JP2009231965A - Image processor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processor that converts an original image in YCC422 format into an expanded image in YCC444 format without generating a color blur nearby a border of colors. <P>SOLUTION: The image processor includes a processing object pixel extraction unit which extracts two pixels allocated with common color difference data on the original image as pixels to be processed, a luminance change detection unit which determines a difference value of luminance between pixels in the pixels to be processed as a first difference, a first comparison unit which compares the first difference and a first threshold with each other, and a color difference data conversion unit which generates two color difference data to be newly allocated to the respective pixels in the pixels to be processed on the basis of the comparison result of the first comparison unit and color difference data between a front pixel and a rear pixel adjacent to each other in the array direction of the respective pixels in the pixels to be processed on the original image. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an image processing apparatus. More specifically, the present invention converts an original image composed of luminance data for each pixel and color difference data for every two pixels into an expanded image composed of luminance data and color difference data for each pixel. The present invention relates to an improvement of an output image processing apparatus.

  There is a YCbCr422 format as a format of image data used for color video signals such as digital television broadcasting. The YCbCr422 format is known as a format with little loss of color information when data is compressed from the RGB format to the YCbCr422 format. The image data in the YCbCr422 format is composed of a luminance (Y) component and a color difference (Cb and Cr) component, and for each 2 × 2 = 4 pixels adjacent to each other on the image, four Y data and two Cb respectively. Data and Cr data. On the other hand, image data in the RGB format is composed of R (red), G (green), and B (blue) color components, and is composed of one piece of luminance data for each pixel. In general, the amount of data per pixel can be suppressed by converting RGB format image data obtained by an image sensor such as a charge coupled device (CCD) into image data in the YCbCr422 format.

  However, when RGB format image data is converted to YCbCr422 format image data, the Y component changes every pixel, whereas the Cb and Cr components both change every 2 pixels. It becomes. For this reason, a shift of one pixel may occur between the change point where the Y component changes greatly and the change point where the Cb or Cr component changes greatly. As described above, when a deviation occurs between the change point of the Y component and the change point of the Cb and Cr components, the change point of the Y component and the change point of the Cb and Cr components may not necessarily coincide depending on the color boundary. Will not. Normally, when YCbCr422 format image data is input to an output device such as a display device, an image processing device such as an RGB converter is used to convert the YCbCr422 format image data to RGB format image data and decompress the image. Is done. In such a conventional image processing apparatus, when YCbCr422 format image data is converted into RGB format image data, color blur occurs near the color boundary having a large degree of change due to the above-described reason. There was a problem that would stand out.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to provide an image processing apparatus capable of suppressing the occurrence of color bleeding near the boundary between colors having a large degree of change. . In particular, an object of the present invention is to provide an image processing apparatus capable of converting an original image in YCbCr422 format into a decompressed image in YCbCr444 format without causing color blurring in the vicinity of the color boundary.

  The image processing apparatus according to the first aspect of the present invention converts an original image composed of luminance data for each pixel and color difference data for each two pixels into an expanded image composed of luminance data and color difference data for each pixel and outputs the image. A processing target pixel extraction unit that extracts two adjacent pixels assigned common color difference data on the original image as processing target pixels, and brightness of each pixel in the processing target pixel A luminance change detection unit that obtains a difference value as a first difference, a first comparison unit that compares the first difference and the first threshold, a comparison result by the first comparison unit, and an adjacent pixel adjacent to the processing target pixel Of these, based on the color difference data of the previous pixel and the subsequent pixel adjacent to each other in the arrangement direction of each pixel in the processing target pixel on the original image. Te, and and a color difference data conversion unit for generating the two color difference data newly assigned to each pixel in the processing target pixel.

  In this image processing apparatus, two pixels to which common color difference data is assigned are extracted as processing target pixels, and a luminance difference value between each pixel in the processing target pixels is obtained. Then, based on the comparison result between the difference value and the first threshold value, and the color difference data of the previous pixel and the subsequent pixel adjacent to each other in the arrangement direction of each pixel in the processing target pixel on the original image, Two color difference data newly assigned to the pixel are generated. According to such a configuration, since two color difference data to be newly assigned are generated based on the comparison result between the luminance difference value and the first threshold value, when the luminance is greatly changed for the processing target pixel, The newly assigned color difference data can be made different from the original color difference data. In addition, since the above two color difference data are generated based on the color difference data of the previous pixel and the subsequent pixel adjacent to the processing target pixel, when making the newly assigned color difference data different from the original color difference data, An appropriate color difference according to the color difference can be assigned.

  In addition to the above configuration, the image processing apparatus according to the second aspect of the present invention obtains a difference value of a color difference between the previous pixel and the processing target pixel as a second difference, and calculates a color difference between the rear pixel and the processing target pixel. The color difference change detection unit that obtains the difference value as the third difference and the larger difference value of the second difference and the third difference are compared with the second threshold value, and the smaller difference value and the second threshold value are compared. A second comparison unit that compares a small third threshold value, and the color difference data conversion unit generates the two color difference data based on the comparison results of the first comparison unit and the second comparison unit. Is done. According to such a configuration, the color difference data to be newly assigned is generated for the processing target pixel depending on whether or not the luminance has changed greatly and whether or not the color difference has changed greatly between the preceding and following pixels. be able to.

  In addition to the above-described configuration, the image processing apparatus according to a third aspect of the present invention is configured such that the color difference data conversion unit includes the color difference data assigned to the pixel to be processed, and the pixel to be processed among the previous pixel and the rear pixel. The two color difference data are generated from the color difference data of adjacent pixels having a large difference value of the color difference between them. According to such a configuration, color difference data to be newly assigned is generated from the color difference data of an adjacent pixel having a large change in color difference and the original color difference data for the processing target pixel, so that the change in color difference near the luminance change point is generated. Can be effectively smoothed.

  In the image processing apparatus according to the fourth aspect of the present invention, in addition to the above configuration, the color difference data conversion unit includes a first difference that exceeds the first threshold value, and a larger difference value that exceeds the second threshold value. In addition, when the smaller difference value is equal to or smaller than the third threshold value, data different from the color difference data assigned to the processing target pixel is generated as the two color difference data. . According to such a configuration, when the luminance of the processing target pixel is greatly changed and the color difference is greatly changed only for one of the preceding and succeeding pixels, the newly assigned color difference data is changed to the original color difference data. Can be different. As a result, the color difference change in the vicinity of the luminance change point can be smoothed, so that it is possible to suppress the occurrence of color blurring near the color boundary where the degree of change is large.

  In addition to the above-described configuration, the image processing apparatus according to the fifth aspect of the present invention is configured such that one of the two color difference data is the original color difference data assigned to the processing target pixel. According to such a configuration, when data different from the original color difference data is assigned to the processing target pixel, one of the two newly assigned color difference data is the original color difference data. It is possible to suppress the deviation of the pixel from becoming larger as the pixel is processed later.

  In addition to the above-described configuration, the image processing apparatus according to a sixth aspect of the present invention is configured such that when the color difference data conversion unit assigns the two color difference data to each pixel in the processing target pixel, Of these, the original color difference data is assigned to a pixel adjacent to an adjacent pixel having a small color difference value with respect to the processing target pixel. According to such a configuration, since the original color difference data is assigned to a pixel adjacent to an adjacent pixel having a small change in color difference for the processing target pixel, the change in color difference near the luminance change point is effectively smoothed. be able to.

  According to the image processing apparatus of the present invention, the color difference data to be newly assigned according to whether or not the luminance is greatly changed for the processing target pixel and whether or not the color difference is greatly changed between the preceding and following pixels. Can be generated. In particular, when the luminance of the pixel to be processed is greatly changed and the color difference is greatly changed only with respect to one of the preceding and succeeding pixels, the newly assigned color difference data can be made different from the original color difference data. . As a result, the color difference change in the vicinity of the luminance change point can be smoothed, so that it is possible to suppress the occurrence of color blurring near the color boundary where the degree of change is large. Therefore, it is possible to realize an image processing apparatus that can convert an original image in the YCbCr422 format into a decompressed image in the YCbCr444 format without causing color blurring in the vicinity of the color boundary.

  FIG. 1 is a system diagram showing a configuration example of an image processing system 1 including an image processing apparatus according to an embodiment of the present invention. The image processing system 1 includes a digital video camera 2, an image processing device 3, a resolution change processing unit 4, and an output device 5, and a video signal photographed by the digital video camera 2 is visualized by the output device 5.

  The digital video camera 2 is a photographing device that photographs a subject and outputs a video signal, and includes a CCD 21 and a color space conversion unit 22. The CCD 21 is an image sensor including a large number of light receiving elements, and brightness data for each of R, G, and B color components is generated as a pixel signal corresponding to the amount of received light.

  The RGB format image data is converted into YCbCr422 format image data by the color space conversion unit 22 and output as a video signal. Here, it is assumed that a moving image having a predetermined frame rate is output as the video signal.

  The video signal in YCbCr422 format input to the image processing device 3 is not limited to the output of the digital video camera 2, and for example, any general composite or component video signal such as a DVD player is assumed. Further, it is assumed that the YCbCr422 format video signal input to the image processing device 3 is quantized from the analog YCbCr422 format to the digital YCbCr422 format via a video decoder or the like as necessary.

  The image processing device 3 is a format conversion device that converts an original image composed of luminance data for each pixel and color difference data for every two pixels into an expanded image composed of luminance data and color difference data for each pixel and outputs the converted image. In the image processing apparatus 3, an operation is performed in which the video signal from the color space conversion unit 22 is processed in time series as an original image, and image data in YCbCr444 or RGB format is generated as an expanded image.

  Here, the image data in the YCbCr444 format is image data composed of four Y data and four Cb data and Cr data, respectively, per 2 × 2 = 4 pixels adjacent to each other on the image. .

  The output device 5 is a device for visualizing the video signal, and the image data from the image processing device 3 and the resolution change processing unit 4 is processed. As such an output device 5, a display device that displays an image on a screen, a printer that prints a still image on paper, or the like is used.

  The resolution change processing unit 4 is a processing device for enlarging or reducing an image as necessary, and has been converted into YCbCr444 or RGB format by the image processing device so as not to deteriorate the image quality due to enlargement and reduction. Later image data is processed. Since both the enlargement process and the reduction process are processes for changing the number of pixels per unit length, the resolution can be changed by these processes.

  FIG. 2 is a block diagram illustrating a configuration example of the image processing apparatus 3 of FIG. The image processing device 3 includes a processing target pixel extraction unit 11, a luminance change detection unit 12, threshold storage units 13 and 16, a first comparison unit 14, a color difference change detection unit 15, a second comparison unit 17, and a color difference data conversion unit 18. Consists of.

  The processing target pixel extraction unit 11 performs an operation of extracting processing target pixels from the original image in order to convert the format of the image data. This processing target pixel is two pixels adjacent to each other to which common color difference data is assigned on the original image. In the processing target pixel extraction unit 11, image data in YCbCr422 format input as a video signal is processed in time series as an original image, and the extracted processing target pixel information is used as the luminance change detection unit 12, the color difference change detection unit 15, and the color difference data. The data is output to the conversion unit 18.

  The luminance change detection unit 12 performs an operation of obtaining a luminance difference value (change amount) between the pixels in the processing target pixel as the first difference. The first comparison unit 14 compares the first difference with the first threshold value stored in the threshold value storage unit 13 and outputs the comparison result to the color difference data conversion unit 18.

  The color difference change detection unit 15 performs an operation of obtaining a difference value (change amount) of a color difference between the previous pixel and the processing target pixel as a second difference among adjacent pixels adjacent to the processing target pixel. In addition, the color difference change detection unit 15 performs an operation of obtaining a difference value of the color difference between the subsequent pixel and the processing target pixel among the adjacent pixels adjacent to the processing target pixel as the third difference.

  The front pixel and the rear pixel are two pixels adjacent in the arrangement direction of each pixel in the processing target pixel on the original image. For example, the front pixel is an adjacent pixel processed immediately before the processing target pixel, and the rear pixel is an adjacent pixel processed immediately after the processing target pixel.

  The second comparison unit 17 compares the larger difference value of the second difference and the third difference with the second threshold value, and compares the smaller difference value with a third threshold value smaller than the second threshold value. The comparison result is output to the color difference data conversion unit 18. Both the second threshold value and the third threshold value are stored in advance in the threshold value storage unit 16.

  The color difference data conversion unit 18 generates two color difference data to be newly assigned to each pixel in the processing target pixel based on the comparison result by the first comparison unit 14 and the second comparison unit 17, and as the color difference component of the YCbCr444 data. The operation to output is performed. The two color difference data newly assigned to the processing target pixel are determined based on the original color difference data assigned to the processing target pixel and the color difference data of one of the previous pixel and the subsequent pixel. Note that the luminance component of the YCbCr422 data is output as it is as the luminance component of the YCbCr444 data.

  Conversion of image data from the YCbCr422 format to the YCbCr444 format is performed by adjusting the color difference component when the change point of the Y component and the change point of the Cb and Cr components are different at the boundary of the color where the degree of change is large. Is called.

The adjustment of the color difference component is performed according to the following rules (a) to (d).
(A) The Y component is not adjusted and the input data is output as it is as output data.
(B) Adjustment is performed for the color difference component, and the Cb component and the Cr component are individually adjusted independently.
(C) Whether or not to perform the adjustment is determined based on the relationship between the change in the Y component and the change in the color difference component.
(D) When adjustment is performed, color difference data different from the original color difference data is generated as two color difference data newly assigned to the processing target pixel. At this time, the original color difference data, that is, the original data is left as it is in one of the two newly assigned color difference data.

  Specifically, the first difference is greater than the first threshold, and the larger difference value of the second difference and the third difference is greater than the second threshold, and the smaller difference value is When the value is equal to or smaller than the third threshold, color difference data different from the color difference data assigned to the processing target pixel is generated. On the other hand, when the first difference is less than or equal to the first threshold and when the first difference is greater than the first threshold, the larger difference value of the second difference and the third difference is less than or equal to the second threshold. Alternatively, if the smaller difference value exceeds the third threshold value, the same color difference data as the original color difference data is generated without adjustment.

  Further, when adjustment is performed, two color difference data based on the color difference data of an adjacent pixel having a large difference value of the color difference between the previous pixel and the rear pixel and the original color difference data. Is determined. That is, the original color difference data is used as it is for one of the two color difference data newly assigned to the processing target pixel. On the other hand, the other color difference data of the two color difference data is obtained from the color difference data of the adjacent pixel having a large color difference value between the previous pixel and the subsequent pixel and the original color difference data. Determined. For example, the average value of the color difference between adjacent pixels and the original color difference is output as the other color difference data of the two color difference data.

  In addition, when assigning two color difference data to each pixel extracted as the processing target pixel, the pixel adjacent to the adjacent pixel having a large color difference difference value between the processing target pixel and the previous pixel and the subsequent pixel. On the other hand, color difference data different from the original color difference data is assigned. That is, the color difference data different from the original color difference data, here, the average value of the color difference of the adjacent pixels and the original color difference is assigned to the pixel on the adjacent pixel side where the difference value of the color difference is large. On the other hand, the original color difference data is assigned to a pixel adjacent to an adjacent pixel having a small color difference difference value between the previous pixel and the rear pixel.

  FIG. 3 is a diagram showing an example of the operation in the image processing apparatus 3 of FIG. 2, and shows the flow of image data in the YCbCr422 format to be processed. The Y component data is data that changes for each pixel. For example, data from Y0 to Y11 is processed in time order in this order.

  On the other hand, the color difference data of the Cb and Cr components is data that changes every two pixels, and one color difference data corresponds to two Y component data.

  When such image data in the YCbCr422 format is converted into image data in the YCbCr444 format, two pixels to which common color difference data is assigned are extracted as processing targets. Then, for the extracted processing target pixel, in order to determine whether or not to adjust the color difference data, the Y component data and the color difference data of the processing target pixel and the adjacent pixels (the previous pixel and the rear pixel) that are adjacent to the processing target pixel. Pixel) color difference data. That is, a 4-pixel section composed of the processing target pixel, the previous pixel, and the rear pixel is an adjustment section for color difference data.

  The image data conversion process is performed while shifting such an adjustment interval by two pixels each time the processing target pixel is changed. However, for the first pixel and the last pixel of the original image, there is no previous pixel or subsequent pixel. Therefore, only when the first pixel or the last pixel is included in the processing target pixel, the first pixel or the last pixel continues. Treat it as

  FIG. 4 is a diagram showing an example of the operation in the image processing apparatus 3 of FIG. 2, and shows how image data is converted from the YCbCr422 format to the YCbCr444 format. When the nth and (n + 1) th pixels are pixels to be processed, the (n−1) th pixel is the previous pixel and the (n + 2) th pixel is the subsequent pixel. The (n−1) th to (n + 2) th pixels are adjustment intervals.

  In the image data conversion processing, input data is output as output data as it is for Y component data, whereas two color difference data newly assigned to each pixel of the processing target pixel is generated for color difference data. . In this example, color difference data is newly assigned to the nth and (n + 1) th pixels to which common color difference data Cb and Cr are assigned.

  When adjusting the color difference component, one of the two color difference data newly assigned to the processing target pixel is the original data, that is, the original color difference data, and the other is the color difference data different from the original color difference data. Adjustments are made. On the other hand, when the color difference component is not adjusted, the two color difference data newly assigned to the processing target pixel are used as the original color difference data, and the original color difference data is directly assigned to each pixel of the processing target pixel.

  Which of the two pixels extracted as the processing target pixel is assigned with the original data is determined based on a change amount of the color difference between the processing target pixel and the adjacent pixels before and after the processing target pixel.

Whether or not to adjust the color difference component is determined by the following conditions (1) and (2).
(1) There is a change in luminance exceeding a certain value (first threshold value) between the pixels to be processed.
(2) Of the amount of change in color difference between the pixel to be processed and the adjacent pixels before and after, the larger amount of change (difference value) exceeds a certain value (second threshold), and the smaller amount of change is Must be below a certain value (third threshold).

  When both of these conditions (1) and (2) are satisfied, the color difference component is adjusted. As the first threshold, for example, the Y component is ITU-R BT. According to the international standard of 601, when quantized as an integer in the range of 16 to 235, or expanded to an 8-bit full range and quantized as an integer in the range of 0 to 255, It is set to about 25.

  Moreover, as said 2nd threshold value and 3rd threshold value, for example, both Cb and Cr components are ITU-R BT. According to the international standard of 601, when quantized as an integer in the range of 16 to 240, or expanded to an 8-bit full range and quantized as an integer in the range of 0 to 255, The second threshold is set to about 40, and the third threshold is set to about 10. However, the setting values of the first to third threshold values can be adjusted as necessary.

  5 to 10 are diagrams showing an example of the operation in the image processing apparatus 3 of FIG. 2, and a specific example of the image data conversion process is shown. FIG. 5 shows a case where the amount of change in luminance (difference value) in the processing target pixel is equal to or less than the first threshold value. In this example, for the Y component of the processing target pixel, the luminance values of the nth pixel and the (n + 1) th pixel are both 100, and the luminance change amount is 0.

  In such a case, the color difference components (Cb and Cr) are not adjusted because the condition (1) is not satisfied, the change in luminance is small, and the color boundary is considered inconspicuous.

  FIG. 6 shows a case where the amount of change in luminance at the processing target pixel exceeds the first threshold, but the color difference change is small before and after. In this example, for the Y component of the pixel to be processed, the luminance value of the nth pixel is 50, and the luminance value of the (n + 1) th pixel is 150. Therefore, the amount of change in luminance is 100 for the processing target pixel.

  However, since the change amount of the color difference between the processing target pixel and the adjacent pixels before and after the pixel is both 0, the change amount of the color difference before and after the processing target pixel is small and does not satisfy the condition (2). No component adjustments are made.

  FIG. 7 shows a case where the amount of change in luminance at the processing target pixel exceeds the first threshold, but the amount of change in color difference is the same before and after. In this example, for the Y component of the pixel to be processed, the luminance value of the nth pixel is 50, and the luminance value of the (n + 1) th pixel is 150. Therefore, the amount of change in luminance is 100 for the processing target pixel.

  However, the amount of change in color difference between the pixel to be processed and the adjacent pixels before and after is 25. Specifically, the color difference value is 100 for the (n-2) th and (n-1) th pixels, and the color difference value is 125 for the nth and (n + 1) th pixels. Further, the color difference value is 150 for the (n + 2) and (n + 3) th pixels, and the amount of change in the color difference between the adjacent pixels before and after is 25. For this reason, the change amount of the color difference is the same before and after the pixel to be processed, and the condition (2) is not satisfied, so the color difference component is not adjusted.

  FIG. 8 shows another example in which the amount of change in luminance at the pixel to be processed exceeds the first threshold, but the amount of change in color difference is the same before and after. In this example, for the Y component of the pixel to be processed, the luminance value of the nth pixel is 50, and the luminance value of the (n + 1) th pixel is 150. Therefore, the amount of change in luminance is 100 for the processing target pixel.

  However, the amount of change in color difference between the pixel to be processed and the adjacent pixels before and after is 50. Specifically, the color difference value is 100 for the (n-2) th and (n-1) th pixels, and the color difference value is 50 for the nth and (n + 1) th pixels. In addition, the color difference value is 100 for the (n + 2) and (n + 3) th pixels, and the amount of change in the color difference between the adjacent pixels before and after is 50. For this reason, the change amount of the color difference is the same before and after the pixel to be processed, and the condition (2) is not satisfied, so the color difference component is not adjusted.

  In the cases as shown in FIGS. 6 to 7, since the change in color difference is stepwise at regular intervals, it is considered that the color boundary is inconspicuous. Further, in the case as shown in FIG. 8, since it is considered that an outline exists at the boundary of the color, in these cases, the color difference component is not adjusted.

  FIG. 9 shows a case where the amount of change in luminance at the processing target pixel exceeds the first threshold and the amount of change in color difference satisfies the condition (2). In this example, for the Y component of the pixel to be processed, the luminance value of the nth pixel is 50, and the luminance value of the (n + 1) th pixel is 150. Therefore, the amount of change in luminance is 100 for the processing target pixel.

  In addition, the amount of change in color difference between the processing target pixel and the adjacent pixels before and after the pixel is 0 on one side and 50 on the other side. Specifically, the color difference value is 100 for the (n-2) th and (n-1) th pixels, and the color difference value is 100 for the nth and (n + 1) th pixels. The color difference value is 150 for the (n + 2) and (n + 3) th pixels, and the change amount of the color difference from the previous pixel is 0, whereas the change amount of the color difference from the subsequent pixel is 50. It has become.

  In such a case, since the conditions (1) and (2) are satisfied, the color difference component is adjusted. When the color difference component is adjusted, original data is assigned to one of the two pixels to be processed. On the other hand, the average value of the color difference value of the adjacent pixel and the color difference value of the processing target pixel, which have a large difference value of the color difference between the previous pixel and the rear pixel, is assigned to the other pixel.

  The pixel to which the average value of the color difference value of the adjacent pixel and the color difference value of the processing target pixel is assigned is a pixel adjacent to the adjacent pixel having a large color difference value between the previous pixel and the subsequent pixel. ing.

  That is, the color difference value of the processing target pixel is compared with the color difference values of the adjacent pixels before and after, and it is determined which of the adjacent pixels before and after the color difference value of the processing target pixel is close. If the color difference value of the previous pixel is close, the original data is assigned to the processing target pixel adjacent to the previous pixel, and the average value is assigned to the processing target pixel adjacent to the subsequent pixel. On the other hand, when it is close to the color difference value of the subsequent pixel, the original data is assigned to the processing target pixel adjacent to the subsequent pixel, and the average value is assigned to the processing target pixel adjacent to the previous pixel.

  In FIG. 9, the color difference value of the processing target pixel is close to the color difference value of the previous pixel, the original data 100 is assigned to the processing target pixel adjacent to the previous pixel, and the processing target pixel adjacent to the subsequent pixel is assigned. Is assigned an average value of 125.

  FIG. 10 shows another example in which the amount of change in luminance at the processing target pixel exceeds the first threshold and the amount of change in color difference satisfies the condition (2). In this example, for the Y component of the pixel to be processed, the luminance value of the nth pixel is 50, and the luminance value of the (n + 1) th pixel is 150. Therefore, the amount of change in luminance is 100 for the processing target pixel.

  Moreover, the amount of change in the color difference between the processing target pixel and the adjacent pixels before and after is 50 for one and 0 for the other. Specifically, the color difference value is 100 for the (n-2) th and (n-1) th pixels, and the color difference value is 150 for the nth and (n + 1) th pixels. Further, the color difference value is 150 for the (n + 2) and (n + 3) th pixels, and the change amount of the color difference from the previous pixel is 50, whereas the change amount of the color difference from the subsequent pixel is 0. It has become.

  In such a case, since the conditions (1) and (2) are satisfied, the color difference component is adjusted. When the color difference component is adjusted, original data is assigned to one of the two pixels to be processed. On the other hand, the average value of the color difference value of the adjacent pixel and the color difference value of the processing target pixel, which have a large difference value of the color difference between the previous pixel and the rear pixel, is assigned to the other pixel.

  The pixel to which the average value of the color difference value of the adjacent pixel and the color difference value of the processing target pixel is assigned is a pixel adjacent to the adjacent pixel having a large color difference value between the previous pixel and the subsequent pixel. ing.

  In FIG. 10, the color difference value of the processing target pixel is close to the color difference value of the subsequent pixel, the original data 150 is assigned to the processing target pixel adjacent to the subsequent pixel, and the processing target pixel adjacent to the previous pixel is assigned. Is assigned an average value of 125.

  According to the present embodiment, the color difference data to be newly assigned is generated for the processing target pixel depending on whether or not the luminance has changed greatly and whether or not the color difference has changed greatly between the previous and subsequent pixels. be able to. In particular, when the luminance of the pixel to be processed is greatly changed and the color difference is greatly changed only with respect to one of the preceding and succeeding pixels, the newly assigned color difference data can be made different from the original color difference data. . As a result, the color difference change in the vicinity of the luminance change point can be smoothed, so that it is possible to suppress the occurrence of color blurring near the color boundary where the degree of change is large.

  Further, when data different from the original color difference data is assigned to the processing target pixel, since one of the two newly assigned color difference data is the original color difference data, the deviation from the color difference data in the original image is processed later. It is possible to suppress an increase in the size of a pixel to be made.

  In this embodiment, it is determined whether or not to adjust the color difference component by comparing the change amount of the color difference between the processing target pixel and the adjacent pixel with the threshold value for each of the previous pixel and the subsequent pixel. However, the present invention is not limited to this example. For example, the change amount of the color difference between the processing target pixel and the adjacent pixels before and after is directly compared between the previous pixel and the subsequent pixel, and it is determined whether or not to adjust the color difference component based on the comparison result. Also good.

1 is a system diagram showing a configuration example of an image processing system 1 including an image processing apparatus according to an embodiment of the present invention. FIG. 2 is a block diagram illustrating a configuration example of the image processing apparatus 3 in FIG. 1. FIG. 3 is a diagram illustrating an example of an operation in the image processing apparatus 3 in FIG. 2, and illustrates a flow of image data in a YCbCr422 format to be processed. FIG. 4 is a diagram illustrating an example of an operation in the image processing apparatus 3 in FIG. 2, and illustrates how image data is converted into the YCbCr444 format. FIG. 3 is a diagram illustrating an example of an operation in the image processing apparatus 3 in FIG. 2, and illustrates a case where the amount of change in luminance (difference value) in the processing target pixel is equal to or less than a first threshold value. FIG. 3 is a diagram illustrating an example of an operation in the image processing apparatus 3 of FIG. 2, in which the amount of change in luminance exceeds a first threshold value but the change in color difference between before and after is small. FIG. 3 is a diagram illustrating an example of an operation in the image processing apparatus 3 of FIG. 2, in which the change amount of luminance exceeds the first threshold value, but the change amount of color difference is the same before and after. It is the figure which showed an example of the operation | movement in the image processing apparatus 3 of FIG. 2, and the other example in case the variation | change_quantity of a color difference is the same before and behind is shown. FIG. 5 is a diagram illustrating an example of an operation in the image processing apparatus 3 of FIG. 2, and illustrates a case where the amount of change in luminance exceeds the first threshold and the amount of change in color difference satisfies the condition (2). It is a figure showing an example of operation in image processing device 3 of Drawing 2, and other examples in case the amount of change of color difference fulfills condition (2) are shown.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Image processing system 2 Digital video camera 3 Image processing apparatus 4 Resolution change process part 5 Output device 11 Processing object pixel extraction part 12 Luminance change detection part 13, 16 Threshold storage part 14 1st comparison part 15 Color difference change detection part 17 2nd Comparison unit 18 Color difference data conversion unit

Claims (6)

In an image processing apparatus for converting an original image composed of luminance data for each pixel and color difference data for every two pixels into an expanded image composed of luminance data and color difference data for each pixel and outputting the same,
A processing target pixel extraction unit that extracts two adjacent pixels assigned common color difference data on the original image as processing target pixels;
A luminance change detection unit that obtains a luminance difference value between the pixels in the processing target pixel as a first difference;
A first comparison unit for comparing the first difference and the first threshold;
The comparison result by the first comparison unit and the color difference data of the previous pixel and the subsequent pixel adjacent in the arrangement direction of each pixel in the processing target pixel on the original image among the adjacent pixels adjacent to the processing target pixel. An image processing apparatus comprising: a color difference data conversion unit that generates two color difference data newly assigned to each pixel in the processing target pixel. A color difference change detection unit that obtains a difference value of a color difference between the previous pixel and the processing target pixel as a second difference and obtains a difference value of the color difference between the rear pixel and the processing target pixel as a third difference;
Of the second difference and the third difference, a larger difference value and the second threshold value are compared, and a second comparison unit that compares the smaller difference value and the third threshold value that is smaller than the second threshold value. Prepared,
The image processing apparatus according to claim 1, wherein the color difference data conversion unit generates the two color difference data based on a comparison result by the first comparison unit and the second comparison unit.   The color difference data conversion unit includes color difference data allocated to the processing target pixel, and color difference data of an adjacent pixel having a large color difference difference value between the processing target pixel and the previous pixel and the rear pixel. The image processing apparatus according to claim 1, wherein the two color difference data are generated from the image data.   In the color difference data converting unit, the first difference exceeds the first threshold, the larger difference value exceeds the second threshold, and the smaller difference value becomes equal to or less than the third threshold. 3. The image processing apparatus according to claim 2, wherein, when the data is different from the color difference data assigned to the processing target pixel, the image processing apparatus generates data as the two color difference data.   The image processing apparatus according to claim 1, wherein one of the two color difference data is original color difference data assigned to the processing target pixel.   When assigning the two color difference data to each pixel in the processing target pixel, the color difference data conversion unit has a small color difference difference value between the processing target pixel and the previous pixel and the rear pixel. 6. The image processing apparatus according to claim 5, wherein the original color difference data is assigned to a pixel adjacent to an adjacent pixel.

Images