JP2005260902A - Image feature value detection apparatus, image quality improving apparatus, display device, and receiver - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image feature value detection apparatus, in which the image feature value of an image to be encoded and decoded is detected and image quality is enhanced using the detected image feature value, and image quality improving apparatus, display device, and receiver. <P>SOLUTION: The image feature value detection apparatus is provided with: a optical signal level averaging means for converting an inputted image signal from an electric signal level to an optical signal level and averaging the image signal of an optical signal level; an electric signal level averaging means for averaging the image signal of an inputted electric signal level; and an output means for converting the image signal of one of the averaged optical signal level and electric signal level into the image signal of the other electric signal level or optical signal level and outputting a difference as an image feature value for at least one pixel. The image quality improving apparatus is provided with a correction means for correcting the inputted image signal in accordance with the outputted image feature value. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an image feature amount detection device, an image quality improvement device, a display device, and a receiver, and in particular, an image feature amount detection device that detects an image feature amount of an image to be encoded and decoded, and uses the detected image feature amount The present invention relates to an image quality improvement device, a display device, and a receiver that improve image quality.

  Conventionally, in encoding processing for reducing the amount of information of a moving image signal, an intra-screen encoded image encoded in the same frame or the same field for each frame or field, and a frame from a past image signal in time. Or, it is classified and coded into three types of images: a forward prediction encoded image that performs prediction between fields, and a bidirectional predictive encoded image that performs prediction between frames or fields from temporally past and future image signals. Is going on.

  It should be noted that the in-screen encoded image, the forward prediction encoded image, and the bidirectional predictive encoded image are MPEG-2 (ISO / IEC 13818) video well known as a moving image encoding method such as BS digital broadcasting. In the system, they are called I picture (Intra coded picture), P picture (Predictive coded picture) and B picture (Bidirectionally predictive coded picture). Therefore, in the following description, the intra-picture encoded image, the forward prediction encoded image, and the bidirectional predictive encoded image are referred to as I picture, P picture, and B picture.

  The I picture can be decoded without referring to other pictures, and is periodically arranged as a point corresponding to random access or an error recovery point. However, the coding efficiency decreases as the frequency of I pictures increases. A P picture is an image for which predictive encoding is performed from an I picture or a P picture located in the past in time.

  The arrangement of the I picture, P picture, and B picture constituting the encoded moving image (hereinafter referred to as the encoded moving image) is, for example, as shown in FIG. FIG. 1 is a diagram illustrating an example of an arrangement of encoded moving images. FIG. 1A is an example of an image arrangement of an I picture and a P picture. N in FIG. 1A is a picture interval of an I picture. M is a picture interval between the I picture and the P picture.

  In addition, by introducing a B picture that performs predictive coding from an I picture or a P picture located in the past and the future in time, high coding efficiency can be obtained even in a screen region where a hidden object appears. . FIG. 1B is an example of an image arrangement of an I picture, a P picture, and a B picture. N in FIG. 1B is a picture interval of the I picture. M is the picture interval of the P picture.

  The B picture is predictively encoded using a reference image obtained by interpolating past and future images in time during bidirectional prediction. Therefore, the B picture is predicted with higher accuracy than the P picture predicted from only one image, and the degree of distortion is reduced even with the same amount of information. Further, the B picture is not used for prediction of other pictures. In an encoded moving image, efficient information amount control is performed in which the amount of code assigned to a B picture is reduced and the amount of code assigned to an I picture or P picture is increased accordingly.

  On the other hand, due to the reduction of the spatial resolution of the predicted image due to the interpolation processing and the reduction of the allocated code amount, the high-frequency component of the image is not transmitted in the B picture, and the spatial resolution is lower than that of other pictures. is there.

  In general, a moving image is encoded in a configuration including bidirectional prediction in order to obtain high encoding efficiency. However, since the spatial resolution of a B picture is relatively low in an encoded moving image, flicker is perceived and disturbed in some images (see, for example, Non-Patent Document 1).

  This is because a display device such as a CRT has a gamma characteristic, and the relationship between the input electric signal level and the output optical signal level is non-linear. Even if the average value is constant at the electrical signal level, the average value does not always become constant at the optical signal level due to the change in the frequency characteristics of the image signal, and the change in luminance between pictures is perceived as flicker. It is.

  FIG. 2 shows the principle of flicker generation. In FIG. 2, when the electrical signal level is converted from the electrical signal level to the optical signal level according to the electrical signal-optical signal conversion (hereinafter referred to as electro-optical conversion) characteristic, the average value at the optical signal level when there is a high-frequency component at the electrical signal level. There is a difference between the value of the optical signal level when the high frequency component is lost at the electric signal level. As a result, an image signal having a high frequency component has a higher average luminance on the display device.

  Therefore, the average luminance of a B picture with a low spatial resolution is lower than the average luminance of an I picture and a P picture with a high spatial resolution, resulting in a difference in brightness between the B picture, the I picture, and the P picture.

As a technique for detecting and reducing such flicker deterioration due to flicker, a level fluctuation detecting device and an image quality improving device are known (for example, Patent Document 1). In this method, an average level of each frame at the optical signal level is obtained, a difference between the frames is detected, and the difference is added and corrected for each pixel in one frame at the electric signal level.
JP-A-9-224250 Yoshiaki Shikaga, Katsunori Aoki, Eisuke Nakasu, “Verification of Bidirectional Prediction Effect by Video Coding—Evaluation and Analysis of Image Quality”, Television Society Journal, Vol. 50, no. 3, 1996, pp. 391-398

  However, flicker detection by the above-described level fluctuation detection device has a problem in that the original luminance difference of each frame is also subject to detection, and only the flicker component is not detected from the change in luminance between frames. Since the original luminance change of the moving image is not an interference, it is necessary to extract only the flicker component that becomes an interference.

  Further, the flicker detection by the level fluctuation detection device is performed in units of one screen, and there is a problem that it cannot be performed for each screen area constituting a part of the screen. Originally, flicker is caused by a picture, and is a disturbance occurring only in a screen area where the picture is presented. In other words, since the deterioration of image quality depends on the screen position where flicker occurs and its area, it is necessary to detect the screen area where flicker occurs in the screen.

  In addition, the flicker correction performed by the image quality improving apparatus is performed in units of one screen, and there is a problem that adaptive correction for each screen area cannot be performed. If adaptive correction is not performed only on the screen area where flicker occurs in the screen, other screen areas where flicker does not occur will also have the effect of correction, and the brightness that can be said to be reverse flicker in that screen area On the contrary, the image quality deteriorates.

  In addition, the flicker correction performed by the image quality improvement apparatus has a problem in that a change in the original brightness of the moving image, such as a scene change, a quick movement of the subject, or a camera flash shooting scene, is also subject to correction. . Unless only the flicker is corrected while leaving the original luminance change of the moving image, the luminance change is disturbed to a normal scene where flicker does not occur, and the image quality deteriorates.

  Note that the necessity of adaptive flicker correction according to the screen area and correction without changing the original luminance of the image is also described in Non-Patent Document 1 described above. The flicker correction by the image quality improvement apparatus is based on the assumption that I, P, and B picture configurations are known and picture type information is obtained from the decoding apparatus.

  Therefore, in order to realize the flicker correction by the image quality improvement apparatus, there is a problem that a modification for incorporating the image quality improvement apparatus into an existing decoding apparatus is necessary. As described above, the conventional level fluctuation detection device and image quality improvement device have not sufficiently performed flicker detection and flicker correction.

  The present invention has been made in view of the above points, and detects an image feature amount of an image to be encoded and decoded, and uses the detected image feature amount to improve an image feature amount detection apparatus and image quality improvement. An object is to provide a device, a display device, and a receiver.

  In order to solve the above-described problems, an image feature quantity detection device according to the present invention converts an input image signal from an electric signal level to an optical signal level according to a predetermined electric signal-optical signal conversion characteristic. Optical signal level averaging means for averaging image signals, electric signal level averaging means for averaging image signals of the inputted electric signal level, one of the averaged optical signal level or image signal of the electric signal level, and the other Output means for converting the image signal to an electrical signal level or optical signal level and outputting the difference as an image feature amount for each at least one pixel.

  The present invention also provides optical signal level averaging means for converting an input image signal from an electric signal level to an optical signal level in accordance with predetermined electric signal-optical signal conversion characteristics, and averaging the image signal at the optical signal level. An electric signal level averaging means for averaging the input image signal of the electric signal level, one of the averaged optical signal level or the image signal of the electric signal level, and the image signal of the other electric signal level or the optical signal level And an output unit that outputs the difference as an image feature amount for each at least one pixel, and a correction unit that corrects the input image signal in accordance with the output image feature amount. To do.

  A display device and a receiver according to the present invention include the above-described image quality improvement device.

  The image feature amount detection apparatus of the present invention outputs the difference between the image signal averaged at the optical signal level and the image signal averaged at the electrical signal level as an image feature amount, thereby changing the frequency characteristics between the screens. It can be detected. For example, it is possible to detect flicker caused by a change in frequency characteristics between screens. As a result, it is possible to detect image feature quantities such as flicker causing image quality degradation for each region, and objectively evaluate and quantify the range and size of image quality degradation.

  The image quality improvement apparatus of the present invention outputs the difference between the image signal averaged at the optical signal level and the image signal averaged at the electrical signal level as an image feature amount, and the image signal is output according to the image feature amount. By correcting, it is possible to reduce image quality degradation such as flicker degradation caused by changes in frequency characteristics between screens.

  In addition, the display device and the receiver of the present invention can output an image whose image quality is improved by the image quality improving device.

  According to the present invention, an image feature amount detection device, an image quality improvement device, a display device, and a receiver that detect an image feature amount of an image to be encoded and decoded and improve the image quality using the detected image feature amount are provided. Can be provided.

  Next, the best mode for carrying out the present invention will be described based on the following embodiments with reference to the drawings. In the following description, an image feature amount detection device that detects flicker as an example of an image feature amount, an image quality improvement device that improves image quality using the detected flicker, a display device, and a receiver will be described. Not limited to.

  The present invention is generally applicable when the frequency characteristics of each picture change due to various moving image processes. In the following description, the moving image processing uses MPEG-2 video as an example of a moving image encoding method. The moving image processing is intended for luminance signals or primary color signals.

  An image feature amount detection apparatus according to the present invention is used when detecting flicker deterioration caused by encoding and decoding processing of a moving image, and detects changes in frequency characteristics between images that are a cause of flicker occurrence. This is detected for each screen area, and the degree of flicker in the moving image and the generation range on the screen can be detected and quantified.

[First embodiment]

  FIG. 3 is a block diagram of the first embodiment of the image feature quantity detecting apparatus according to the present invention. The image feature amount detection apparatus includes a non-linear conversion unit 1, a narrow band LPF (Low pass filter) 2, a non-linear inverse conversion unit 3, a narrow band LPF 4 and a subtraction unit 5. A moving image signal at an electric signal level, which is an input signal to the image feature quantity detection device, is input to the nonlinear converter 1 and the narrowband LPF 4.

  When the moving image signal is input, the non-linear conversion unit 1 performs non-linear conversion on the level value for each pixel, and outputs a moving image signal having an optical signal level obtained by non-linear conversion of the level value for each pixel to the narrowband LPF 2. Here, the non-linear conversion means conversion according to the gamma characteristic which is the electro-optical conversion characteristic of the display device.

  When a moving image signal at an optical signal level is input, the narrowband LPF 2 performs a narrowband low-pass filtering process by a spatial convolution operation for each pixel. FIG. 4 is a configuration diagram of an example of a convolution mask used by the narrow band LPF 2.

  The convolution mask shown in FIG. 4 is composed of 9 × 9 grid cells, and each cell corresponds to a pixel. In the cells, multipliers corresponding to the respective pixels are described. The pixel that performs the spatial convolution operation is a pixel corresponding to the central cell of the convolution mask. The level value of the pixel subjected to the spatial convolution calculation is calculated by adding 81 values obtained by multiplying the level value of each pixel and a multiplier corresponding to the pixel, and dividing the value by 64.

  The narrowband LPF 2 outputs a moving image signal having an optical signal level subjected to narrowband low-pass filtering processing to the nonlinear inverse transform unit 3. When the moving image signal of the optical signal level is input, the nonlinear inverse converting unit 3 outputs the moving image signal of the electric signal level obtained by nonlinearly converting the level value for each pixel to the subtracting unit 5. Here, the nonlinear inverse transformation is a transformation having a relation of inverse function with the nonlinear transformation performed by the nonlinear transformation unit 1.

  On the other hand, when a moving image signal having an electric signal level is input to the narrowband LPF 4, a narrowband low-pass filtering process by a spatial convolution operation similar to the narrowband LPF 2 is performed for each pixel. The narrowband LPF 4 outputs a moving image signal having an electric signal level subjected to narrowband low-pass filtering processing to the subtracting unit 5.

  That is, the moving image signal input to the subtracting unit 5 through the non-linear conversion unit 1, the narrow band LPF 2, and the non-linear inverse conversion unit 3 is subjected to a narrow band low-pass filtering process by a spatial convolution operation at the optical signal level. It is a thing. The moving image signal input to the subtracting unit 5 through the narrow band LPF 4 is obtained by performing narrow band low-pass filtering processing by spatial convolution calculation at the electric signal level.

  The subtracting unit 5 subtracts, for each pixel, the moving image signal having the electric signal level input from the non-linear inverse converting unit 3 and the moving image signal having the electric signal level input from the narrowband LPF 4, and uses the difference as a flicker component. Output. Note that the flicker component output from the subtracting unit 5 corresponds to, for example, the difference between the average optical signal level when there is an AC component in FIG. 2 and the optical signal level when the AC component is lost.

  In the image feature amount detection apparatus of FIG. 3, by comparing flicker components of pixels having the same spatial position in successive frames or fields, the spatial position, in other words, the extent of occurrence of flicker in the pixel can be detected.

  FIG. 5 is a graph showing flicker components obtained at two points having different spatial positions in a moving image signal in which flicker is actually generated. In the graph of FIG. 5, the horizontal axis represents the frame number of successive frames, and the vertical axis represents the flicker component of each frame obtained at points A and B having different spatial positions. For example, in the graph of FIG. 5, flicker occurs at point A where the change in flicker component between frames is large, and flicker does not occur at point B where the change in flicker component between frames is small. The image feature amount detection apparatus in FIG. 3 can quantitatively obtain the flicker component of the moving image signal for each pixel.

[Second Embodiment]

  FIG. 6 is a block diagram of a second embodiment of the image feature quantity detecting apparatus according to the present invention. The image feature amount detection apparatus includes a blocking unit 6, a nonlinear conversion unit 7, an average value calculation unit 8, a nonlinear inverse conversion unit 9, an average value calculation unit 10, and a subtraction unit 11. A moving image signal having an electric signal level, which is an input signal to the image feature quantity detection device, is input to the blocking unit 6.

  When the moving image signal is input, the blocking unit 6 performs a blocking process for dividing the screen into a plurality of blocks on the moving image signal, and converts the moving image signal subjected to the blocking process to the nonlinear conversion unit 7 and It outputs to the average value calculation part 10.

  When the moving image signal is input, the non-linear conversion unit 7 performs non-linear conversion on the level value for each pixel, and outputs a moving image signal having an optical signal level obtained by non-linear conversion of the level value for each pixel to the average value calculation unit 8. When a moving image signal having an optical signal level is input, the average value calculation unit 8 calculates an average value (hereinafter referred to as a pixel level average value) obtained by averaging pixel level values for each block, and the pixel level average value Is output to the nonlinear inverse transform unit 9. When the pixel level average value of the optical signal level is input, the non-linear inverse conversion unit 9 outputs the pixel level average value of the electrical signal level obtained by nonlinear reverse conversion of the pixel level average value of the optical signal level to the subtraction unit 11. . On the other hand, when a moving image signal having an electric signal level is input, the average value calculating unit 10 calculates a pixel level average value and outputs the electric signal level pixel level average value to the subtracting unit 11.

  That is, the pixel level average value input to the subtraction unit 11 via the nonlinear conversion unit 7, the average value calculation unit 8 and the nonlinear inverse conversion unit 9 is an optical signal level process for averaging pixel level values for each block. It is what I did. The pixel level average value input to the subtraction unit 11 via the average value calculation unit 10 is obtained by performing processing for averaging the pixel level value for each block at the electric signal level.

  The subtraction unit 11 subtracts the pixel level average value of the electric signal level input from the nonlinear inverse conversion unit 9 and the pixel level average value of the electric signal level input from the average value calculation unit 10 for each block, and the difference Is output as a flicker component. Note that the flicker component output from the subtracting unit 11 corresponds to, for example, the difference between the average optical signal level when there is an AC component in FIG. 2 and the optical signal level when the AC component is lost.

  In the image feature amount detection apparatus of FIG. 6, by comparing flicker components of blocks having the same spatial position in successive frames or fields, the spatial position, in other words, the extent of occurrence of flicker in the block can be detected. The image feature amount detection apparatus in FIG. 6 can quantitatively obtain the flicker component of the moving image signal for each block.

[Third embodiment]

  FIG. 7 is a block diagram of a third embodiment of the image feature quantity detecting apparatus according to the present invention. The image feature amount detection apparatus includes a non-linear conversion unit 12, a narrow-band LPF 13, a narrow-band LPF 14, a non-linear conversion unit 15, and a subtraction unit 16. A moving image signal having an electric signal level, which is an input signal to the image feature quantity detection device, is input to the nonlinear converter 12 and the narrowband LPF 14.

  The nonlinear converter 12 and the narrow band LPF 13 are the same as the nonlinear converter 1 and the narrow band LPF 2 in FIG. That is, the moving image signal at the electrical signal level input to the nonlinear conversion unit 12 is converted into a moving image signal at the optical signal level by the nonlinear conversion unit 12 and then narrowband by spatial convolution calculation for each pixel by the narrowband LPF 13. The low-pass filtering process is performed. The narrowband LPF 13 outputs the moving image signal of the optical signal level subjected to the narrowband low-pass filtering processing to the subtracting unit 16.

  The narrow band LPF 14 is the same as the narrow band LPF 4 in FIG. That is, the moving image signal of the electric signal level input to the narrowband LPF 14 is subjected to narrowband low-pass filtering processing by spatial convolution calculation for each pixel. The narrowband LPF 14 outputs a moving image signal having an electric signal level that has undergone narrowband low-pass filtering processing to the non-linear converter 15.

  When a moving image signal having an electric signal level is input, the nonlinear conversion unit 15 performs a nonlinear conversion of the level value for each pixel and performs a nonlinear conversion of the level value for each pixel, similarly to the nonlinear conversion unit 12. The image signal is output to the subtracting unit 16.

  That is, the moving image signal input to the subtraction unit 16 via the nonlinear conversion unit 12 and the narrowband LPF 13 is obtained by performing a narrowband low-pass filtering process by a spatial convolution operation at an optical signal level. The moving image signal input to the subtraction unit 16 via the narrowband LPF 14 and the nonlinear conversion unit 15 is obtained by performing a narrowband low-pass filtering process by a spatial convolution operation at an electric signal level.

  The subtracting unit 16 subtracts the moving image signal of the optical signal level input from the narrowband LPF 13 and the moving image signal of the optical signal level input from the nonlinear conversion unit 15 for each pixel, and outputs the difference as a flicker component. To do. Note that the flicker component output from the subtracting unit 16 corresponds to, for example, the difference between the average optical signal level when there is an AC component in FIG. 2 and the optical signal level when the AC component is lost.

  In the image feature amount detection apparatus of FIG. 7, by comparing flicker components of pixels having the same spatial position in successive frames or fields, the spatial position, in other words, the extent of occurrence of flicker in the pixel can be detected. The image feature amount detection apparatus in FIG. 7 can quantitatively obtain the flicker component of the moving image signal for each pixel.

[Fourth embodiment]

  FIG. 8 is a block diagram of a fourth embodiment of the image feature quantity detection apparatus according to the present invention. The image feature amount detection apparatus includes a blocking unit 17, a nonlinear conversion unit 18, an average value calculation unit 19, an average value calculation unit 20, a nonlinear conversion unit 21, and a subtraction unit 22. A moving image signal having an electric signal level, which is an input signal to the image feature quantity detection device, is input to the blocking unit 17.

  The blocking unit 17 is the same as the blocking unit 6 in FIG. That is, the moving image signal of the electric signal level input to the blocking unit 17 is output to the nonlinear conversion unit 18 and the average value calculation unit 20 after being subjected to blocking processing.

  Further, the nonlinear conversion unit 18 and the average value calculation unit 19 are the same as the nonlinear conversion unit 7 and the average value calculation unit 8 of FIG. That is, the moving image signal at the electrical signal level input to the nonlinear conversion unit 18 is converted into a moving image signal at the optical signal level by the nonlinear conversion unit 18, and then the average value calculation unit 19 calculates the pixel level average value. . The average value calculation unit 19 outputs the calculated pixel level average value of the optical signal level to the subtraction unit 22.

  Moreover, the average value calculation part 20 is the same as the average value calculation part 10 of FIG. In other words, the pixel level average value is calculated for the moving image signal of the electric signal level input to the average value calculation unit 20. The average value calculation unit 20 outputs the pixel level average value of the electric signal level to the non-linear conversion unit 21.

  When the pixel level average value of the electrical signal level is input, the nonlinear conversion unit 21 performs nonlinear conversion on the pixel level average value of the electrical signal level and performs nonlinear conversion on the pixel level average value of the electrical signal level. The level average value is output to the subtraction unit 22.

  That is, the pixel level average value input to the subtraction unit 22 via the nonlinear conversion unit 18 and the average value calculation unit 19 is obtained by performing the process of averaging the pixel level values for each block at the optical signal level. The pixel level average value input to the subtraction unit 22 via the average value calculation unit 20 and the non-linear conversion unit 21 is obtained by performing the process of averaging the pixel level value for each block at the electric signal level.

  The subtracting unit 22 subtracts the pixel level average value of the optical signal level input from the average value calculating unit 19 and the pixel level average value of the optical signal level input from the nonlinear conversion unit 21 for each block, and calculates the difference. Output as flicker component. Note that the flicker component output from the subtracting unit 22 corresponds to, for example, the difference between the average optical signal level when there is an AC component in FIG. 2 and the optical signal level when the AC component is lost.

  In the image feature quantity detection apparatus of FIG. 8, by comparing flicker components of blocks having the same spatial position in successive frames or fields, the spatial position, in other words, the extent of occurrence of flicker in the block can be detected. The image feature amount detection apparatus of FIG. 8 can quantitatively obtain the flicker component of the moving image signal for each block.

  According to the image feature amount detection apparatus of the first to fourth embodiments, it is possible to quantitatively obtain the flicker component of the moving image signal for each screen area such as a pixel or a block.

The image quality improvement apparatus according to the present invention is a screen area for flicker degradation that is generated by encoding and decoding processing of moving images and perceived as a change in frequency characteristics between images as a change in luminance at the optical signal level. Therefore, the image quality can be improved by performing the adaptive flicker correction according to the above and not subjecting the original luminance change of the moving image such as a scene change to the flicker correction.
[First embodiment]

  FIG. 9 is a block diagram of a first embodiment of the image quality improving apparatus according to the present invention.

  The image quality improvement apparatus shown in FIG. 9 includes an image feature quantity detection device including a non-linear conversion unit 31, a narrowband LPF 32, a non-linear inverse conversion unit 33, a narrowband LPF 34 and a subtraction unit 35, and a subtraction unit 36. A moving image signal having an electric signal level, which is an input signal to the image quality improving apparatus, is input to the non-linear conversion unit 31, the narrowband LPF 34, and the subtraction unit 36.

  Note that the image feature quantity detection device including the nonlinear transformation unit 31, the narrowband LPF 32, the nonlinear inverse transformation unit 33, the narrowband LPF 34, and the subtraction unit 35 is the same as the image feature quantity detection device of FIG. To do. The subtractor 35 outputs flicker component data (hereinafter referred to as flicker component data). The flicker component data output from the subtractor 35 is output to the subtractor 36.

  The subtracting unit 36 subtracts the flicker component data input from the subtracting unit 35 from the input moving image signal for each pixel having the same spatial and temporal position, and outputs it as a corrected moving image signal.

The image quality improvement apparatus shown in FIG. 9 repeatedly performs the signal processing described above to adapt to the flicker deterioration caused by the change in the frequency characteristics of the moving image for each screen area by removing the flicker component from the input moving image signal. The corrected moving image signal with improved image quality and improved image quality can be output.
[Second Embodiment]

  FIG. 10 is a block diagram of a second embodiment of the image quality improving apparatus according to the present invention.

  The image quality improvement apparatus shown in FIG. 10 includes an image feature amount detection apparatus including a blocking unit 41, a nonlinear conversion unit 42, an average value calculation unit 43, a nonlinear inverse conversion unit 44, an average value calculation unit 45, and a subtraction unit 46, and A subtraction unit 47 and an image reconstruction unit 48 are included. A moving image signal having an electric signal level, which is an input signal to the image quality improving apparatus, is input to the blocking unit 41.

  When the moving image signal is input, the blocking unit 41 performs a blocking process for dividing the screen into a plurality of blocks on the moving image signal, and converts the moving image signal subjected to the blocking process to the nonlinear conversion unit 42, The result is output to the average value calculation unit 45 and the subtraction unit 47. Note that the image feature quantity detection device including the blocking unit 41, the nonlinear conversion unit 42, the average value calculation unit 43, the nonlinear inverse transformation unit 44, the average value calculation unit 45, and the subtraction unit 46 is the image feature quantity detection device of FIG. Since it is the same as that, description is abbreviate | omitted. The subtractor 46 outputs the flicker component to the subtractor 47.

  The subtracting unit 47 subtracts the flicker component input from the subtracting unit 46 from the blocked moving image signal input from the blocking unit 41 for each pixel of the block having the same spatial and temporal position, The difference is output to the image reconstruction unit 48.

  When the moving image signal subjected to the blocking process is input from the subtracting unit 47, the image reconstructing unit 48 performs a process of reconstructing the moving image by returning the blocked moving image signal to the original position. And output as a corrected moving image signal.

The image quality improvement apparatus in FIG. 10 can adaptively reduce flicker degradation caused by changes in the frequency characteristics of moving images for each screen area by repeatedly performing the above signal processing.
[Third embodiment]

  FIG. 11 is a block diagram of a third embodiment of the image quality improving apparatus according to the present invention. The image quality improvement apparatus shown in FIG. 11 includes an image feature amount detection device including a non-linear conversion unit 51, a narrow-band LPF 52, a narrow-band LPF 53, a non-linear conversion unit 54, and a subtraction unit 55, a subtraction unit 56, and a non-linear inverse conversion unit 57. Is done. A moving image signal having an electric signal level, which is an input signal to the image quality improvement apparatus, is input to the nonlinear converter 51 and the narrowband LPF 53.

  Note that the image feature quantity detection device including the nonlinear transformation unit 51, the narrowband LPF 52, the narrowband LPF 53, the nonlinear transformation unit 54, and the subtraction unit 55 is the same as the image feature quantity detection device of FIG. . The subtractor 55 outputs the flicker component data to the subtractor 56.

  The subtracting unit 56 subtracts the flicker component data input from the subtracting unit 55 from the moving image signal from the non-linear converting unit 51 for each pixel having the same spatial and temporal position, and calculates the difference as a non-linear inverse converting unit. To 57.

  When a moving image signal having an optical signal level is input, the nonlinear inverse converting unit 57 performs nonlinear inverse conversion of the moving image signal having the optical signal level into a moving image signal having an electric signal level, and outputs the converted moving image signal as a corrected moving image signal.

The image quality improvement apparatus in FIG. 11 can adaptively reduce flicker degradation caused by changes in the frequency characteristics of moving images for each screen area by repeatedly performing the above signal processing.
[Fourth embodiment]

  FIG. 12 is a block diagram of a fourth embodiment of the image quality improving apparatus according to the present invention. The image quality improvement apparatus of FIG. 12 includes an image feature quantity detection apparatus including a blocking unit 61, a nonlinear conversion unit 62, an average value calculation unit 63, an average value calculation unit 64, a nonlinear conversion unit 65, and a subtraction unit 66, and a subtraction unit 67. And a non-linear inverse transformation unit 68 and an image reconstruction unit 69. A moving image signal having an electric signal level, which is an input signal to the image quality improving apparatus, is input to the blocking unit 61.

  When the moving image signal is input, the blocking unit 61 performs a blocking process for dividing the screen into a plurality of blocks on the moving image signal, and converts the moving image signal subjected to the blocking process to the nonlinear conversion unit 62, The result is output to the average value calculation unit 64. Note that the image feature quantity detection device including the blocking unit 61, the nonlinear conversion unit 62, the average value calculation unit 63, the average value calculation unit 64, the nonlinear conversion unit 65, and the subtraction unit 66 is the same as the image feature quantity detection device of FIG. The description is omitted because it is similar. The subtraction unit 66 outputs the flicker component data to the subtraction unit 67.

  The subtracting unit 67 subtracts the flicker component data input from the subtracting unit 66 from the blocked moving image signal input from the nonlinear converting unit 62 for each pixel of the block having the same spatial and temporal positions. The result is output to the nonlinear inverse transform unit 68.

  When a moving image signal having an optical signal level is input, the nonlinear inverse converting unit 68 performs nonlinear inverse conversion on the moving image signal having the optical signal level into a moving image signal having an electric signal level, and outputs the moving image signal to the image reconstruction unit 69. .

  When the moving image signal subjected to the blocking process is input from the non-linear inverse transform unit 68, the image reconstructing unit 69 returns the blocked moving image signal to the original position and reconstructs the moving image. And output as a corrected moving image signal.

  The image quality improvement apparatus in FIG. 12 can adaptively reduce flicker degradation caused by changes in the frequency characteristics of moving images for each screen area by repeatedly performing the above signal processing.

[Fifth embodiment]

  FIG. 13 is a block diagram of a fifth embodiment of the image quality improving apparatus according to the present invention.

  The image quality improvement apparatus includes an image feature amount detection device including a nonlinear conversion unit 101, a narrowband LPF 102, a nonlinear inverse conversion unit 103, a narrowband LPF 104, and a subtraction unit 105, a memory 106, a subtraction unit 107, a switch 109, And an adder 110. A moving image signal having an electric signal level, which is an input signal to the image quality improvement apparatus, is input to the nonlinear conversion unit 101, the narrowband LPF 104, and the addition unit 110.

  Note that the image feature quantity detection device including the nonlinear transformation unit 101, the narrowband LPF 102, the nonlinear inverse transformation unit 103, the narrowband LPF 104, and the subtraction unit 105 is the same as the image feature quantity detection device of FIG. To do. The subtraction unit 105 outputs flicker component data (hereinafter referred to as flicker component data) to the memory 106 and the subtraction unit 107.

  The memory 106 operates with reference to the input identification signal. Here, the identification signal is used to identify a frame or field (hereinafter referred to as a high-luminance frame or field) having the highest average optical signal level output from the display device among a predetermined number of frames or fields. Is. The high luminance frame or field is a frame or field indicated by the identification signal.

  For example, the high luminance frame or field is an I picture or a P picture. In the case of an I picture or a P picture, an identification signal for identifying the I picture or the P picture can be obtained from a decoding device (not shown).

  The memory 106 writes the flicker component data only when the flicker component data of the high-luminance frame or field is input. At this time, there is no reading of flicker component data from the memory 106. Further, when flicker component data other than a high-luminance frame or field is input, the memory 106 does not write the flicker component data. At this time, the memory 106 outputs the held flicker component data to the subtraction unit 107.

  The subtraction unit 107 subtracts the flicker component data input from the memory 106 and the flicker component data input from the subtraction unit 105 for each pixel having the same spatial position, and uses the difference as a flicker correction value as a switch. Output to 109.

  For example, when the identification signal indicates a high-luminance frame or field, since the flicker component data is not read from the memory 106, the subtractor 107 outputs the flicker component data input from the subtractor 105 to the switch 109 as a flicker correction value. To do.

  On the other hand, when the identification signal indicates a frame or field other than the high-intensity frame or field, the subtraction unit 107 receives the high-intensity frame or field flicker component data input from the memory 106 and the high-intensity frame input from the subtraction unit 105. Alternatively, a flicker correction value obtained by subtracting the flicker component data of a frame or field other than the field for each pixel is output to the switch 109.

  The switch 109 operates with reference to the input identification signal. The switch 109 outputs the flicker correction value input from the subtractor 107 to the adder 110 when the identification signal indicates a frame or field other than the high luminance frame or field. The switch 109 does not output the flicker correction value input from the subtractor 107 to the adder 110 when the identification signal indicates a high-luminance frame or field.

  The adding unit 110 performs a process of adding the moving image signal of the electric signal level input as the input signal and the flicker correction value input from the switch 109 for each pixel having the same spatial and temporal position. Do.

  For example, when the identification signal indicates a high-luminance frame or field, since the flicker correction value is not input from the switch 109, the adding unit 110 uses the moving image signal of the electric signal level input as the input signal as the corrected moving image signal. Output.

  On the other hand, when the identification signal indicates a frame or field other than the high-luminance frame or field, the adding unit 110 receives the flicker correction value input from the switch 109 and the moving image signal at the electric signal level input as the input signal. Then, addition is performed for each pixel having the same spatial and temporal positions, and a corrected moving image signal is output.

  13 uses the difference between the flicker component data of the high-luminance frame or field and the flicker component data of the frame other than the high-luminance frame or field or the flicker component data of the field as a flicker correction value, After the flicker correction value is added to the moving image signal of the field, it is output as a corrected moving image signal. Note that the image quality improving apparatus in FIG. 13 outputs a corrected moving image signal without adding a flicker correction value to a moving image signal of a high-luminance frame or field.

  The image quality improvement apparatus in FIG. 13 repeatedly performs the signal processing described above, thereby reducing flicker degradation caused by a change in the frequency characteristics of the moving image for each screen area by correcting the level value for each pixel using spatial convolution calculation. Can be adaptively reduced.

[Sixth embodiment]

  FIG. 14 is a block diagram of a sixth embodiment of the image quality improving apparatus according to the present invention.

  The image quality improvement apparatus in FIG. 14 includes an image feature amount detection apparatus including a blocking unit 111, a nonlinear conversion unit 112, an average value calculation unit 113, a nonlinear inverse conversion unit 114, an average value calculation unit 115, and a subtraction unit 116, and a memory 117. And a subtractor 118, a switch 120, an adder 121, and an image reconstruction unit 122. A moving image signal having an electric signal level, which is an input signal to the image quality improvement apparatus, is input to the blocking unit 111.

  When the moving image signal is input, the blocking unit 111 performs a blocking process for dividing the screen into a plurality of blocks on the moving image signal, and converts the moving image signal subjected to the blocking process to the nonlinear conversion unit 112, The result is output to average value calculation section 115 and addition section 121. Note that the image feature quantity detection device including the blocking unit 111, the nonlinear transformation unit 112, the average value calculation unit 113, the nonlinear inverse transformation unit 114, the average value calculation unit 115, and the subtraction unit 116 is the image feature quantity detection device of FIG. Since it is the same as that of FIG. The subtractor 116 outputs the flicker component data to the memory 117 and the subtractor 118.

  The memory 117 operates with reference to the input identification signal, and writes the flicker component data only when the flicker component data of the high-luminance frame or field is input. At this time, the memory 117 does not read the held flicker component data. In addition, when the flicker component data of a frame or field other than the high-luminance frame or field is input, the memory 117 does not write the flicker component data. At this time, the memory 117 outputs the held flicker component data to the subtractor 118.

  The subtractor 118 subtracts the flicker component data input from the memory 117 and the flicker component data input from the subtractor 116 for each block having the same spatial position, and uses the difference as a flicker correction value for the switch. 120 is output.

  For example, when the identification signal indicates a high-luminance frame or field, since the flicker component data is not read from the memory 117, the subtractor 118 outputs the flicker component data input from the subtractor 116 as a flicker correction value to the switch 120. To do.

  On the other hand, when the identification signal indicates a frame or field other than the high-luminance frame or field, the subtraction unit 118 and the high-luminance frame or field flicker component data input from the memory 117 and the high-luminance frame input from the subtraction unit 116 Alternatively, a flicker correction value obtained by subtracting the flicker component data of a frame or field other than the field for each block is output to the switch 120.

  The switch 120 operates with reference to the input identification signal. The switch 120 outputs the flicker correction value input from the subtractor 118 to the adder 121 when the identification signal indicates a frame or field other than the high luminance frame or field. The switch 120 does not output the flicker correction value input from the subtractor 118 to the adder 121 when the identification signal indicates a high-luminance frame or field.

  The adding unit 121 performs a process of adding the moving image signal subjected to the blocking process and the flicker correction value input from the switch 120 for each pixel of a block having the same spatial and temporal positions.

  For example, when the identification signal indicates a high-luminance frame or field, since the flicker correction value is not input from the switch 120, the adding unit 121 outputs the moving image signal subjected to the blocking process to the image reconstruction unit 122.

  On the other hand, when the identification signal indicates a frame or a field other than the high-luminance frame or field, the adding unit 121 performs spatial and spatial conversion between the flicker correction value input from the switch 120 and the moving image signal subjected to the blocking process. It adds for every pixel of the block with the same temporal position, and outputs it to the image reconstruction part 122. FIG.

  When the moving image signal subjected to the blocking process is input from the adder 121, the image reconstructing unit 122 performs a process of reconstructing the moving image by returning the blocked moving image signal to the original position. And output as a corrected moving image signal.

  14 uses the difference between the flicker component data of the high-luminance frame or field and the flicker component data of the frame or field other than the high-luminance frame or field as a flicker correction value, A flicker correction value is added to the field moving image signal for each pixel of the block, and then output as a corrected moving image signal. Note that the image quality improvement apparatus in FIG. 14 outputs a corrected moving image signal without adding a flicker correction value to a moving image signal of a high-luminance frame or field.

  The image quality improvement apparatus in FIG. 14 adaptively performs flicker degradation due to a change in the frequency characteristics of moving images for each screen region by correcting the level value for each pixel using a block by repeatedly performing the above signal processing. Can be reduced.

[Seventh embodiment]

  FIG. 15 is a block diagram of a seventh embodiment of the image quality improving apparatus according to the present invention.

  The image quality improvement apparatus in FIG. 15 is a configuration example in which the picture interval M between the I picture and the P picture is 3, and correction is performed in units of frames. The image quality improvement apparatus includes an image feature amount detection apparatus including a nonlinear conversion unit 123, a narrowband LPF 124, a nonlinear inverse conversion unit 125, a narrowband LPF 126, and a subtraction unit 127, a delay 128, an average value calculation unit 129, and a delay 130. , Subtractor 131, switch 133, and subtractor 134. A moving image signal having an electric signal level, which is an input signal to the image quality improvement apparatus, is input to the nonlinear converter 123, the narrowband LPF 126, and the subtractor 134.

  Note that the image feature quantity detection device including the nonlinear transformation unit 123, the narrowband LPF 124, the nonlinear inverse transformation unit 125, the narrowband LPF 126, and the subtraction unit 127 is the same as the image feature quantity detection device of FIG. To do. The subtraction unit 127 outputs the flicker component data to the delay 128, the average value calculation unit 129, and the subtraction unit 131.

  The delay 128 delays one frame of the flicker component data of the input high luminance frame or a frame other than the high luminance frame. The average value calculator 129 receives the flicker component data from the subtractor 127 and the flicker component data delayed by one frame from the delay 128. That is, the two flicker component data input to the average value calculation unit 129 have a time difference of one frame. The average value calculation unit 129 calculates the average value of the two input flicker component data for each pixel and outputs it to the delay 130.

  The delay 130 delays the average value of the two flicker component data input from the average value calculation unit 129 by one frame and outputs the result to the subtraction unit 131. The subtractor 131 subtracts the flicker component data input from the subtractor 127 and the flicker component data input from the delay 130 for each pixel having the same spatial position, and opens and closes the difference as a flicker correction value. Output to the device 133.

  The switch 133 operates with reference to the input identification signal. The switch 133 outputs the flicker correction value input from the subtraction unit 131 to the subtraction unit 134 when the identification signal indicates a high luminance frame. The switch 133 does not output the flicker correction value input from the subtraction unit 131 to the subtraction unit 134 when the identification signal indicates a frame other than the high-luminance frame.

  For example, when the identification signal indicates a high-luminance frame, the flicker correction value output from the subtraction unit 131 includes the flicker component data of the high-luminance frame input from the subtraction unit 127 and the frames other than the high-luminance frame input from the delay 130. The average of the flicker component data is subtracted for each pixel having the same spatial position.

  The subtracting unit 134 performs a process of subtracting, for each pixel having the same spatial and temporal position, a moving image signal having an electric signal level input as an input signal and a flicker correction value input from the switch 133. Do.

  For example, when the identification signal indicates a frame other than the high-luminance frame, since the flicker correction value is not input from the switch 133, the subtracting unit 134 converts the moving image signal at the electric signal level input as the input signal to the corrected moving image signal. Output as. On the other hand, when the identification signal indicates a high-luminance frame, the subtracting unit 134 converts the flicker correction value input from the switch 133 and the moving image signal of the electric signal level input as the input signal, spatially and temporally. Subtraction is performed for each pixel having the same position, and a corrected moving image signal is output.

  The image quality improvement apparatus in FIG. 15 is a configuration example when the picture interval M is 3, but the configuration is the same even when the value of the picture interval M changes or when correction is performed in units of fields. Yes. In other words, the average value calculation unit 129 adjusts the arrangement and delay amount of the delay of the preceding stage or the subsequent stage of the average value calculation unit 129 so that the average of flicker component data of consecutive frames or fields other than the luminance frame or field can be calculated. do it.

  The image quality improvement apparatus in FIG. 15 uses the difference between the flicker component data of a high-luminance frame or field and the average of the flicker component data of a frame or field other than the high-luminance frame or field as a flicker correction value, and After subtracting the flicker correction value from the image signal, it is output as a corrected moving image signal. Note that the image quality improvement apparatus in FIG. 15 outputs a corrected moving image signal without subtracting the flicker correction value from the moving image signal of a frame or field other than the high luminance frame or field.

  The image quality improvement apparatus in FIG. 15 repeats the above signal processing, thereby reducing flicker degradation caused by changes in the frequency characteristics of moving images for each screen area by correcting the level value for each pixel using spatial convolution calculation. Can be adaptively reduced.

[Eighth embodiment]

  FIG. 16 is a block diagram of an eighth embodiment of the image quality improving apparatus according to the present invention.

  The image quality improvement apparatus in FIG. 16 is a configuration example when the picture interval M between the I picture and the P picture is 3, and correction is performed in units of frames.

  The image quality improvement apparatus of FIG. 16 includes an image feature amount detection apparatus including a blocking unit 135, a nonlinear conversion unit 136, an average value calculation unit 137, a nonlinear inverse conversion unit 138, an average value calculation unit 139, and a subtraction unit 140; And an average value calculation unit 142, a delay 143, a subtraction unit 144, a switch 146, a subtraction unit 147, and an image reconstruction unit 148. A moving image signal having an electric signal level, which is an input signal to the image quality improving apparatus, is input to the blocking unit 135.

  When the moving image signal is input, the blocking unit 135 performs a blocking process for dividing the screen into a plurality of blocks on the moving image signal, and converts the moving image signal subjected to the blocking process to the nonlinear conversion units 136 and 136. The result is output to the average value calculation unit 139 and the subtraction unit 147. Note that the image feature quantity detection device including the blocking unit 135, the nonlinear transformation unit 136, the average value calculation unit 137, the nonlinear inverse transformation unit 138, the average value calculation unit 139, and the subtraction unit 140 is the image feature quantity detection device of FIG. Since it is the same as that, description is abbreviate | omitted. The subtraction unit 140 outputs the flicker component data to the delay 141, the average value calculation unit 142, and the subtraction unit 144.

  The delay 141 delays one frame of the flicker component data of the input high luminance frame or a frame other than the high luminance frame. The average value calculation unit 142 receives the flicker component data from the subtraction unit 140 and the flicker component data delayed by one frame from the delay 141. That is, the two flicker component data input to the average value calculation unit 142 have a time difference of one frame. The average value calculation unit 142 calculates the average value of the two input flicker component data for each block, and outputs it to the delay 143.

  The delay 143 delays the average value of the two flicker component data input from the average value calculation unit 142 by one frame and outputs the result to the subtraction unit 144. The subtraction unit 144 subtracts the flicker component data input from the subtraction unit 140 and the flicker component data input from the delay 143 for each block having the same spatial position, and opens and closes the difference as a flicker correction value. Output to the device 146.

  The switch 146 operates with reference to the input identification signal. The switch 146 outputs the flicker correction value input from the subtraction unit 144 to the subtraction unit 147 when the identification signal indicates a high luminance frame. The switch 146 does not output the flicker correction value input from the subtraction unit 144 to the subtraction unit 147 when the identification signal indicates a frame other than the high-luminance frame.

  For example, when the identification signal indicates a high-luminance frame, the flicker correction value output from the subtraction unit 144 is a frame other than the high-luminance frame input from the flicker component data of the high-luminance frame input from the subtraction unit 140 and the delay 143. The average of the flicker component data is subtracted for each block having the same spatial position.

  The subtracting unit 147 subtracts the moving image signal of the electric signal level input as the input signal and the flicker correction value input from the switch 146 for each pixel of the block having the same spatial and temporal positions. Process.

  For example, when the identification signal indicates a frame other than the high-luminance frame, since the flicker correction value is not input from the switch 146, the subtracting unit 147 converts the moving image signal of the electric signal level input as the input signal into the corrected moving image signal. Output as. On the other hand, when the identification signal indicates a high-luminance frame, the subtracting unit 147 calculates the flicker correction value input from the switch 146 and the moving image signal of the electric signal level input as the input signal, spatially and temporally. Subtraction is performed for each pixel in the block having the same position, and the corrected moving image signal is output.

  When the blocked moving image signal is input from the subtracting unit 147, the image reconstructing unit 148 performs processing for restoring the blocked moving image signal to the original position and reconstructing the moving image. And output as a corrected moving image signal.

  The image quality improvement apparatus in FIG. 16 is a configuration example in the case where the picture interval M is 3, but it has the same configuration even when the value of the picture interval M changes or when correction is performed in units of fields. Yes. In other words, the average value calculation unit 142 adjusts the arrangement and delay amount of delays before or after the average value calculation unit 142 so that the average of flicker component data of consecutive frames or fields other than the luminance frame or field can be calculated. do it.

  The image quality improvement apparatus in FIG. 16 uses the difference between the flicker component data of a high-luminance frame or field and the average of the flicker component data of a frame or field other than the high-luminance frame or field as a flicker correction value, and After subtracting the flicker correction value from the image signal, it is output as a corrected moving image signal. Note that the image quality improving apparatus in FIG. 16 outputs a corrected moving image signal without subtracting the flicker correction value from the moving image signal of a frame or field other than the high-luminance frame or field.

  The image quality improvement apparatus in FIG. 16 repeatedly performs the signal processing described above, so that flicker degradation caused by a change in the frequency characteristics of the moving image is adaptive for each screen region by correcting the level value for each pixel using a block. Can be reduced.

[Ninth embodiment]

  FIG. 17 is a block diagram of a ninth embodiment of the image quality improving apparatus according to the present invention.

  The image quality improvement apparatus of FIG. 17 includes an image feature amount detection device including a non-linear conversion unit 149, a narrow-band LPF 150, a narrow-band LPF 151, a non-linear conversion unit 152, and a subtraction unit 153, a memory 154, a subtraction unit 155, and a switch 157. And an adder 158 and a non-linear inverse transformer 159. A moving image signal having an electric signal level, which is an input signal to the image quality improvement apparatus, is input to the nonlinear conversion unit 149 and the narrowband LPF 151.

  Note that the image feature quantity detection device including the nonlinear transformation unit 149, the narrow band LPF 150, the narrow band LPF 151, the nonlinear transformation unit 152, and the subtraction unit 153 is the same as the image feature quantity detection device in FIG. . The subtraction unit 153 outputs the flicker component data to the memory 154 and the subtraction unit 156.

  The memory 154 operates with reference to the input identification signal. The memory 154 writes the flicker component data only when the flicker component data of the high luminance frame or field is input. At this time, there is no reading of flicker component data from the memory 154. Further, when flicker component data other than a high-luminance frame or field is input, the memory 154 does not write the flicker component data. At this time, the memory 154 outputs the held flicker component data to the subtraction unit 155.

  The subtracting unit 155 subtracts the flicker component data input from the memory 154 and the flicker component data input from the subtracting unit 153 for each pixel having the same spatial position, and uses the difference as a flicker correction value. To 157.

  For example, when the identification signal indicates a high-luminance frame or field, since the flicker component data is not read from the memory 154, the subtractor 155 outputs the flicker component data input from the subtractor 153 to the switch 157 as a flicker correction value. To do.

  On the other hand, when the identification signal indicates a frame or field other than the high-intensity frame or field, the subtraction unit 155 receives the high-intensity frame or field flicker component data input from the memory 154 and the high-intensity frame input from the subtraction unit 153. Alternatively, a flicker correction value obtained by subtracting the flicker component data of the frame or field other than the field for each pixel is output to the switch 157.

  The switch 157 operates with reference to the input identification signal. The switch 157 outputs the flicker correction value input from the subtractor 155 to the adder 158 when the identification signal indicates a frame or field other than the high luminance frame or field. The switch 157 does not output the flicker correction value input from the subtraction unit 155 to the addition unit 158 when the identification signal indicates a high-luminance frame or field.

  The adder 158 adds the optical signal level moving image signal input from the nonlinear converter 149 and the flicker correction value input from the switch 157 for each pixel having the same spatial and temporal positions. Process.

  For example, when the identification signal indicates a high-luminance frame or field, since the flicker correction value is not input from the switch 157, the adder 158 performs non-linear inverse conversion on the moving image signal at the optical signal level input from the non-linear converter 149. Output to the unit 159.

  On the other hand, when the identification signal indicates a frame or a field other than the high-luminance frame or field, the adder 158 and the flicker correction value input from the switch 157 and the moving image signal of the optical signal level input from the nonlinear converter 149 Are added to each pixel having the same spatial and temporal position, and output to the nonlinear inverse transform unit 159.

  When a moving image signal having an optical signal level is input, the nonlinear inverse converting unit 159 performs nonlinear inverse conversion on the moving image signal having the optical signal level to a moving image signal having an electric signal level, and outputs the moving image signal as a corrected moving image signal.

  17 uses the difference between the flicker component data of the high-luminance frame or field and the flicker component data of the frame or field other than the high-luminance frame or field as the flicker correction value, After the flicker correction value is added to the moving image signal of the field, it is output as a corrected moving image signal. Note that the image quality improvement apparatus in FIG. 17 outputs a corrected moving image signal without adding a flicker correction value to a moving image signal of a high-luminance frame or field.

  The image quality improvement apparatus in FIG. 17 repeats the signal processing described above, thereby reducing flicker degradation caused by changes in the frequency characteristics of moving images for each screen area by correcting the level value for each pixel using spatial convolution calculation. Can be adaptively reduced.

[Tenth embodiment]

  FIG. 18 is a block diagram of a tenth embodiment of the image quality improving apparatus according to the present invention.

  The image quality improvement apparatus of FIG. 18 includes an image feature quantity detection device including a block forming unit 160, a nonlinear conversion unit 161, an average value calculation unit 162, an average value calculation unit 163, a nonlinear conversion unit 164, and a subtraction unit 165, a memory 166, , A subtractor 167, a switch 169, an adder 170, a non-linear inverse transform unit 171, and an image reconstruction unit 172. A moving image signal having an electric signal level, which is an input signal to the image quality improvement apparatus, is input to the blocking unit 160.

  When the moving image signal is input, the blocking unit 160 performs a blocking process for dividing the screen into a plurality of blocks on the moving image signal, and converts the moving image signal subjected to the blocking process to the nonlinear conversion unit 161, The result is output to the average value calculation unit 163. Note that the image feature quantity detection device including the blocking unit 160, the nonlinear conversion unit 161, the average value calculation unit 162, the average value calculation unit 163, the nonlinear conversion unit 164, and the subtraction unit 165 is the same as the image feature quantity detection device of FIG. The description is omitted because it is similar. The subtraction unit 165 outputs the flicker component data to the memory 166 and the subtraction unit 167.

  The memory 166 operates with reference to the input identification signal, and writes the flicker component data only when the flicker component data of the high-luminance frame or field is input. At this time, the memory 166 does not read the held flicker component data. Further, when flicker component data of a frame or field other than the high luminance frame or field is input, the memory 166 does not write the flicker component data. At this time, the memory 166 outputs the held flicker component data to the subtraction unit 167.

  The subtracting unit 167 subtracts the flicker component data input from the memory 166 and the flicker component data input from the subtracting unit 165 for each block having the same spatial position, and uses the difference as a flicker correction value. To 169.

  For example, when the identification signal indicates a high-luminance frame or field, since the flicker component data is not read from the memory 166, the subtractor 167 outputs the flicker component data input from the subtractor 165 to the switch 169 as a flicker correction value. To do.

  On the other hand, when the identification signal indicates a frame or field other than the high-luminance frame or field, the subtraction unit 167 outputs the high-luminance frame or field flicker component data input from the memory 166 and the high-luminance frame input from the subtraction unit 165. Alternatively, the flicker correction value obtained by subtracting the flicker component data of the frame or field other than the field for each block is output to the switch 169.

  The switch 169 operates with reference to the input identification signal. The switch 169 outputs the flicker correction value input from the subtractor 167 to the adder 170 when the identification signal indicates a frame or field other than the high luminance frame or field. The switch 169 does not output the flicker correction value input from the subtractor 167 to the adder 170 when the identification signal indicates a high-luminance frame or field.

  The adder 170 has the same spatial and temporal position for the optical signal level moving image signal input from the nonlinear converter 161 and the flicker correction value input from the switch 169. Addition processing is performed for each pixel of a certain block.

  For example, when the identification signal indicates a high-luminance frame or field, since the flicker correction value is not input from the switch 169, the adder 170 outputs the moving image signal subjected to the blocking process to the non-linear inverse converter 171.

  On the other hand, when the identification signal indicates a frame or field other than the high-luminance frame or field, the adder 170 converts the flicker correction value input from the switch 169 and the moving image signal subjected to the block processing into spatial and Add each pixel of a block having the same temporal position and output to the nonlinear inverse transform unit 171.

  When a moving image signal having an optical signal level is input, the nonlinear inverse converting unit 171 performs nonlinear inverse conversion of the moving image signal having the optical signal level into a moving image signal having an electric signal level, and outputs the moving image signal to the image reconstruction unit 172. .

  When the moving image signal subjected to the blocking process is input from the non-linear inverse transform unit 171, the image reconstructing unit 172 returns the blocked moving image signal to the original position and reconstructs the moving image. And output as a corrected moving image signal.

  The image quality improvement apparatus in FIG. 18 uses the difference between the flicker component data of the high-luminance frame or field and the flicker component data of the frame or field other than the high-luminance frame or field as the flicker correction value, A flicker correction value is added to the field moving image signal for each pixel of the block, and then output as a corrected moving image signal. Note that the image quality improving apparatus in FIG. 18 outputs a corrected moving image signal without adding a flicker correction value to the moving image signal of a high-luminance frame or field.

  The image quality improvement apparatus in FIG. 18 performs the above signal processing repeatedly so that flicker degradation caused by changes in the frequency characteristics of moving images is adaptive for each screen area by correcting the level value for each pixel using a block. Can be reduced.

[Eleventh embodiment]

  FIG. 19 is a block diagram of an eleventh embodiment of the image quality improving apparatus according to the present invention.

  The image quality improvement apparatus in FIG. 19 is a configuration example when the picture interval M between the I picture and the P picture is 3, and correction is performed in units of frames. The image quality improvement apparatus includes an image feature amount detection device including a non-linear conversion unit 173, a narrow-band LPF 174, a narrow-band LPF 175, a non-linear conversion unit 176, and a subtraction unit 177, a delay 178, an average value calculation unit 179, a delay 180, A subtracting unit 181, a switch 183, a subtracting unit 184, and a nonlinear inverse transform unit 185 are configured. A moving image signal having an electric signal level, which is an input signal to the image quality improvement apparatus, is input to the nonlinear converter 173 and the narrowband LPF 175.

  Note that the image feature quantity detection device including the nonlinear transformation unit 173, the narrow band LPF 174, the narrow band LPF 175, the nonlinear transformation unit 176, and the subtraction unit 177 is the same as the image feature quantity detection device of FIG. .

  The subtraction unit 177 outputs the flicker component data to the delay 178, the average value calculation unit 179, and the subtraction unit 181. The delay 178 delays one frame of the flicker component data of the input high-luminance frame or a frame other than the high-luminance frame. The average value calculation unit 179 receives the flicker component data from the subtraction unit 177 and the flicker component data delayed by one frame from the delay 178.

  That is, the two flicker component data input to the average value calculation unit 179 have a time difference of one frame. The average value calculation unit 179 calculates the average value of the two input flicker component data for each pixel and outputs it to the delay 180.

  The delay 180 delays the average value of the two flicker component data input from the average value calculation unit 179 by one frame and outputs it to the subtraction unit 181. The subtraction unit 181 subtracts the flicker component data input from the subtraction unit 177 and the flicker component data input from the delay 180 for each pixel having the same spatial position, and opens and closes the difference as a flicker correction value. Output to the device 183.

  The switch 183 operates with reference to the input identification signal. The switch 183 outputs the flicker correction value input from the subtraction unit 181 to the subtraction unit 184 when the identification signal indicates a high luminance frame. The switch 183 does not output the flicker correction value input from the subtraction unit 181 to the subtraction unit 184 when the identification signal indicates a frame other than the high luminance frame.

  For example, when the identification signal indicates a high luminance frame, the flicker correction value output from the subtraction unit 181 includes the flicker component data of the high luminance frame input from the subtraction unit 177 and the frames other than the high luminance frame input from the delay 180. The average of the flicker component data is subtracted for each pixel having the same spatial position.

  The subtracting unit 184 subtracts the moving image signal of the optical signal level input from the nonlinear conversion unit 173 and the flicker correction value input from the switch 183 for each pixel having the same spatial and temporal position. Process.

  For example, when the identification signal indicates a frame other than a high-luminance frame, since the flicker correction value is not input from the switch 183, the subtracting unit 184 converts the moving image signal of the optical signal level input from the nonlinear converting unit 173 into a nonlinear inverse signal. The data is output to the conversion unit 185.

  On the other hand, when the identification signal indicates a high-luminance frame, the subtractor 184 calculates the flicker correction value input from the switch 183 and the moving image signal at the optical signal level input from the nonlinear converter 173, spatially and temporally. Each pixel having the same general position is subtracted and output to the nonlinear inverse transform unit 185.

  When a moving image signal having an optical signal level is input, the nonlinear inverse conversion unit 185 performs nonlinear inverse conversion on the moving image signal having the optical signal level into a moving image signal having an electric signal level, and outputs the corrected moving image signal as a corrected moving image signal.

  The image quality improvement apparatus in FIG. 19 is a configuration example in the case where the picture interval M is 3, but the configuration is the same even when the value of the picture interval M changes or when correction is performed in units of fields. Yes. In other words, the average value calculation unit 179 adjusts the arrangement and delay amount of delays before and after the average value calculation unit 179 so that the average of flicker component data of consecutive frames or fields other than the luminance frame or field can be calculated. do it.

  19 uses the difference between the flicker component data of the high-luminance frame or field and the average of the flicker component data of the frame or field other than the high-luminance frame or field as the flicker correction value, and the moving image of the high-luminance frame or field. After subtracting the flicker correction value from the image signal, it is output as a corrected moving image signal. Note that the image quality improving apparatus in FIG. 19 outputs a corrected moving image signal without subtracting the flicker correction value from the moving image signal of a frame or field other than the high luminance frame or field.

  The image quality improvement apparatus in FIG. 19 repeatedly performs the signal processing described above, thereby reducing flicker degradation caused by changes in the frequency characteristics of moving images for each screen area by correcting the level value for each pixel using spatial convolution calculation. Can be adaptively reduced.

[Twelfth embodiment]

  FIG. 20 is a block diagram of a twelfth embodiment of the image quality improving apparatus according to the present invention.

  The image quality improvement apparatus in FIG. 20 is a configuration example when the picture interval M between the I picture and the P picture is 3, and correction is performed in units of frames.

  The image quality improvement apparatus in FIG. 20 includes an image feature amount detection apparatus including a blocking unit 186, a nonlinear conversion unit 187, an average value calculation unit 188, an average value calculation unit 189, a nonlinear conversion unit 190, and a subtraction unit 191, a delay 192, , An average value calculation unit 193, a delay 194, a subtraction unit 195, a switch 197, a subtraction unit 198, a non-linear inverse transformation unit 199, and an image reconstruction unit 200. A moving image signal having an electric signal level, which is an input signal to the image quality improving apparatus, is input to the blocking unit 186.

  When the moving image signal is input, the blocking unit 186 performs blocking processing for dividing the screen into a plurality of blocks on the moving image signal, and converts the moving image signal subjected to the blocking processing to the nonlinear conversion unit 187 and The result is output to the average value calculation unit 189. Note that the image feature quantity detection device including the blocking unit 186, the nonlinear conversion unit 187, the average value calculation unit 188, the average value calculation unit 189, the nonlinear conversion unit 190, and the subtraction unit 191 is the same as the image feature quantity detection device of FIG. The description is omitted because it is similar. The subtracting unit 191 outputs the flicker component data to the delay 192, the average value calculating unit 193, and the subtracting unit 195.

  The delay 192 delays one frame for flicker component data of an input high-luminance frame or a frame other than the high-luminance frame. The average value calculator 193 receives the flicker component data from the subtractor 191 and the flicker component data delayed by one frame from the delay 192. That is, the two flicker component data input to the average value calculation unit 193 has a difference of one frame in time. The average value calculation unit 193 calculates the average value of the two input flicker component data for each block and outputs it to the delay 194.

  The delay 194 delays the average value of the two flicker component data input from the average value calculation unit 193 by one frame and outputs it to the subtraction unit 195. The subtraction unit 195 subtracts the flicker component data input from the subtraction unit 191 and the flicker component data input from the delay 194 for each block having the same spatial position, and opens and closes the difference as a flicker correction value. Output to the unit 197.

  The switch 197 operates with reference to the input identification signal. The switch 197 outputs the flicker correction value input from the subtraction unit 195 to the subtraction unit 198 when the identification signal indicates a high luminance frame. The switch 197 does not output the flicker correction value input from the subtraction unit 195 to the subtraction unit 198 when the identification signal indicates a frame other than the high luminance frame.

  For example, when the identification signal indicates a high-luminance frame, the flicker correction value output from the subtraction unit 195 includes the flicker component data of the high-luminance frame input from the subtraction unit 191 and the frames other than the high-luminance frame input from the delay 194. The average of the flicker component data is subtracted for each block having the same spatial position.

  The subtracting unit 198 has the same spatial and temporal position between the moving image signal of the optical signal level subjected to the blocking processing input from the nonlinear conversion unit 187 and the flicker correction value input from the switch 197. A subtraction process is performed for each pixel of a certain block.

  For example, when the identification signal indicates a frame other than the high-luminance frame, since the flicker correction value is not input from the switch 197, the subtractor 198 has the optical signal level subjected to the blocking processing input from the nonlinear converter 187. The moving image signal is output to the nonlinear inverse transform unit 199.

  On the other hand, when the identification signal indicates a high-luminance frame, the subtraction unit 198 receives the flicker correction value input from the switch 197 and the moving image signal of the optical signal level subjected to the blocking processing input from the nonlinear conversion unit 187. Is subtracted for each pixel of the block having the same spatial and temporal positions, and output to the nonlinear inverse transform unit 199.

  When a moving image signal having an optical signal level is input, the nonlinear inverse converting unit 199 performs nonlinear inverse conversion on the moving image signal having the optical signal level into a moving image signal having an electric signal level, and outputs the moving image signal to the image reconstruction unit 200. .

  When the moving image signal subjected to the block processing is input from the nonlinear inverse transform unit 199, the image reconstructing unit 200 returns the blocked moving image signal to the original position and reconstructs the moving image. And output as a corrected moving image signal.

  The image quality improvement apparatus in FIG. 20 is a configuration example when the picture interval M is 3, but the configuration is the same even when the value of the picture interval M changes or when correction is performed in units of fields. Yes. In other words, the average value calculation unit 193 adjusts the arrangement and delay amount of delays before and after the average value calculation unit 193 so that the average of flicker component data of consecutive frames or fields other than the luminance frame or field can be calculated. do it.

  The image quality improvement apparatus of FIG. 20 uses a difference between the flicker component data of a high-luminance frame or field and the average of the flicker component data of a frame or field other than the high-luminance frame or field as a flicker correction value, and After subtracting the flicker correction value from the image signal, it is output as a corrected moving image signal. Note that the image quality improvement apparatus in FIG. 20 outputs a corrected moving image signal without subtracting the flicker correction value from the moving image signal of a frame or field other than the high luminance frame or field.

  The image quality improvement apparatus in FIG. 20 performs the above signal processing repeatedly so that flicker degradation caused by changes in the frequency characteristics of moving images is adaptive for each screen region by correcting the level value for each pixel using a block. Can be reduced.

  In the image quality improvement apparatuses of the fifth to twelfth embodiments, the identification signal is obtained from outside the image quality improvement apparatus. However, a determination unit that outputs the identification signal may be provided. Here, the configuration of FIG. 21 in which the determination unit is provided in the image quality improvement apparatus of the fifth embodiment will be described, but the image quality improvement apparatus of other embodiments can also be handled with the same configuration.

[Thirteenth embodiment]

  FIG. 21 is a block diagram of a thirteenth embodiment of the image quality improving apparatus according to the present invention. 21 is the same as the image quality improvement apparatus in FIG. 13 except for the determination unit 210, and thus the same parts are denoted by the same reference numerals and description thereof is omitted.

  The determination unit 210 receives the flicker component data output from the subtraction unit 105 and the moving image signal at the electrical signal level input to the image quality improvement apparatus. The determination unit 210 outputs an identification signal for identifying a high-luminance frame or field to the memory 106 and the switch 109 based on the input flicker component data.

  The determination unit 210 delays the flicker component data input from the subtraction unit 105 and outputs the delayed data to the memory 106 and the subtraction unit 107. Further, the determination unit 210 delays the input moving image signal of the electric signal level and outputs the delayed moving image signal to the addition unit 110.

  Here, a configuration example of the determination unit 210 will be described. FIG. 22 is a configuration diagram of an example of the determination unit 210. The determination unit 210 in FIG. 22 is a configuration example in the case of obtaining a high-luminance frame having the highest average optical signal level output from the display device among three consecutive frames and outputting an identification signal. .

  22 includes an average value calculation unit 211, a delay 212, a delay 213, a maximum value determination unit 214, a delay 215, and a delay 216. The flicker component data input to the determination unit 210 is input to the average value calculation unit 211 and the delay 215. Further, the moving image signal at the electric signal level input to the determination unit 210 is input to the delay 216.

  The average value calculation unit 211 calculates an average value (hereinafter referred to as “one frame average value”) of the input flicker component data for each frame, and outputs it to the delay 212 and the maximum value determination unit 214. When the average value of one frame is input from the average value calculation unit 212, the delay 212 outputs a delay of one frame to the delay 213 and the maximum value determination unit 214. When the average value of one frame is input from the delay 212, the delay 213 outputs a delay for one frame to the maximum value determination unit 214. The determination unit 210 repeatedly performs the above operation.

  In other words, the maximum value determination unit 214 is always input with an average value of three consecutive frames. The maximum value determination unit 214 compares the 1-frame average values of three consecutive frames, and outputs an identification signal when the 1-frame average value input from the average value calculation unit 211 is the largest.

  When the flicker component data is input, the delay 215 outputs the data with a delay of one frame. The delay performed by the delay 215 is to match the delay in the calculation in the average value calculation unit 211.

  Similarly, when a moving image signal having an electric signal level is input to the delay 216, the delay 216 outputs it with a delay of one frame. The delay performed by the delay 216 is to match the delay in the calculation in the average value calculation unit 211.

  22 outputs an identification signal for identifying a high-luminance frame when the average value of one frame output from the average value calculation unit 211 is the largest. Alternatively, an identification signal for identifying a high-luminance frame can be output when the average value of one frame output from the delay 213 is the largest.

  For example, when outputting an identification signal for identifying a high-luminance frame when the average value of one frame output from the delay 212 is the largest, in addition to the delays of the delays 215 and 216, one frame worth respectively. Need to add a delay. When outputting an identification signal for identifying a high-luminance frame when the average value of one frame output from the delay 213 is the largest, in addition to the delays of the delays 215 and 216, delays of two frames respectively. Need to be added.

  As described above, the determination unit 210 in FIG. 22 can output an identification signal for identifying a frame that visually appears to have the highest luminance due to the influence of the gamma characteristic of the display device among three consecutive frames. Therefore, the image quality improvement apparatus in FIG. 21 can perform flicker correction without obtaining an identification signal from the outside of the image quality improvement apparatus.

  The determination unit 210 in FIG. 22 is a configuration example in the case of obtaining a high-luminance frame having the highest average optical signal level output from the display device among three consecutive frames and outputting an identification signal. However, any number of frames or fields can be handled with the same configuration. That is, the maximum value determination unit 214 adjusts the delay amount and arrangement of the delays constituting the determination unit 210 so that the average value of flicker component data of an arbitrary number of frames or fields can be compared, and the average value of the flicker component is determined. An identification signal for identifying the largest frame or field may be output.

  Note that FIG. 22 illustrates the configuration of the determination unit 210 when the number of frames serving as the calculation unit of the identification signal is three. However, in actual operation, it is required that the number of frames or fields that are the calculation unit of the identification signal be equal to or greater than the picture interval M at the time of encoding in terms of the number of frames.

  For example, if the encoding apparatus has performed encoding with a picture interval M of up to 3, the number of frames serving as a calculation unit of identification information is required to be 3 or more. In addition, the smaller the number of frames as a unit for calculating the identification information, the more accurate the flicker correction can be performed in terms of time. Therefore, it is desirable that the number of frames be the smallest number more than the necessary number.

  In the image quality improvement apparatuses of the fifth to thirteenth embodiments, a determination unit that performs determination on the calculated flicker correction value may be provided. Here, the configuration of FIG. 23 in which the determination unit is provided in the image quality improvement apparatus of the fifth embodiment will be described. However, image quality improvement apparatuses of other embodiments can be handled with the same configuration.

[14th embodiment]

  FIG. 23 is a block diagram of a fourteenth embodiment of the image quality improving apparatus according to the present invention. 23 is the same as the image quality improving apparatus in FIG. 13 except for the determination unit 108, the same parts are denoted by the same reference numerals and the description thereof is omitted.

  The determination unit 108 performs the following determination on the flicker correction value input from the subtraction unit 107. By performing the determination with respect to the flicker correction value, it is possible to prevent flicker correction with respect to a change in the original luminance of the moving image, such as a quick movement of the subject or a scene change.

  For example, since it is impossible for the flicker correction value to take a negative value for normal flicker, the determination unit 108 sets the negative flicker correction value to 0. Further, the determination unit 108 determines the upper limit value of the flicker correction value in consideration of the gamma characteristic of the display device and the like, and sets the flicker correction value exceeding the upper limit value that is unlikely to be normally taken to be 0. Then, the determination unit 108 outputs the flicker correction value to the switch 109. If the flicker correction value is 0, the adding unit 110 does not perform flicker correction on the pixel.

  Therefore, the image quality improvement apparatus in FIG. 23 performs flicker correction for a change in the original luminance of a moving image, and can prevent image quality deterioration due to erroneous correction. That is, the image quality improvement apparatus in FIG. 23 can leave only the original luminance change of the moving image as a correction target, leaving only the luminance change that causes flicker degradation.

  In the image quality improvement apparatuses according to the fifth to fourteenth embodiments, a determination unit that performs determination on a moving image signal (hereinafter referred to as a processing signal) having an electric signal level subjected to narrowband low-pass filtering processing is provided. You can also Here, the configuration of FIG. 24 in which the determination unit is provided in the image quality improvement apparatus of the fifth embodiment will be described. However, the image quality improvement apparatus of other embodiments can be handled with the same configuration.

[15th embodiment]

  FIG. 24 is a block diagram of a fifteenth embodiment of image quality improving apparatus according to the present invention. 24 is the same as the image quality improvement apparatus in FIG. 13 except for the determination unit 300, and therefore, the same parts are denoted by the same reference numerals and description thereof is omitted.

  The determination unit 300 receives a processing signal input from the narrowband LPF 104 and an identification signal input from the outside of the image quality improvement apparatus. The determination unit 300 calculates a difference in the processing signal output from the narrowband LPF 104 between the high luminance frame or field and a frame or field other than the high luminance frame or field, and according to the absolute value of the difference Determine whether to perform flicker correction.

  Specifically, the determination unit 300 calculates the difference between the processing signals output from the narrowband LPF 104 between the high luminance frame or field and the high luminance frame or frame other than the field, and the absolute value of the difference is calculated. When the value is equal to or greater than the predetermined value, it is determined that the original luminance change of the moving image is present, and it is determined that the flicker correction is not performed. In addition, when the absolute value of the difference is equal to or less than the predetermined value, the determination unit 300 determines that flicker correction is to be performed by determining that the change is not an original luminance change of the moving image.

  The determination unit 300 outputs to the switch 109 a switch control signal indicating a determination result of whether or not to perform flicker correction. In addition, the determination unit 300 outputs the processing signal input from the narrowband LPF 104 to the subtraction unit 105.

  The switch 109 outputs the flicker correction value input from the subtractor 107 to the adder 110 when the switch control signal represents a determination result of performing flicker correction. Further, the switch 109 does not output the flicker correction value input from the subtractor 107 to the adder 110 when the switch control signal indicates a determination result that does not perform flicker correction.

  By performing the determination on the difference between the processing signals output from the narrow band LPF 104, it is possible to prevent flicker correction with respect to a change in the original luminance of the moving image such as a quick movement of the subject or a scene change.

  Here, a configuration example of the determination unit 300 will be described. FIG. 25 is a configuration diagram of an example of the determination unit 300. 25 includes a memory 301, a subtraction unit 302, a difference determination unit 303, and a comparison unit 304. The processing signal input to the determination unit 300 is input to the memory 301 and the subtraction unit 302. The identification signal input to the determination unit 300 is input to the memory 301 and the comparison unit 304.

  The memory 301 operates with reference to the input identification signal. The memory 301 writes a processing signal only when a processing signal of a high luminance frame or field is input. At this time, no processing signal is read from the memory 301. Further, when a processing signal other than a high-luminance frame or field is input, the memory 301 does not write the processing signal. At this time, the memory 301 outputs the retained processing signal to the subtraction unit 302.

  The subtraction unit 302 outputs a difference between the processing signal input from the memory 301 and the processing signal input from the narrowband LPF 104 to the difference determination unit 303. When the absolute value of the difference input from the subtraction unit 302 is greater than or equal to a predetermined value, the difference determination unit 303 outputs an addition availability signal indicating that flicker correction is not performed to the comparison unit 304. Further, when the absolute value of the difference input from the subtraction unit 302 is equal to or less than a predetermined value, the difference determination unit 303 outputs an addition availability signal indicating that flicker correction is performed to the comparison unit 304.

  The comparison unit 304 compares the identification signal input from the outside of the determination unit 300 with the addition permission signal input from the difference determination unit 303, and performs a logical operation. Specifically, the comparison unit 304 indicates a switch control signal for closing the switch 109 when the identification signal indicates a frame or field other than the high-luminance frame or field and the addition / non-permission signal indicates that flicker correction is performed. In other cases, a switch control signal for opening the switch 109 is output.

  Therefore, the image quality improvement apparatus shown in FIG. 24 can prevent flicker correction due to an erroneous correction by performing flicker correction for a change in the original luminance of a moving image. That is, the image quality improvement apparatus in FIG. 24 can leave only a change in the original luminance of the moving image as a correction target, leaving only a change in luminance that causes flicker degradation.

  The image quality improving apparatus according to the present invention can be applied to a display device, a receiver, a conversion device connected between the display device and the receiver, and the like.

  FIG. 26 is a block diagram of an example of a display device provided with an image quality improving apparatus according to the present invention. A digital broadcasting television 400 as an example of a display device includes a TS decoding unit 401, an MPEG-2 demultiplexing unit 402, a data decoding unit 403, a video decoding unit 404, an audio decoding unit 405, a display unit 406, and a speaker unit 407. And an image quality improvement apparatus 500.

  Broadcast waves received by an antenna (not shown) are input to the TS decoding unit 401. In the case of satellite digital broadcasting, the input broadcast wave is a signal frequency-converted to an intermediate frequency band (IF band).

  The TS decoding unit 401 performs channel selection and detection, decodes a transport stream (hereinafter referred to as TS) signal, and outputs the decoded signal to the MPEG-2 multiplexing / demultiplexing unit 402. The MPEG-2 demultiplexing unit 402 separates the input TS signal into video, audio, and data element signals, and the video, audio, and data element signals are converted into a video decoding unit 404 and an audio decoding unit 405, respectively. And output to the data decoding unit 403.

  The data decoding unit 403 decodes the data signal. In the video decoding unit 404, the video signal is decoded. Also, the audio decoding unit 405 decodes the audio signal. The data signal is appropriately presented in cooperation with video and audio. The video signal (moving image signal) decoded by the video decoding unit 404 is input to the image quality improvement apparatus 500 according to the present invention.

  The image quality improving apparatus 500 performs the above-described processing for improving the image quality on the input video signal, and outputs it to the display unit 406. The audio decoding unit 405 outputs the decoded audio signal to the speaker unit 407.

  The display unit 406 displays the input video signal as a video. The speaker unit 407 outputs the input audio signal as audio. Therefore, the digital broadcast television 400 of FIG. 26 can adaptively reduce the flicker degradation caused by the change in the frequency characteristic of the moving image for each screen area.

  26 may be accommodated in a single casing or may be distributed and accommodated in a plurality of casings. An example in which various processing units are accommodated in a plurality of cases is shown in FIGS.

  FIG. 27 is a block diagram of an example of a receiver provided with an image quality improving apparatus according to the present invention. The receiver 600 includes a TS decoding unit 401, an MPEG-2 demultiplexing unit 402, a data decoding unit 403, a video decoding unit 404, an audio decoding unit 405, and an image quality improvement apparatus 500. The display device 601 includes a display unit 406 and a speaker unit 407.

  27 are the same as the various processing units constituting the display device 400 of FIG. 26, and the same parts are denoted by the same reference numerals and description thereof is omitted. . The receiver 600 in FIG. 27 can adaptively reduce flicker degradation caused by changes in the frequency characteristics of moving images for each screen area. Therefore, the image quality of the moving image displayed on the display device 601 can be improved.

  FIG. 28 is a block diagram of an example of a conversion apparatus provided with an image quality improvement apparatus according to the present invention. The conversion device 701 is connected between the receiver 700 and the display device 702. The receiver 700 in FIG. 28 includes a TS decoding unit 401, an MPEG-2 demultiplexing unit 402, a data decoding unit 403, a video decoding unit 404, and an audio decoding unit 405. The conversion device 701 includes an image quality improvement device 500 according to the present invention. The display device 702 includes a display unit 406 and a speaker unit 407.

  Note that the various processing units constituting the receiver 700, the conversion device 701, and the display device 702 in FIG. 28 are the same as the various processing units constituting the display device 400 in FIG. Description is omitted. The conversion device 701 in FIG. 28 can adaptively reduce flicker degradation caused by changes in the frequency characteristics of moving images for each screen area. Therefore, the image quality of the moving image displayed on the display device 702 can be improved.

  In the present invention, spatial convolution calculation and screen blocking are applied to flicker degradation that appears as a luminance change that did not exist in the original moving image due to the change in the frequency characteristic of the decoded moving image due to the gamma characteristic of the display device. Flicker correction is used.

  According to the present invention, it is possible to quantitatively obtain the degree of occurrence of flicker due to fluctuations in the frequency characteristics of moving images for each screen area. In addition, according to the present invention, the correction is performed only on the screen area where flicker occurs, the correction value is determined, and the original luminance change of the moving image is not corrected. Flicker correction is possible. Furthermore, according to the present invention, flicker correction can be performed without obtaining information about a frame or a field from a decoding device or the like.

  The present invention is not limited to the specifically disclosed embodiments, and various modifications and changes can be made without departing from the scope of the claims.

It is a figure showing an example of arrangement | positioning of an encoding moving image. It is a figure showing the generation principle of flicker. It is a block diagram of 1st Example of the image feature-value detection apparatus by this invention. It is a block diagram of an example of the convolution mask which narrowband LPF uses. FIG. 6 is a graph showing flicker components obtained at two points having different spatial positions in a moving image signal in which flicker is actually generated. It is a block diagram of 2nd Example of the image feature-value detection apparatus by this invention. It is a block diagram of 3rd Example of the image feature-value detection apparatus by this invention. It is a block diagram of 4th Example of the image feature-value detection apparatus by this invention. 1 is a configuration diagram of a first embodiment of an image quality improvement apparatus according to the present invention. It is a block diagram of 2nd Example of the image quality improvement apparatus by this invention. It is a block diagram of 3rd Example of the image quality improvement apparatus by this invention. It is a block diagram of 4th Example of the image quality improvement apparatus by this invention. It is a block diagram of 5th Example of the image quality improvement apparatus by this invention. It is a block diagram of 6th Example of the image quality improvement apparatus by this invention. It is a block diagram of 7th Example of the image quality improvement apparatus by this invention. It is a block diagram of 8th Example of the image quality improvement apparatus by this invention. It is a block diagram of 9th Example of the image quality improvement apparatus by this invention. It is a block diagram of 10th Example of the image quality improvement apparatus by this invention. It is a block diagram of 11th Example of the image quality improvement apparatus by this invention. It is a block diagram of 12th Example of the image quality improvement apparatus by this invention. It is a block diagram of 13th Example of the image quality improvement apparatus by this invention. It is a block diagram of an example of a determination part. It is a block diagram of 14th Example of the image quality improvement apparatus by this invention. It is a block diagram of 15th Example of the image quality improvement apparatus by this invention. It is a block diagram of an example of a determination part. It is a block diagram of an example of the display apparatus provided with the image quality improvement apparatus by this invention. It is a block diagram of an example of the receiver provided with the image quality improvement apparatus by this invention. It is a block diagram of an example of the converter provided with the image quality improvement apparatus by this invention.

Explanation of symbols

1,7,12,15,18,21 Nonlinear conversion unit 2,4,13,14 Narrow band LPF
3 Nonlinear inverse transformation unit 5, 11, 16, 22, 107, 131, 134, 181, 184 Subtraction unit 6, 17 Blocking unit 8, 10, 19, 20, 129, 179 Average value calculation unit 9, 159, 185 Nonlinear inverse transform unit 106 Memory 109, 133, 183 Switch 110 Adder unit 122, 148 Image reconstruction unit 128, 130, 178, 180, Delay 108, 210, 300 Judgment unit 400 Digital broadcasting television 500 Image quality improvement device 600 , 700 Receiver 601 702 Display device 701 Conversion device

Claims (23)

An optical signal level averaging means for converting an input image signal from an electric signal level to an optical signal level according to a predetermined electric signal-optical signal conversion characteristic, and averaging the image signal at the optical signal level;
Electrical signal level averaging means for averaging the image signals of the input electrical signal level;
Output means for converting one of the averaged optical signal level or electrical signal level image signal into the other electrical signal level or optical signal level image signal, and outputting the difference as an image feature amount for each at least one pixel; An image feature quantity detection device characterized by comprising: The optical signal level averaging means includes a non-linear conversion unit that converts an input image signal from an electric signal level to an optical signal level according to predetermined electric signal-optical signal conversion characteristics;
A first low-pass filter unit that performs a low-pass filter process on the image signal converted to the optical signal level by the nonlinear conversion unit;
The electrical signal level averaging means has a second low-pass filter unit that performs low-pass filter processing on the input image signal of the electrical signal level,
The output means includes a non-linear inverse conversion unit that converts an optical signal level image signal that has passed through the first low-pass filter unit from an optical signal level to an electrical signal level according to a predetermined electrical signal-optical signal conversion characteristic; ,
An image that calculates a difference between an image signal converted to an electric signal level by the non-linear inverse conversion unit and an image signal at an electric signal level that has passed through the second low-pass filter unit for each pixel, and outputs the difference as an image feature amount The image feature quantity detection apparatus according to claim 1, further comprising a feature quantity output unit. Further comprising a dividing means for dividing the input image signal into a region composed of a predetermined number of pixels;
The optical signal level averaging means includes a non-linear conversion unit that converts the regionalized image signal from an electric signal level to an optical signal level according to predetermined electric signal-optical signal conversion characteristics;
A first average value calculation unit that averages the image signal converted into the optical signal level by the nonlinear conversion unit for each region;
The electrical signal level averaging means has a second average value calculation unit that averages the image signal of the electrical signal level that has been regionalized for each region,
The output means includes a non-linear inverse conversion unit that converts the image signal of the optical signal level averaged by the first average value calculation unit from an optical signal level to an electrical signal level according to a predetermined electrical signal-optical signal conversion characteristic; ,
The difference between the image signal converted to the electric signal level by the non-linear inverse conversion unit and the image signal of the electric signal level averaged by the second average value calculation unit is calculated for each region and output as an image feature amount The image feature quantity detection apparatus according to claim 1, further comprising a feature quantity output unit. The optical signal level averaging means includes a first non-linear converter that converts an input image signal from an electric signal level to an optical signal level according to predetermined electric signal-optical signal conversion characteristics;
A first low-pass filter unit that performs low-pass filter processing on the image signal converted to the optical signal level by the first nonlinear conversion unit;
The electrical signal level averaging means has a second low-pass filter unit that performs low-pass filter processing on the input image signal of the electrical signal level,
The output means converts the image signal of the electric signal level averaged by the second low-pass filter unit from an electric signal level to an optical signal level according to a predetermined electric signal-optical signal conversion characteristic. And
The difference between the image signal of the electrical signal level that has passed through the first low-pass filter unit and the image signal converted to the optical signal level by the second nonlinear conversion unit is calculated for each pixel and output as an image feature amount The image feature amount detection apparatus according to claim 1, further comprising an image feature amount output unit. Further comprising a dividing means for dividing the input image signal into a region composed of a predetermined number of pixels;
The optical signal level averaging means includes a first non-linear conversion unit that converts the regionized image signal from an electric signal level to an optical signal level according to predetermined electric signal-optical signal conversion characteristics;
A first average value calculation unit that averages the image signal converted into the optical signal level by the first nonlinear conversion unit for each region;
The electrical signal level averaging means has a second average value calculation unit that averages the image signal of the electrical signal level that has been regionalized for each region,
The output means converts the image signal of the optical signal level averaged by the second average value calculation unit from an optical signal level to an electric signal level according to a predetermined electric signal-optical signal conversion characteristic. When,
The difference between the image signal of the electric signal level averaged by the first average value calculation unit and the image signal converted to the electric signal level by the second nonlinear conversion unit is calculated for each region and output as an image feature amount. The image feature amount detection apparatus according to claim 1, further comprising an image feature amount output unit. An optical signal level averaging means for converting an input image signal from an electric signal level to an optical signal level according to a predetermined electric signal-optical signal conversion characteristic, and averaging the image signal at the optical signal level;
Electrical signal level averaging means for averaging the image signals of the input electrical signal level;
Output means for converting one of the averaged optical signal level or electrical signal level image signal into the other electrical signal level or optical signal level image signal, and outputting the difference as an image feature amount for each at least one pixel; ,
An image quality improving apparatus comprising: correction means for correcting the input image signal in accordance with the output image feature amount. The correction unit includes a correction unit that corrects the input image signal for each pixel having the same spatial position between screens using the image feature amount output from the output unit. The image quality improving apparatus according to claim 6. A dividing unit that divides the input image signal into a region including a predetermined number of pixels;
The correction unit includes a correction unit that corrects the input image signal for each pixel in a region having the same spatial position between screens using the image feature amount output from the output unit. The image quality improving apparatus according to claim 6. The predetermined number of continuous screens of the input image signals are composed of a high brightness screen with the highest average optical signal level and other screens,
The correction unit calculates the difference between the image feature amount of the high brightness screen output from the output unit and the image feature amount of the screen other than the high brightness screen for each pixel having the same spatial position between the screens, A correction value output unit that outputs the correction value of the pixel;
The image quality improvement apparatus according to claim 6, further comprising: a correction unit that corrects the input image signal for each pixel having the same spatial position between the screens using the output correction value. . Further comprising a dividing means for dividing the input image signal into a region composed of a predetermined number of pixels;
The predetermined number of continuous screens of the input image signals are composed of a high brightness screen with the highest average optical signal level and other screens,
The correction unit calculates the difference between the image feature amount of the high brightness screen output from the output unit and the image feature amount of the screen other than the high brightness screen for each region having the same spatial position between the screens, A correction value output unit that outputs the correction value of the region;
7. The image quality according to claim 6, further comprising: a correction unit that corrects the input image signal for each pixel in a region having the same spatial position between the screens using the output correction value. Improvement device. The optical signal level averaging means includes a non-linear conversion unit that converts an input image signal from an electric signal level to an optical signal level according to predetermined electric signal-optical signal conversion characteristics;
A first low-pass filter unit that performs a low-pass filter process on the image signal converted to the optical signal level by the nonlinear conversion unit;
The electrical signal level averaging means has a second low-pass filter unit that performs low-pass filter processing on the input image signal of the electrical signal level,
The output means includes a non-linear inverse conversion unit that converts an optical signal level image signal that has passed through the first low-pass filter unit from an optical signal level to an electrical signal level according to a predetermined electrical signal-optical signal conversion characteristic; ,
The difference between the image signal converted to the electric signal level by the nonlinear inverse conversion unit and the image signal of the electric signal level that has passed through the second low-pass filter unit is calculated for each pixel and output as an image feature amount The image quality improving apparatus according to claim 7, further comprising a feature amount output unit. The optical signal level averaging means includes a non-linear conversion unit that converts the regionalized image signal from an electric signal level to an optical signal level according to predetermined electric signal-optical signal conversion characteristics;
A first average value calculation unit that averages the image signal converted into the optical signal level by the nonlinear conversion unit for each region;
The electrical signal level averaging means has a second average value calculation unit that averages the image signal of the electrical signal level that has been regionalized for each region,
The output means includes a non-linear inverse conversion unit that converts the image signal of the optical signal level averaged by the first average value calculation unit from an optical signal level to an electrical signal level according to a predetermined electrical signal-optical signal conversion characteristic; ,
The difference between the image signal converted to the electric signal level by the non-linear inverse conversion unit and the image signal of the electric signal level averaged by the second average value calculation unit is calculated for each region and output as an image feature amount 11. The image quality improving apparatus according to claim 8, further comprising a feature amount output unit. The correction means uses an image feature amount output from the output means to convert an image signal converted from an electric signal level to an optical signal level according to the electric signal-optical signal conversion characteristics between each screen. A correction unit that corrects each pixel having the same position;
And a non-linear inverse conversion unit that converts the optical signal level image signal corrected by the correction unit from an optical signal level to an electrical signal level according to predetermined electrical signal-optical signal conversion characteristics. 6. The image quality improving apparatus according to 6. A dividing unit that divides the input image signal into a region including a predetermined number of pixels;
The correction unit uses the image feature amount output from the output unit to convert an image signal converted from an electric signal level to an optical signal level according to the electric signal-optical signal conversion characteristics between spatial positions between screens. A correction unit that corrects each pixel in a region where the
And a non-linear inverse conversion unit that converts the optical signal level image signal corrected by the correction unit from an optical signal level to an electrical signal level in accordance with predetermined electrical signal-optical signal conversion characteristics. 6. The image quality improving apparatus according to 6. The predetermined number of continuous screens of the input image signals are composed of a high brightness screen with the highest average optical signal level and other screens,
The correction unit calculates the difference between the image feature amount of the high brightness screen output from the output unit and the image feature amount of the screen other than the high brightness screen for each pixel having the same spatial position between the screens, A correction value output unit that outputs the correction value of the pixel;
Using the output correction value, the image signal converted from the electric signal level to the optical signal level according to the electric signal-optical signal conversion characteristic is corrected for each pixel having the same spatial position between the screens. A correction unit to perform,
And a non-linear inverse conversion unit that converts the optical signal level image signal corrected by the correction unit from an optical signal level to an electrical signal level in accordance with predetermined electrical signal-optical signal conversion characteristics. 6. The image quality improving apparatus according to 6. Further comprising a dividing means for dividing the input image signal into a region composed of a predetermined number of pixels;
The predetermined number of continuous screens of the input image signals are composed of a high brightness screen with the highest average optical signal level and other screens,
The correction unit calculates the difference between the image feature amount of the high brightness screen output from the output unit and the image feature amount of the screen other than the high brightness screen for each region having the same spatial position between the screens, A correction value output unit that outputs the correction value of the region;
Using the output correction value, an image signal converted from an electric signal level to an optical signal level in accordance with the electric signal-optical signal conversion characteristics for each pixel in a region where the spatial position is the same between the screens. A correction unit to correct,
And a non-linear inverse conversion unit that converts the optical signal level image signal corrected by the correction unit from an optical signal level to an electrical signal level in accordance with predetermined electrical signal-optical signal conversion characteristics. 6. The image quality improving apparatus according to 6. The optical signal level averaging means includes a first non-linear converter that converts an input image signal from an electric signal level to an optical signal level according to predetermined electric signal-optical signal conversion characteristics;
A first low-pass filter unit that performs low-pass filter processing on the image signal converted to the optical signal level by the first nonlinear conversion unit;
The electrical signal level averaging means has a second low-pass filter unit that performs low-pass filter processing on the input image signal of the electrical signal level,
The output means converts the image signal of the electric signal level averaged by the second low-pass filter unit from an electric signal level to an optical signal level according to a predetermined electric signal-optical signal conversion characteristic. And
The difference between the image signal of the electrical signal level that has passed through the first low-pass filter unit and the image signal converted to the optical signal level by the second nonlinear conversion unit is calculated for each pixel and output as an image feature amount 16. The image quality improving apparatus according to claim 13, further comprising an image feature amount output unit. The optical signal level averaging means includes a first non-linear conversion unit that converts the regionized image signal from an electric signal level to an optical signal level according to predetermined electric signal-optical signal conversion characteristics;
A first average value calculation unit that averages the image signal converted into the optical signal level by the first nonlinear conversion unit for each region;
The electrical signal level averaging means has a second average value calculation unit that averages the image signal of the electrical signal level that has been regionalized for each region,
The output means converts the image signal of the optical signal level averaged by the second average value calculation unit from an optical signal level to an electric signal level according to a predetermined electric signal-optical signal conversion characteristic. When,
The difference between the image signal of the electric signal level averaged by the first average value calculation unit and the image signal converted to the electric signal level by the second nonlinear conversion unit is calculated for each region and output as an image feature amount. 17. The image quality improving apparatus according to claim 14, further comprising an image feature amount output unit. A first determination unit that calculates an average of the image feature amounts output from the output unit for each screen and determines a screen having the largest average image feature amount as a high-luminance screen among the predetermined number of screens; The image quality improvement apparatus according to claim 7, further comprising:   The correction means further includes second determination means for determining whether or not the correction value of the pixel is within a predetermined range, and corrects the image signal when the correction value is within the predetermined range. The image quality improvement apparatus according to claim 7, wherein the image quality improvement apparatus is an image quality improvement apparatus.   The correction means is a pixel or region in which the spatial position is the same between the screens for the difference between the image feature quantity of the high brightness screen output from the electrical signal level averaging means and the image feature quantity of a screen other than the high brightness screen. And a third determination means for determining whether or not the difference is within a predetermined range, and correcting the image signal when the difference is within the predetermined range. The image quality improving apparatus according to any one of claims 7 to 18.   A display device comprising the image quality improvement device according to any one of claims 6 to 21.   A receiver comprising the image quality improving apparatus according to any one of claims 6 to 21.

Images