparatus 10a according to the first embodiment of the present invention are step ST301 (addition) indicating processing related to the color space conversion unit 15, and the edge extraction unit 11c. Step ST101 indicating the processing related to the above, and Step ST102 indicating the processing related to the specific color region determination means 21a of the noise removal selection unit 12d. However, the added step ST301 and the changed steps ST101 and ST102 are edges corresponding to the blocks of the image processing apparatus 10e according to the fifth embodiment of the present invention or the image processing apparatus 10f according to the sixth embodiment of the present invention. The operation is the same as or equivalent to that of the extraction unit 11c, the noise removal selection unit 12d, and the color space conversion unit 15, and the explanation of the operation of other same or equivalent parts is omitted.

  Thus, in the image processing apparatus according to the seventh embodiment of the present invention, color space conversion is applied, based on color information converted into hue, saturation, and brightness, which are parameters closer to human senses. It is possible to remove noise with high accuracy according to visual characteristics.

  As described above, the image processing apparatus according to Embodiment 7 of the present invention switches between specific color noise removal and normal noise removal having different intensities based on edge information for the target pixel included in the specific color region. Therefore, the same effect as that of the image processing apparatus according to Embodiment 1 of the present invention can be obtained.

  Further, the image processing apparatus according to Embodiment 7 of the present invention performs noise removal after applying color space conversion to the input image signal to the edge extraction unit and the noise removal selection unit. The same effects as those of the image processing apparatuses according to the fifth and sixth embodiments can be obtained.

Embodiment 8 FIG.
In the eighth embodiment of the present invention, the image signal input to the noise removal processing unit 13a of the image processing apparatus 10g according to the seventh embodiment of the present invention is converted into an image signal that has been color space converted by the color space conversion unit 15. The processing device 10h will be described. Further, the image processing apparatus 10h is described as a configuration including a color space reverse conversion unit 16 that performs color space reverse conversion with respect to the color space conversion of the color space conversion unit 15 on the image signal from which noise has been removed by the noise removal processing unit 13b. To do.

  FIG. 15 is a block diagram showing an example of an image processing apparatus 10h according to the eighth embodiment of the present invention. A color preceding the noise removal processing unit 13a in the block diagram (FIG. 14) of the seventh embodiment of the present invention is shown. The color space reverse conversion unit 16 is added to the subsequent stage of the space conversion unit 15 and the noise removal processing unit 13a. The outline of the function of each block based on this block diagram is the same as that of the block diagram of FIG. 14 according to the seventh embodiment of the present invention, and the edge extraction unit 11c and noise removal selection unit 12d assigned the same reference numerals The color space conversion unit 15 performs the same or equivalent functions, and a description thereof will be omitted. The color space conversion unit 15 is shown as being shared outside the edge extraction unit 11c, the noise removal selection unit 12d, and the noise removal processing unit 13b, but the edge extraction unit 11c and the noise removal selection unit 12d are shown. In addition, each of the noise removal processing unit 13b may include a color space conversion unit 15 independently. Alternatively, one of the edge extraction unit 11b, the noise removal selection unit 12c, and the noise removal processing unit 13a may include the color space conversion unit 15 independently, and the remaining two may be used while being shared.

  The noise removal processing unit 13b in FIG. 15 applies noise removal processing selected by the noise removal selection unit 12d to the image signal color space converted by the color space conversion unit 15 to remove noise. The color space reverse conversion unit 16 performs color space reverse conversion on the color space conversion performed by the color space conversion unit 15 on the image signal from which noise has been removed by the noise removal processing unit 13b.

  However, the image processing device 10h according to the eighth embodiment of the present invention may output the image signal from which noise is removed by the noise removal processing unit 13b without passing through the color space inverse transformation unit 16. This is because, for example, the input image signal is given by the luminance / color difference signal YCbCr, converted into hue, saturation, and brightness by the color space conversion unit 15, removed from noise, and then encoded and stored as it is by an external encoder. It can be used when communicating. At this time, if necessary, the color space inverse conversion unit 16 is provided on the reproduction device side, and inversely converts the hue, saturation, and lightness decoded by the decoder into the same luminance color difference signal YCbCr as the input image signal. Also good. Also. The input / output image signal is not limited to the luminance / color difference signal YCbCr but may be an RGB signal or the like.

  Next, the operation of each unit of the image processing apparatus 10h will be described with reference to the flowchart of FIG. In the flowchart of FIG. 16, the portions that perform operations different from those of the respective units of the image processing apparatus 10g according to the seventh embodiment of the present invention are steps ST104 and ST105 indicating the processing related to the noise removal processing unit 13b, and the color space inverse conversion unit. 16 is a step ST302 (addition) showing processing according to FIG. The explanation of the operation of other identical or equivalent parts is omitted.

  First, similarly to Embodiment 7, steps ST301 and steps ST101 to ST103 are performed, and the noise removal selection unit 12d selects a noise removal process to be applied by the noise removal processing unit 13b.

  Next, the noise removal processing unit 13b in FIG. 15 performs color space conversion on the normal noise removal processing (step ST104) or the specific color noise removal processing (step ST105) based on the noise removal processing selected by the noise removal selection unit 12d. The unit 15 applies the color space converted image signal to remove noise.

  In addition, the color space inverse conversion unit 16 in FIG. 15 performs inverse color space conversion on the image signal from which noise has been removed by the noise removal processing unit 13b, and inversely converts the image signal into an image signal in the same color space as the input image signal.

  Thus, in the image processing apparatus according to Embodiment 8 of the present invention, color space conversion is applied, based on color information converted into hue, saturation, and brightness, which are parameters closer to human senses. By performing noise removal, an edge is extracted according to visual characteristics, a specific color region is determined, and noise removal with high accuracy is possible.

  As described above, the image processing apparatus according to the eighth embodiment of the present invention switches between specific color noise removal and normal noise removal with different intensities based on edge information for the pixel of interest included in the specific color region. Therefore, the same effect as that of the image processing apparatus according to Embodiment 1 of the present invention can be obtained.

  Further, the image processing apparatus according to the eighth embodiment of the present invention performs noise removal after applying color space conversion to the input image signal to the edge extraction unit and the noise removal selection unit. The effects similar to those of the image processing apparatus according to the fifth to seventh embodiments can be obtained.

  In addition, since the image processing apparatus according to the eighth embodiment of the present invention performs noise removal after applying color space conversion to the input image signal to the noise removal processing unit, it uses parameters closer to human senses. Based on the color information converted into a certain hue, saturation, and lightness, a highly accurate noise removal effect can be obtained according to the visual characteristics.

  It should be noted that the same or equivalent functions of each block having the configuration described in the image processing apparatus according to the embodiment of the present invention may be applied in combination with other image processing apparatuses in the embodiment of the present invention. Absent. For example, the following image processing apparatus can be configured. Further, it is obvious that an image processing apparatus in which many functions are simultaneously mounted is also conceivable.

  The function of setting the specific color determination range based on the edge information performed by the specific color determination unit of the noise removal selection unit of the image processing apparatus according to the second to fourth embodiments of the present invention includes the edge information as the noise removal selection unit. May be applied to the image processing apparatus according to the fifth to eighth embodiments of the present invention.

  Also, the specification by the statistical value calculated from the color signal of the surrounding pixels included in the specific color determination range performed by the specific color determination means of the noise removal selection unit of the image processing apparatus according to the third and fourth embodiments of the present invention The function of determining the color area is to store and store edge information as peripheral pixel information in the edge information storage means of the edge extraction unit, and input the peripheral pixel information to the noise removal selection unit, thereby implementing the embodiment of the present invention. The present invention may be applied to image processing apparatuses according to Embodiments 5 to 8.

  The function of reducing the color difference of the input image signal performed by the color difference reduction processing unit of the image processing apparatus according to the fourth embodiment of the present invention is implemented from the first embodiment, the second embodiment, and the fifth embodiment of the present invention. If the input image signal to the image processing apparatus according to Embodiment 8 is an image signal composed of a luminance color difference signal, a color difference reduction processing unit may be provided in the same manner.

  The function of converting the color section performed by the color space conversion unit of the image processing apparatus according to the fifth to eighth embodiments of the present invention is the same as that of the image processing apparatus according to the first to fourth embodiments of the present invention. Similarly, the color space conversion unit may be provided in the corresponding unit and applied. In addition to the color space conversion unit, a color space inverse conversion unit may be similarly provided in the corresponding unit.

  The function of selecting the noise removal processing by the noise removal selection unit of the image processing apparatus according to the first to eighth embodiments of the present invention is based on whether or not the target pixel is in the specific color region based on the specific color region information. After the determination, the edge information is used to select the specific color noise removal for the target pixel that does not have an edge around the target pixel determined to be in the specific color region. The edge information is used to determine whether or not there is an edge around the pixel of interest, and for the pixel of interest determined to have no edge around it, the specific color region based on the specific color region information It is also possible to select to apply specific color noise removal to a target pixel in

  Note that the edge extraction unit of the image processing apparatus according to Embodiments 1 to 8 of the present invention has been described as a configuration that extracts edges from an image signal, but already extracted edge information is input together with the image signal. It may be a configuration. In this case, in the first to second embodiments and the fifth to eighth embodiments, the edge information is sequentially input together with the target pixel of the image signal in units of pixels, so that the edge extraction unit is not provided. There may be. In the third and fourth embodiments, the edge extraction unit is not provided with an edge extraction unit, and the edge information storage unit is a specific color determination that may be set by the specific color region determination unit of the noise removal selection unit. The configuration may be such that edge information as peripheral image information is stored in units of ranges or in units of one frame of image.

  When applying these functions described in the image processing apparatus according to the embodiment of the present invention to remove noise from an image signal, edge information is used to suppress blurring of the image without blurring the edge. And image processing for effectively removing noise of the image signal. In particular, a large effect can be obtained in removing noise from a high-sensitivity image in which a lot of noise is likely to appear as an image signal.

  As described above, the image processing apparatus according to the embodiment of the present invention switches and applies specific color noise removal and normal noise removal with different intensities based on edge information to the target pixel included in the specific color region. Therefore, it is possible to suppress the blurring of the edge in the area where the edge exists around the target pixel, maintain the resolution, and obtain the effect of performing appropriate noise removal for the specific color area where the edge does not exist. it can.

  Further, the image processing apparatus according to the embodiment of the present invention sets a specific color determination range of peripheral pixels used for calculation of a statistical value based on edge information for a target pixel included in the specific color region, Since the statistical value calculated from the color signal of the target pixel and surrounding pixels is within the specific color area or outside the specific color area, the influence of local noise is suppressed, so the target pixel alone As a result, it is possible to obtain the effect of improving the determination accuracy of the specific color region as compared with the above determination.

  In addition, the image processing apparatus according to the embodiment of the present invention uses the color signal of the target pixel and the peripheral pixel as a statistical value calculation target only for the peripheral pixel having no edge of the specific color determination range based on the peripheral pixel information. It is determined whether the statistical value calculated from is within the specific color area or outside the specific color area, and the influence of local noise is suppressed. The effect of improving the determination accuracy of the specific color region can be obtained rather than calculating and determining the statistical value.

  Further, the image processing apparatus according to the embodiment of the present invention performs noise removal after applying the color difference reduction process to the color difference signal of the input image signal. A high noise removal effect can be obtained.

  In addition, the image processing apparatus according to the embodiment of the present invention performs noise removal after applying color space conversion to the input image signal to the noise removal selection unit, and thus is a parameter closer to a human sense. A specific color region can be determined according to visual characteristics based on color information converted into hue, saturation, and brightness, and a highly accurate noise removal effect can be obtained.

  In addition, since the image processing apparatus according to the embodiment of the present invention performs noise removal after applying color space conversion to the input image signal to the edge extraction unit, the hue that is a parameter closer to human senses. Edges can be extracted according to visual characteristics based on color information converted into saturation and brightness, and a highly accurate noise removal effect can be obtained.

  In addition, the image processing apparatus according to the embodiment of the present invention performs noise removal after applying color space conversion to the input image signal to the noise removal processing unit, and thus is a parameter closer to a human sense. Based on the color information converted into hue, saturation, and lightness, a highly accurate noise removal effect can be obtained according to visual characteristics.

It is a block diagram which shows an example of a structure of the image processing apparatus by Embodiment 1 of this invention. It is a block diagram which shows an example of a structure of the noise removal selection part of the image processing apparatus by Embodiment 1 of this invention. It is a flowchart which shows an example of the process sequence of the image processing apparatus by Embodiment 1 of this invention and Embodiment 2 of this invention. It is a block diagram which shows an example of a structure of the noise removal selection part of the image processing apparatus by Embodiment 2 of this invention. It is a block diagram which shows an example of a structure of the image processing apparatus by Embodiment 3 of this invention. It is a block diagram which shows an example of a structure of the edge extraction part of the image processing apparatus by Embodiment 3 of this invention. It is a block diagram which shows an example of a structure of the noise removal selection part of the image processing apparatus by Embodiment 3 of this invention. It is a flowchart which shows an example of the process sequence of the image processing apparatus by Embodiment 3 of this invention. It is a block diagram which shows an example of a structure of the image processing apparatus by Embodiment 4 of this invention. It is a flowchart which shows an example of the process sequence of the image processing apparatus by Embodiment 4 of this invention. It is a block diagram which shows an example of a structure of the image processing apparatus by Embodiment 5 of this invention. It is a flowchart which shows an example of the process sequence of the image processing apparatus by Embodiment 5 of this invention from Embodiment 5 of this invention. It is a block diagram which shows an example of a structure of the image processing apparatus by Embodiment 6 of this invention. It is a block diagram which shows an example of a structure of the image processing apparatus by Embodiment 7 of this invention. It is a block diagram which shows an example of a structure of the image processing apparatus by Embodiment 8 of this invention. It is a flowchart which shows an example of the process sequence of the image processing apparatus by Embodiment 8 of this invention.

Explanation of symbols

10a, 10b, 10c, 10d, 10e, 10f, 10g, 10h Image processing apparatus 11a, 11b, 11c Edge extraction unit 12, 12a, 12b, 12c Noise removal selection unit 13a, 13b Noise removal processing unit 14 Color difference reduction processing unit 15 Color space conversion unit 16 Color space inverse conversion unit 21a, 21b Specific color area determination unit 22 Noise removal determination unit 31 Edge extraction unit 32 Edge information storage unit

Claims (13)

An edge extraction unit that extracts edge information about the presence or absence of an edge around the target pixel of the input image signal;
Based on the specific color area information, it is determined whether the color signal of the target pixel is within the specific color area or outside the specific color area, and the target pixel is used by using the determination result and the edge information extracted by the edge extraction unit. A noise removal selection unit for selecting a noise removal process to be applied;
An image processing apparatus comprising: a noise removal processing unit that applies the noise removal processing selected by the noise removal selection unit to the target pixel. The noise removal selection unit
Specific color region determination means for determining whether the color signal of the pixel of interest is within the specific color region or outside the specific color region based on the specific color region information;
The specific color area determination means determines that the specific color area is in the specific color area, and uses the edge information extracted by the edge extraction unit to select specific color noise removal for a pixel of interest that does not have an edge in the vicinity. The image processing apparatus according to claim 1, further comprising noise removal determination means for selecting normal noise removal for the target pixel. The noise removal selection unit
Based on the edge information extracted by the edge extraction unit, a specific color determination range for the target pixel is set, and the statistical values of color signals of peripheral pixels and the target pixel in the specific color determination range are included in the specific color region information. A specific color area determination means for determining whether the specific color area is inside or outside the specific color area,
The specific color area determination means determines that the specific color area is in the specific color area, and uses the edge information extracted by the edge extraction unit to select specific color noise removal for a pixel of interest that does not have an edge in the vicinity. The image processing apparatus according to claim 1, further comprising noise removal determination means for selecting normal noise removal for the target pixel. The edge extraction unit
Comprising edge information storage means for storing the extracted edge information as peripheral pixel information;
The specific color area determination unit is configured to generate the peripheral pixel and the target pixel corresponding to the peripheral pixel information indicating that no edge exists in the specific color determination range based on the peripheral pixel information stored in the edge information storage unit. The image processing apparatus according to claim 3, wherein a statistical value is calculated from the color signal. A color difference reduction processing unit that performs color difference reduction processing on the color difference signal of the input image signal is provided.
The edge extraction unit extracts edge information using the input image signal subjected to the color difference reduction processing by the color difference reduction processing unit as input,
The noise removal selection unit selects a noise removal process to be applied to the pixel of interest using the input image signal subjected to the color difference reduction processing by the color difference reduction processing unit as an input,
The image processing apparatus according to claim 1, wherein the noise removal processing unit removes noise using the input image signal subjected to the color difference reduction processing by the color difference reduction processing unit as an input. A color space conversion unit that converts an input image signal into an image signal of another color space;
The image processing apparatus according to claim 1, wherein the noise removal selection unit selects a noise removal process to be applied to the target pixel by using the input image signal color space converted by the color space conversion unit as an input. A color space conversion unit that converts an input image signal into an image signal of another color space;
The image processing apparatus according to claim 1, wherein the edge extraction unit extracts edge information using the input image signal color space converted by the color space conversion unit as an input. A color space conversion unit that converts an input image signal into an image signal of another color space;
The image processing apparatus according to claim 1, wherein the noise removal processing unit removes noise using the input image signal color-space converted by the color space conversion unit as an input. 9. A color space reverse conversion unit that reversely converts an image signal from which noise has been removed by the noise removal processing unit into the same color space as an input image signal that has been color space converted by the color space conversion unit. Image processing apparatus. An edge extraction step for extracting edge information related to the presence or absence of an edge around the target pixel of the input image signal;
Based on the specific color area information, it is determined whether the color signal of the target pixel is within the specific color area or outside the specific color area, and the target pixel is used by using the determination result and the edge information extracted by the edge extraction step. A noise removal selection step for selecting a noise removal process to be applied;
An image processing method comprising: a noise removal processing step of applying the noise removal processing selected in the noise removal selection step to the target pixel. The noise removal selection step includes:
A specific color region determination step for determining whether the color signal of the pixel of interest is within the specific color region or outside the specific color region based on the specific color region information;
This specific color area determination step determines that the specific color area is within the specific color area, and uses the edge information extracted by the edge extraction step to select specific color noise removal for a pixel of interest that does not have an edge in the vicinity. The image processing method according to claim 10, further comprising: a noise removal determination step of selecting normal noise removal for the target pixel. The noise removal selection step includes:
A specific color determination range for the pixel of interest is set based on the edge information extracted in the edge extraction step, and a statistical value of color signals of surrounding pixels in the specific color determination range is specified based on the specific color region information A specific color region determination step for determining whether the color region is inside or outside the specific color region;
This specific color area determination step determines that the specific color area is within the specific color area, and uses the edge information extracted by the edge extraction step to select specific color noise removal for a pixel of interest that does not have an edge in the vicinity. The image processing method according to claim 10, further comprising: a noise removal determination step of selecting normal noise removal for the target pixel. The edge extraction step includes
An edge information storage step of storing the extracted edge information as peripheral pixel information;
In the specific color region determination step, the peripheral pixel and the target pixel corresponding to peripheral pixel information indicating that no edge exists in the specific color determination range based on the peripheral pixel information stored in the edge information storage step The image processing method according to claim 12, wherein a statistical value is calculated from the color signal.

Images