JP2006229623A - Encoder and method therefor, decoder and method therefor, recording medium, program, and image processing system and method therefor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To increase a deterioration level of image data by repeating encoding and decoding. <P>SOLUTION: A motion estimating unit 114 carries out motion search by using only pixels from top to the 3rd places of a frequency distribution of an original searching block in a digital image data Vdg1, and calculates motion vector. In a residual vector quantization unit 116; a residual after the motion estimation is subjected to vector quantization, and quantization code data is obtained. In a data synthesizing unit 117; the motion vector, quantization code data, and a code block 121 used for vector quantization, are generated as encoded data Vcd to a rear stage. The pixel used for the vector search changes a lot because of phase dislocation which is added to the image data Vdg1 as an object, and it becomes difficult to perform motion estimation correctly. The encoder and the decoder defined above can be applied to an image processing system for decoding image data from a recording medium and displaying it or encoding the image data and recording it in the recording medium. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an encoding device and method, a decoding device and method, a recording medium, a program, an image processing system and a method, and in particular, performs motion estimation by paying attention to a pixel frequency distribution in a block, and sets a residual as a vector quantum. The encoding apparatus and method, the decoding apparatus and method, and the recording medium that can suppress copying while maintaining good quality without degrading the quality of output by the data before copying And a program, an image processing system, and a method.

  FIG. 1 shows a configuration example of a conventional image processing system 1. The image processing system 1 includes a playback device 11 that outputs analog image data Van, and a display 12 that displays an image corresponding to the image data Van output from the playback device 11.

  The playback device 11 includes a decoding unit 21 and a D / A (Digital-to-Analog) conversion unit 22. The decoding unit 21 decodes encoded image data reproduced from a recording medium such as an optical disk (not shown), and supplies the decoded digital image data to the D / A conversion unit 22. The D / A conversion unit 22 converts the digital image data from the decoding unit 21 into analog image data, and supplies the analog image data Van to the display 12.

  The display 12 is composed of, for example, a CRT (Cathode Ray Tube) or an LCD (Liquid Crystal Display).

  Conventionally, there has been a risk of unauthorized copying using analog image data Van output from the playback device 11 of such an image processing system 1.

  That is, the analog image data Van output from the reproduction apparatus 11 is converted into digital image data Vdg by the A / D conversion unit 31 and supplied to the encoding unit 32. The encoding unit 32 encodes the digital image data Vdg and supplies the encoded image data Vcd to the recording unit 33. The recording unit 33 records the encoded image data Vcd on a recording medium such as an optical disk.

  In order to prevent unauthorized copying using analog image data Van that is performed as described above, when copyright protection has been conventionally performed, analog image data Van is scrambled and output ( For example, Patent Document 1) or the output of analog image data Vdg has been suppressed. However, in this case, there has been a problem that normal video does not move to the display 12.

  In addition, when encoding and decoding are performed using ADRC (Adaptive Dynamic Range Coding) shown in Patent Document 2, image data deteriorates due to a decrease in dynamic range due to encoding and decoding. The degree of reduction of the dynamic range by ADRC is not so much deterioration. In addition, in the case of ADRC, although it can be applied to a moving image, it does not utilize the characteristics of motion, so that a great deterioration cannot be obtained for the moving image.

  Therefore, a method for preventing unauthorized copying using an analog image signal without causing inconvenience such as no image being displayed has been proposed by the present applicant (see, for example, Patent Document 3).

JP 2001-245270 A Japanese Patent Laid-Open No. 61-144989 JP 2004-289585 A

  In the method described in Patent Literature 3, attention is paid to analog noise (distortion) such as phase shift of a digital image signal obtained by A / D conversion of the analog image signal, and attention is paid to analog noise for the digital image signal. By performing this encoding, it is impossible to copy while maintaining good quality without degrading the quality of the image before copying, thereby preventing unauthorized copying using analog image signals, but digital content In recent years, when the distribution of such information has become common, there has been a demand for a proposal of another method for preventing unauthorized copying as described above.

  The present invention has been made in view of such a situation, so that copying can be easily suppressed while maintaining good quality without deteriorating the quality of output by data before copying. To do.

  The encoding apparatus according to the present invention represents a block based on a blocking unit that blocks a first frame in input image data, and a frequency distribution of pixel values of the blocks that are blocked by the blocking unit. A pixel extraction unit that extracts pixels to be used, a motion estimation unit that estimates a motion vector of the block using the pixels extracted by the pixel extraction unit, and a motion vector estimated by the motion estimation unit. Difference calculating means for calculating a difference between the image data of the second frame to be processed and the image data in the block, and vector quantization means for performing vector quantization on the difference calculated by the difference calculating means. It is characterized by.

  The image processing apparatus further includes noise adding means for adding noise to the image data and outputting the image data to which the noise is added, and the blocking means blocks the first frame of the image data to which the noise is added by the noise adding means. Can be.

  The motion estimation means can estimate a motion vector by searching for a block having the minimum sum of squares of pixel differences from the second frame using the pixels extracted by the pixel extraction means. .

  The code book used for vector quantization by the vector quantization means can be generated based on input image data.

  Codebook used for vector quantization by vector quantization means, quantized data obtained by vector quantization by vector quantization means, and motion vector estimated by motion estimation means as encoded data It is possible to further include data output means for outputting the data.

  The pixel extraction means can extract pixels having a frequency distribution of pixel values of the block that are within the top three as pixels representing the block.

  The encoding method of the present invention includes a blocking step for blocking the first frame in the input image data, and a block based on the frequency distribution of the pixel values of the blocks blocked by the processing of the blocking step. A pixel extraction step for extracting a pixel representing the pixel, a motion estimation step for estimating a motion vector of the block using the pixel extracted by the processing of the pixel extraction step, and a motion vector estimated by the processing of the motion estimation step A difference calculation step for calculating a difference between the image data of the second frame corresponding to the block and the image data in the block, and vector quantization on the difference calculated by the processing of the difference calculation step. And a vector quantization step.

  The image processing method further includes a noise addition step of adding noise to the image data and inputting the noise-added image data to the blocking step processing. In the blocking step processing, the image to which noise has been added by the noise addition step processing The first frame of data can be blocked.

  In the processing of the motion estimation step, the motion vector is estimated by searching the second frame for the block having the minimum sum of squares of pixel differences using the pixels extracted by the processing of the pixel extraction step. can do.

  The code book used for vector quantization by the processing of the vector quantization step can be generated based on input image data.

  Codebook used for vector quantization by vector quantization step processing, quantized data obtained by vector quantization by vector quantization step processing, and motion vector estimated by motion estimation step processing A data output step of outputting the encoded data to the subsequent stage can be further included.

  In the process of the pixel extraction step, the pixels having the highest frequency distribution of the pixel values of the block can be extracted as the pixels representing the block.

  A recording medium on which the first program of the present invention is recorded includes a blocking step for blocking the first frame in the input image data, and pixel values of blocks blocked by the processing of the blocking step. Based on the frequency distribution, a pixel extraction step for extracting pixels representative of the block, a motion estimation step for estimating a motion vector of the block using the pixels extracted by the processing of the pixel extraction step, and processing of the motion estimation step The difference calculation step for calculating the difference between the image data of the second frame corresponding to the block and the image data in the block using the motion vector estimated by the step, and the difference calculated by the processing of the difference calculation step And a vector quantization step for performing vector quantization.

  The first program of the present invention is based on a blocking step for blocking the first frame in the input image data, and a frequency distribution of pixel values of the blocks blocked by the processing of the blocking step. A pixel extraction step for extracting a pixel representing the block, a motion estimation step for estimating a motion vector of the block using the pixels extracted by the processing of the pixel extraction step, and a motion vector estimated by the processing of the motion estimation step Is used to calculate the difference between the image data of the second frame corresponding to the block and the image data in the block, and the vector quantization is performed on the difference calculated by the processing of the difference calculation step. A vector quantization step to be performed.

  The decoding device according to the present invention includes a motion vector obtained using pixels representing a block extracted based on a frequency distribution of pixel values in a block in image data, and a plurality of frames calculated using the motion vector. Data input means for inputting quantized data obtained by performing vector quantization on the difference value between them, and extracting the difference value by inverse quantization of the quantized data input by the data input means The second frame image is obtained by adding the difference value extracted by the difference value extraction unit to the image data of the first frame in accordance with the motion vector input by the difference value extraction unit and the data input unit. And a data adding means for generating data.

  Noise addition means for adding noise to the image data generated by the data addition means and outputting the image data with the noise added to the subsequent stage can be further provided.

  The data input means also inputs a codebook used for vector quantization, and the difference value extraction means may inversely quantize the quantized data using the codebook input by the data input means. it can.

  According to the motion vector input by the data input means, the apparatus further comprises a motion compensation means for predicting the image data of the prediction block from the image data of the first frame, and the data addition means is a prediction predicted by the motion compensation means. By adding the difference value extracted by the difference value extraction means to the image data of the block, the image data of the second frame can be generated.

  The decoding method of the present invention includes a motion vector obtained using pixels representing a block extracted based on a frequency distribution of pixel values in a block in image data, and a plurality of frames calculated using the motion vector. A data input step for inputting quantized data obtained by performing vector quantization on the difference value between, and the difference value by dequantizing the quantized data input by the processing of the data input step By adding the difference value extracted by the difference value extraction step process to the image data of the first frame according to the difference value extraction step to be extracted and the motion vector input by the data input step process, And a data adding step for generating image data of the second frame.

  It is possible to further include a noise addition step of adding noise to the image data generated by the data addition step processing and outputting the noise-added image data to the subsequent stage.

  In the data input step processing, the code book used for vector quantization is also input. In the difference value extraction step processing, the quantized data is inversely quantized using the code book input by the data input step processing. To be able to.

  The method further includes a motion compensation step of predicting the image data of the prediction block from the image data of the first frame in accordance with the motion vector input by the data input step processing. The image data of the second frame can be generated by adding the difference value extracted by the processing of the difference value extraction step to the image data of the prediction block predicted by the processing.

  A recording medium on which the second program of the present invention is recorded includes a motion vector obtained by using a pixel representing a block extracted based on a frequency distribution of pixel values in a block in image data, and a motion vector The data input step for inputting the quantized data obtained by performing vector quantization on the difference value between multiple frames calculated using, and the quantized data input by the data input step processing are reversed. In accordance with the difference value extraction step for extracting the difference value by quantization and the motion vector input by the processing of the data input step, the image data of the first frame is extracted by the processing of the difference value extraction step. And a data addition step of generating image data of the second frame by adding the difference values.

  The second program of the present invention is calculated using a motion vector obtained by using a pixel representing a block extracted based on a frequency distribution of pixel values in a block in image data, and the motion vector. A data input step for inputting quantized data obtained by performing vector quantization on a difference value between multiple frames, and a difference by dequantizing the quantized data input by the data input step processing The difference value extracted by the process of the difference value extraction step is added to the image data of the first frame in accordance with the difference vector extraction step for extracting the value and the motion vector input by the process of the data input step. And a data addition step of generating image data of the second frame.

  In the first image processing system of the present invention, the encoding device includes a blocking unit that blocks the first frame in the input image data, and the frequency of the pixel values of the blocks blocked by the blocking unit. Based on the distribution, pixel extraction means for extracting pixels representing the block, motion estimation means for estimating a motion vector of the block using the pixels extracted by the pixel extraction means, and motion estimated by the motion estimation means Using the vector, the difference calculation means for calculating the difference between the image data of the second frame corresponding to the block and the image data in the block, and vector quantization on the difference calculated by the difference calculation means Vector quantization means.

  It is possible to further include noise adding means for adding noise to the image data from the decoding device and inputting the image data to which the noise has been added to the encoding device.

  According to the first image processing method of the present invention, the encoding method of the encoding device is blocked by the block step for blocking the first frame in the input image data and the block step processing. Based on the frequency distribution of the pixel value of the block, a pixel extraction step for extracting a pixel representing the block, a motion estimation step for estimating a motion vector of the block using the pixel extracted by the processing of the pixel extraction step, Using the motion vector estimated by the motion estimation step processing, the difference calculation step for calculating the difference between the image data of the second frame corresponding to the block and the image data in the block, and the calculation by the processing of the difference calculation step And a vector quantization step for performing vector quantization on the difference obtained.

  It is possible to further include a noise adding step of adding noise to the image data from the decoding device and inputting the image data to which the noise has been added to the encoding device.

  In the second image processing system according to the present invention, the decoding device includes a motion vector obtained by using a pixel representing a block extracted based on a frequency distribution of pixel values in a block in image data, and a motion vector. Data input means for inputting quantized data obtained by performing vector quantization on the difference value between multiple frames calculated using, and inverse quantization of the quantized data input by the data input means By adding the difference value extracted by the difference value extracting means to the image data of the first frame in accordance with the motion vector input by the data input means and the difference value extracting means for extracting the difference value by And data adding means for generating image data of the second frame.

  It is possible to further include noise adding means for adding noise to the image data from the decoding device and outputting the image data to which the noise has been added to the encoding device.

  According to a second image processing method of the present invention, the decoding method of the decoding device includes a motion vector obtained by using a pixel representing a block extracted based on a frequency distribution of pixel values in a block in image data. A data input step for inputting quantized data obtained by performing vector quantization on a difference value between a plurality of frames calculated using a motion vector, and a quantization input by processing of the data input step A difference value extraction step for extracting a difference value by dequantizing the data, and extraction of the first frame image data by the difference value extraction step processing according to the motion vector input by the data input step processing And a data addition step of generating image data of the second frame by adding the difference values thus obtained.

  It is possible to further include a noise adding step of adding noise to the image data from the decoding device and outputting the image data to which the noise has been added to the encoding device.

  In the first aspect of the present invention, the first frame in the input image data is blocked, and pixels representing the block are extracted and extracted based on the frequency distribution of the pixel values of the blocked blocks. The motion vector of the block is estimated using the obtained pixels. Then, using the estimated motion vector, a difference between the image data of the second frame corresponding to the block and the image data in the block is calculated, and vector quantization is performed on the calculated difference.

  In the second aspect of the present invention, a motion vector obtained using pixels representing a block extracted based on a frequency distribution of pixel values in a block in image data, and a plurality of motion vectors calculated using the motion vector. Quantized data obtained by performing vector quantization on the difference value between frames is input. Then, a difference value is extracted by dequantizing the input quantized data, and by adding the extracted difference value to the image data of the first frame according to the input motion vector, Image data of the second frame is generated.

  In the third aspect of the present invention, the first frame in the input image data is blocked, and pixels representing the block are extracted and extracted based on the frequency distribution of the pixel values of the blocked blocks. The motion vector of the block is estimated using the obtained pixels. Then, using the estimated motion vector, a difference between the image data of the second frame corresponding to the block and the image data in the block is calculated, and vector quantization is performed on the calculated difference.

  In the fourth aspect of the present invention, a motion vector obtained using a pixel representing a block extracted based on a frequency distribution of pixel values in a block in image data, and a plurality of motion vectors calculated using the motion vector. Quantized data obtained by performing vector quantization on the difference value between frames is input. Then, a difference value is extracted by dequantizing the input quantized data, and by adding the extracted difference value to the image data of the first frame according to the input motion vector, Image data of the second frame is generated.

  According to the present invention, the degree of degradation of image data can be increased by repeating encoding and decoding. Thus, according to the present invention, it is possible to easily suppress copying while maintaining good quality without degrading the quality of output by data before copying.

  Embodiments of the present invention will be described below. Correspondences between constituent elements described in the claims and specific examples in the embodiments of the present invention are exemplified as follows. This description is to confirm that specific examples supporting the invention described in the claims are described in the embodiments of the invention. Accordingly, although there are specific examples that are described in the embodiment of the invention but are not described here as corresponding to the configuration requirements, the specific examples are not included in the configuration. It does not mean that it does not correspond to a requirement. On the contrary, even if a specific example is described here as corresponding to a configuration requirement, this means that the specific example does not correspond to a configuration requirement other than the configuration requirement. not.

  Further, this description does not mean that all the inventions corresponding to the specific examples described in the embodiments of the invention are described in the claims. In other words, this description is an invention corresponding to the specific example described in the embodiment of the invention, and the existence of an invention not described in the claims of this application, that is, in the future, a divisional application will be made. Nor does it deny the existence of an invention added by amendment.

  The encoding device according to claim 1 (for example, the encoding device 63 in FIG. 2) is a blocking unit (for example, the blocking unit 111 in FIG. 5) that blocks the first frame in the input image data. ), A pixel extraction unit (for example, the motion estimation block generation unit 113 in FIG. 5) that extracts a pixel representing the block based on the frequency distribution of the pixel values of the blocks that are blocked by the blocking unit, and pixel extraction Corresponding to a block using a motion estimation unit (for example, the motion estimation unit 114 in FIG. 5) that estimates a motion vector of a block using the pixels extracted by the unit, and a motion vector estimated by the motion estimation unit Difference calculation means for calculating the difference between the image data of the second frame and the image data in the block (for example, the residual calculation unit 115 in FIG. 5), and difference calculation means For a greater calculated difference, characterized in that it comprises a vector quantization means for performing vector quantization (e.g., residual vector quantization unit 116 of FIG. 5).

  The encoding apparatus according to claim 2 further includes noise adding means (for example, an A / D conversion unit 81 in FIG. 2) for adding noise to the image data and outputting the image data to which the noise has been added. The converting means blocks the first frame of the image data to which noise is added by the noise adding means.

  In the encoding device according to claim 4, a code book (for example, the vector quantization code book 121 in FIG. 5) used for vector quantization by the vector quantization means is generated based on input image data. It is characterized by that.

  The encoding apparatus according to claim 5 is a codebook used for vector quantization by vector quantization means, quantized data obtained by performing vector quantization by vector quantization means, and estimation by motion estimation means. The apparatus further includes data output means (for example, the data synthesizer 117 in FIG. 5) that outputs the motion vector thus processed as encoded data to the subsequent stage.

  The encoding method according to claim 7 is blocked by processing of a blocking step (for example, step S21 in FIG. 10) for blocking the first frame in the input image data and the blocking step. Based on the frequency distribution of the pixel values of the block, a pixel extraction step (for example, step S22 in FIG. 10) for extracting a pixel representative of the block, and a pixel motion extracted using the pixel extraction process. Using the motion estimation step (for example, step S23 in FIG. 10) for estimating the vector and the motion vector estimated by the processing of the motion estimation step, the image data of the second frame corresponding to the block, and the image data in the block A difference calculating step (for example, step S24 in FIG. 10) for calculating the difference between On the difference calculated by the processing of flops, characterized in that it comprises a vector quantization step of performing vector quantization (e.g., step S25 in FIG. 10).

  The encoding method according to claim 8 further includes a noise adding step (for example, step S4 in FIG. 4) for adding noise to the image data and inputting the noise-added image data to the processing of the blocking step. The blocking step processing is characterized in that the first frame of image data to which noise has been added by the noise addition step processing is blocked.

  The encoding method according to claim 11 includes a codebook used for vector quantization by a vector quantization step process, quantized data obtained by performing vector quantization by a vector quantization step process, and motion It further includes a data output step (for example, step S26 in FIG. 10) that outputs the motion vector estimated by the processing of the estimation step to the subsequent stage as encoded data.

  Note that the recording medium according to claim 13 and the program according to claim 14 are basically the same processing as the encoding method according to claim 7 described above, and are therefore repeated, so that the description thereof is omitted. To do.

  The decoding device according to claim 15 (for example, the encoding device 63 in FIG. 2) is obtained by using a pixel representing a block extracted based on a frequency distribution of pixel values in a block in image data. Data input means for inputting a motion vector and quantized data obtained by performing vector quantization on a difference value between a plurality of frames calculated using the motion vector (for example, the data decomposition unit 211 in FIG. 17) ), Difference value extraction means (for example, the residual inverse vector quantization unit 212 in FIG. 17) for extracting a difference value by inverse quantization of the quantized data input by the data input means, and data input means By adding the difference value extracted by the difference value extraction means to the image data of the first frame according to the input motion vector, the image data of the second frame is added. Data addition means for generating data (e.g., the residual addition unit 215 of FIG. 17), characterized in that it comprises a.

  17. The decoding apparatus according to claim 16, wherein noise is added to the image data generated by the data addition means, and noise addition means for outputting the image data with the noise added to a subsequent stage (for example, D / A conversion in FIG. 2). Part 85).

  In the decoding device according to claim 17, the data input means also inputs a code book used for vector quantization (for example, the vector quantization code book 121 of FIG. 5), and the difference value extraction means receives the data input. The quantized data is inversely quantized using the code book input by the means.

  The decoding apparatus according to claim 18 is a motion compensation unit that predicts image data of a prediction block from image data of a first frame in accordance with a motion vector input by a data input unit (for example, motion of FIG. 17). The second frame image data by adding the difference value extracted by the difference value extraction means to the image data of the prediction block predicted by the motion compensation means. Is generated.

  The decoding method according to claim 19 is calculated using a motion vector obtained by using a pixel representing a block extracted based on a frequency distribution of pixel values in a block in image data, and the motion vector. A data input step (for example, step S211 in FIG. 21) for inputting quantized data obtained by performing vector quantization on a difference value between a plurality of frames, and a quantum input by processing of the data input step A difference value extraction step (for example, step S212 in FIG. 21) for extracting a difference value by inverse quantization of the quantized data, and an image of the first frame according to the motion vector input by the data input step processing The image data of the second frame is generated by adding the difference value extracted by the difference value extraction step to the data. Data addition step (e.g., step S214 in FIG. 21), characterized in that it comprises a.

  The decoding method according to claim 20 adds a noise to the image data generated by the processing of the data addition step, and outputs the noise-added image data to a subsequent stage (for example, step S7 in FIG. 4). (Or S2)).

  The decoding method according to claim 22 is a motion compensation step of predicting image data of a prediction block from image data of a first frame in accordance with a motion vector input by the data input step (for example, FIG. 21). In the data addition step processing, the difference value extracted by the difference value extraction step processing is added to the image data of the prediction block predicted by the motion compensation step processing. Image data of two frames is generated.

  Note that the recording medium according to claim 23 and the program according to claim 24 are basically the same processing as the decoding method according to claim 19 described above, and are therefore repeated, so the description thereof is omitted. .

  In the image processing system according to claim 25 (for example, the image processing system 51 in FIG. 2), the encoding device (for example, the encoding unit 82 in FIG. 2) uses the first frame in the input image data. Blocking means for blocking (for example, the blocking unit 111 in FIG. 5) and pixel extraction means for extracting pixels representing blocks based on the frequency distribution of the pixel values of the blocks blocked by the blocking means ( For example, the motion estimation block generation unit 113 in FIG. 5, the motion estimation unit (for example, the motion estimation unit 114 in FIG. 5) that estimates the motion vector of the block using the pixels extracted by the pixel extraction unit, A difference for calculating a difference between the image data of the second frame corresponding to the block and the image data in the block using the motion vector estimated by the estimation means An output unit (for example, the residual calculation unit 115 in FIG. 5) and a vector quantization unit for performing vector quantization on the difference calculated by the difference calculation unit (for example, the residual vector quantization unit 116 in FIG. 5). ).

  An image processing system according to a twenty-sixth aspect adds noise to image data from a decoding device (for example, the decoding unit 71 or the decoding unit 84 in FIG. 2), and the image data with the noise added to the encoding device. It further comprises an input noise adding means (for example, the A / D conversion unit 81 in FIG. 2).

  The image processing method according to claim 27, wherein the encoding method of the encoding device includes a blocking step (for example, step S21 in FIG. 10) for blocking the first frame in the input image data, The pixel extraction step (for example, step S22 in FIG. 10) for extracting the pixel representative of the block based on the frequency distribution of the pixel values of the block that has been blocked by the conversion step processing, and the pixel extraction step processing A second frame corresponding to the block using the motion estimation step (for example, step S23 in FIG. 10) for estimating the motion vector of the block using the obtained pixels and the motion vector estimated by the processing of the motion estimation step. The difference calculating step for calculating the difference between the image data of the image and the image data in the block (for example, the step of FIG. And-up S24), on the difference calculated by the processing of the difference calculation step, characterized in that it comprises a vector quantization step of performing vector quantization (e.g., step S25 in FIG. 10).

  The image processing method according to claim 28 adds noise to the image data from the decoding device, and inputs the noise-added image data to the encoding device (for example, step S4 in FIG. 4). Is further included.

  In the image processing system according to claim 29, the decoding device (for example, the decoding unit 84 in FIG. 2) represents a block extracted based on a frequency distribution of pixel values in the block in the input image data. A data input means for inputting quantized data obtained by performing vector quantization on a motion vector obtained using a pixel and a difference value between a plurality of frames calculated using the motion vector (for example, 17 and a difference value extraction unit (for example, a residual inverse vector quantization unit 212 in FIG. 17) that extracts a difference value by dequantizing the quantized data input by the data input unit. ) And the difference value extracted by the difference value extraction means to the image data of the first frame in accordance with the motion vector input by the data input means. Data addition means for generating image data of the second frame (e.g., the residual addition unit 215 of FIG. 17), characterized in that it comprises a.

  The image processing system according to claim 30 adds noise to the image data from the decoding device, and outputs noise added image data to the encoding device (for example, D / A in FIG. 2). A conversion unit 85) is further provided.

  The image processing method according to claim 31 is obtained by using a pixel representing a block extracted based on a frequency distribution of pixel values in a block in input image data. A data input step (for example, step S211 in FIG. 21) for inputting a motion vector and quantized data obtained by performing vector quantization on a difference value between a plurality of frames calculated using the motion vector; A difference value extraction step (for example, step S212 in FIG. 21) for extracting a difference value by inverse quantization of the quantized data input by the data input step processing, and a motion input by the data input step processing According to the vector, adding the difference value extracted by the processing of the difference value extraction step to the image data of the first frame Data adding step of generating image data of the second frame (e.g., step S214 in FIG. 21), characterized in that it comprises a.

  The image processing method according to a thirty-second aspect adds a noise to the image data from the decoding device and outputs the noise-added image data to the encoding device (for example, step S7 in FIG. 4). Or S2)).

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 2 shows a configuration example of an image processing system 51 to which the present invention is applied. The image processing system 51 again uses a playback device 61 that outputs analog image data Van1, a display 62 that displays an image corresponding to the image data Van1 output from the playback device 61, and analog image data Van1. The encoding unit 63 is configured to perform encoding processing and record encoded image data Vcd (hereinafter also referred to as encoded data Vcd) on a recording medium such as an optical disk (not shown).

  The playback device 61 includes a decoding unit 71 and a D / A (Digital-to-Analog) conversion unit 72. The decoding unit 71 decodes encoded image data reproduced from a recording medium such as an optical disc (not shown), and supplies the decoded digital image data Vdg0 to the D / A conversion unit 72. The D / A conversion unit 72 converts the digital image data Vdg0 from the decoding unit 71 into analog image data Van1, and supplies the converted analog image data Van1 to the display 62.

  The display 62 is composed of, for example, a CRT (Cathode Ray Tube) or an LCD (Liquid Crystal Display), and displays an image corresponding to the analog image data Van1 from the D / A converter 72.

  The encoding device 63 includes an A / D (Analog-to-Digital) conversion unit 81, an encoding unit 82, a recording unit 83, a decoding unit 84, a D / A conversion unit 85, and a display 86.

  The A / D converter 81 converts the analog image data Van1 from the playback device 61 into digital image data Vdg1, and supplies the digital image data Vdg1 to the encoder 82.

  The encoding unit 82 encodes the digital image data Vdg1 from the A / D conversion unit 81 and supplies the encoded data Vcd to the recording unit 83 or the decoding unit 84. In the encoding unit 82, the same encoding process as that of the encoded image data obtained by reproducing from the recording medium in the reproducing apparatus 61 is executed. That is, the encoding unit 82 performs motion search by block matching using only the pixels of the top three frequency distributions of the search source block of the digital image data Vdg1, and calculates the residual after motion estimation as a vector quantum. As a result, encoded data Vcd is obtained. Details of the configuration of the encoding unit 82 will be described later.

  The recording unit 83 records the encoded data Vcd from the encoding unit 82 on a recording medium such as an optical disc (not shown). The encoded data Vcd recorded on the recording medium by the recording unit 83 may be read by the recording unit 83 and supplied to the decoding unit 84.

  The decoding unit 84 decodes the encoded data Vcd from the encoding unit 82 or the recording unit 83, and supplies the decoded digital image data Vdg 2 to the D / A conversion unit 85. In the decoding unit 84, a decoding process similar to the decoding in the decoding unit 71 is executed. That is, the decoding unit 84 is configured by an output block obtained by adding a prediction block and a residual block obtained by performing motion compensation and inverse vector quantization using the encoded data Vcd from the encoding unit 82, respectively. Get digital image data Vdg2. Details of the configuration of the decoding unit 84 will be described later.

  The D / A conversion unit 85 converts the digital image data Vdg2 from the decoding unit 84 into analog image data Van2, and supplies the converted analog image data Van2 to the display 86. The display 86 is composed of, for example, a CRT or LCD, and displays an image corresponding to the analog image data Van2 from the D / A converter 85.

  In the image processing system 51, when the D / A conversion unit 72 or the D / A conversion unit 85 converts the data into analog data, the A / D conversion unit 81 converts the data into digital data, and the D / A conversion unit. When data is communicated through a communication path between the A / D converter 81 and the A / D converter 81, noise (such as distortion of high-frequency components) such as white noise or the phase of the image data is shifted in the converted image data. This causes distortion (hereinafter referred to as phase shift). That is, high-frequency component distortion due to white noise and distortion due to phase shift are added to the converted image data as noise. These white noise and phase shift (distortion due to phase difference) are collectively referred to as analog noise (or analog distortion).

  Here, the distortion of the high frequency component caused by white noise will be described. In the process of converting digital image data into analog image data, white noise with a substantially uniform frequency component is added to the image data. The level of white noise changes randomly in time series, and its distribution almost follows a normal distribution. That is, the level of white noise added to the analog image data corresponding to each pixel changes randomly.

  Therefore, for example, in the digital image data Vdg0 before conversion, even if the pixel values of a plurality of pixels in one horizontal line have the same value, they are D / A converted by the D / A converter 72, Further, the pixel value of the pixel having the same value in the digital image data image data Vdg1 after A / D conversion by the A / D conversion unit 81 is centered on the original value (the same value). As a result, the image data is distorted with high frequency components. Similarly, not only the horizontal direction but also the vertical direction is distorted by high frequency components. In addition, depending on the degree of dispersion of the level of white noise added to each pixel, distortion of components other than high-frequency components may occur.

  As described above, in the D / A conversion unit 72 and the D / A conversion unit 85, white noise is added in the process of converting digital image data into analog image data. In the two dimensions, data distortion occurs. Note that the noise added to the image data is not limited to white noise but also includes other colored noise.

  On the other hand, the distortion due to the phase shift adds distortion due to the phase shift to the image data Van1, and when the A / D conversion is performed on the image data Van1, the position where the image data is quantized shifts, and thus the image data Vdg1 The pixel position is, for example, shifted by φh in the horizontal direction and φv in the vertical direction compared to the pixel position of the original image data Vdg0. The horizontal phase shift width φh is smaller or larger than the horizontal pixel interval, while the vertical phase shift width φv is an integral multiple of the vertical pixel interval. Further, the phase shift may occur only in one direction in the horizontal direction or the vertical direction.

  When the phase shift occurs, the display position of the image is shifted by the range of the phase shift. However, the display position shift is slight, and the image quality of the image that can be visually recognized by the user is almost the same. Has no effect.

  As described above, the analog image data Van1 output from the D / A converter 72 and the digital image data Vdg1 from the A / D converter 81 have white noise and phase as compared with the digital image data Vdg0. The analog image data Van2 output from the D / A converter 85 is further accompanied by white noise and a phase shift as compared with the digital image data Vdg1.

  Although the degree of deterioration of image quality due to the white noise and the phase shift is not so great, the encoding unit 82 uses the digital image data Vdg1 with a phase shift, among the analog noises, to perform the encoding process. By being executed, accurate estimation of motion and vector quantization in the encoding unit 82 is suppressed.

  As a result, the image quality of the encoded data Vcd obtained from the encoding unit 82 and the analog image data Van2 obtained from the decoding unit 84 are greatly deteriorated compared to the image quality of the digital image data Vdg0 and Vdg1. Thus, it is possible to contribute to the prevention of analog copy while displaying an image with little deterioration in image quality.

  Furthermore, as described above, white noise and phase shift occur during analog and digital conversion, and therefore have little effect when copying digital data. Therefore, according to the image processing system 51, only analog copying can be restricted, and the image quality of image data can be deteriorated during analog copying.

  In the image processing system 51 of FIG. 2, white noise addition and phase shift are not caused during analog conversion in the D / A conversion unit 72 or D / A conversion unit 85 or in digital conversion in the A / D conversion unit 81. Although it occurs naturally, it is possible to forcibly apply (add) the occurrence of white noise or phase shift more than naturally occurring.

  As a result, the effect of preventing analog copy can be further improved.

  FIG. 3 shows a frame configuration of image data processed in the image processing system 51 according to the present invention.

  In the example of FIG. 3, a frame of image data is shown along the time axis. The image data is composed of reference frames (hatched in the figure) of the 0th frame and the 5th frame and frames other than the reference frame. The reference frame interval is 5 frame intervals, and can be set by the user.

  In the image processing system 51, among these frames, the reference frame uses ADRC (Adaptive Dynamic Range Coding) shown in Patent Document 2 in which the dynamic range is reduced by encoding and decoding, for example. Intra-frame encoding is performed. Then, inter-frame encoding described below is executed for frames other than the reference frame. That is, the following description is directed to interframe coding.

  Next, an example of processing in the image processing system 51 will be described with reference to the flowchart of FIG.

  In step S1, the decoding unit 71 decodes encoded image data reproduced from a recording medium such as an optical disc (not shown), and supplies the decoded digital image data Vdg0 to the D / A conversion unit 72. The process proceeds to step S2. In step S1, processing similar to the decoding processing in step S6 described later is executed.

  In step S2, the D / A conversion unit 72 converts the digital image data Vdg0 from the decoding unit 71 into analog image data Van1, and the converted analog image data Van1 is displayed on the display 62 and the A / D conversion unit. The process proceeds to step S3.

  Thus, in step S3, an image corresponding to the analog image data Van1 is displayed on the display 62.

  In step S4, the A / D conversion unit 81 converts the analog image data Van1 from the D / A conversion unit 72 into digital image data Vdg1, supplies the digital image data Vdg1, and proceeds to step S5. That is, by the conversion of the D / A conversion unit 72 in step S2 and the conversion of the A / D conversion unit 81 in step S4, the digital image data Vdg1 has white noise and phase shift compared to the digital image data Vdg0. Is added.

  In step S5, the encoding unit 82 encodes the digital image data Vdg1 from the A / D conversion unit 81, supplies the encoded data Vcd to the decoding unit 84, and proceeds to step S6. Details of the processing of the encoding unit 82 will be described later.

  With the encoding process in step S5, motion search by block matching is performed using only the pixels in the top three frequency distributions of the search source block of the digital image data Vdg1 to which white noise and phase shift are added. The residual after the motion estimation is vector quantized to obtain encoded data Vcd, and the obtained encoded data Vcd is supplied to the decoding unit 84.

  In step S6, the decoding unit 84 decodes the encoded data Vcd from the image data encoding unit 82, supplies the decoded digital image data Vdg2 to the D / A conversion unit 85, and proceeds to step S7. Details of the processing of the decoding unit 84 will be described later.

  An output block obtained by adding the prediction block and the residual block obtained by performing motion compensation and inverse vector quantization using the encoded data Vcd encoded by the encoding unit 82 by the decoding processing in step S6 is obtained. Digital image data Vdg2 that is generated and configured by the generated output block is acquired.

  In step S7, the D / A conversion unit 85 converts the digital image data Vdg2 from the decoding unit 84 into analog image data Van2, and supplies the converted analog image data Van2 to the display 86. Proceed to S8.

  As a result, in step S8, an image corresponding to the analog image data Van2 is displayed on the display 86, and the image processing is ended in the image processing system 51.

  As described above, in the image processing system 51 according to the present invention, the above-described encoding process is performed using the digital image data Vdg1 to which white noise and phase shift are added. In addition, accurate motion estimation and vector quantization are suppressed. Further, since the decoding process is executed using the encoded data Vcd obtained by encoding the digital image data Vdg1 to which the white noise and the phase shift are added, the motion compensation and the residual are particularly caused by the phase shift. Accurate compensation is suppressed.

  As a result, the encoded data Vcd obtained from the encoding unit 82 and the digital image data Vdg2 from the decoding unit 84 obtained by decoding the encoded data Vcd rather than the image displayed on the display 62 in step S4 are digital image data Vdg0. Since the image quality is greatly deteriorated as compared with the analog image data Van1, the image quality of the image displayed on the display 86 in step S8 is deteriorated, and the analog copy can be prevented.

  It should be noted that the encoded data Vcd whose image quality has greatly deteriorated is read out from the recording medium recorded by the recording unit 83, and the decoded result is also equivalent to the image displayed on the display 86 in step S8. .

  Therefore, in step S1 described above, the image data encoded by the encoding unit 32 is read from the recording medium recorded by the recording unit 83, and the encoding of steps S5 and S6 is performed again on the decoded image data. The image quality of the decoded image data is further deteriorated than that of the digital image data Vdg2. That is, every time encoding and decoding according to the present invention are repeated, the image quality of the image data obtained as a result deteriorates more and more.

  As described above, it is possible to contribute to the prevention of analog copy.

  Next, details of the configuration of the encoding unit 82 in FIG. 2 will be described.

  FIG. 5 is a block diagram showing a configuration of the encoding unit 82. The encoding unit 82 receives the digital image data Vdg1 accompanied by white noise or phase shift from the A / D conversion unit 81, encodes the input digital image data Vdg1, and the encoded data Vcd becomes the subsequent stage. To the recording unit 83 or the decoding unit 84.

  The encoding unit 82 includes a blocking unit 111, a frame memory 112, a motion estimation block generation unit 113, a motion estimation unit 114, a residual calculation unit 115, a residual vector quantization unit 116, and a data synthesis unit 117. .

  The digital image data Vdg1 from the A / D conversion unit 81 is input to the blocking unit 111 and the frame memory 112. The blocking unit 111 reads an input image, divides it into designated block sizes (for example, 4 × 4 pixels or 8 × 8 pixels), and performs motion estimation for each block using image data of the designated block size as an input block. This is supplied to the block generator 113 and the residual calculator 115.

  The frame memory 112 accumulates image data of the previous frame (hereinafter also referred to as the previous frame) and supplies the image data to the motion estimation unit 114 and the residual calculation unit 115.

  The motion estimation block generation unit 113 generates a motion estimation block for motion estimation performed by the motion estimation unit 114. That is, the motion estimation block generation unit 113 reads the input block supplied from the blocking unit 111 and generates an intra-block frequency distribution of pixels of the input block. The motion estimation block generation unit 113 generates a pixel of the motion estimation block based on the generated intra-block frequency distribution, and supplies the generated motion estimation block to the motion estimation unit 114.

  The motion estimation unit 114 reads the previous frame from the frame memory 112 and the motion estimation block from the motion estimation block generation unit 113, and uses the pixel value (other than 0) of the motion estimation block to perform the motion of the previous frame by block matching. As a result, the motion vector is calculated, and the calculated motion vector is supplied to the residual calculation unit 115 and the data synthesis unit 117.

  The residual calculation unit 115 calculates a residual after motion estimation. That is, the residual calculation unit 115 reads the input block from the blocking unit 111, the motion vector from the motion estimation unit 114, and the previous frame from the frame memory 112. The residual calculation unit 115 generates a pixel value of the prediction block using the motion vector and the previous frame, and obtains the prediction block. The residual calculation unit 115 supplies the obtained prediction block and the residual of the input block to the residual vector quantization unit 116 as a residual block.

  The residual vector quantization unit 116 reads the residual block from the residual calculation unit 115 and performs vector quantization using the LBG algorithm using the vector quantization code book 121. The residual vector quantization unit 116 supplies the quantized code data obtained by vector quantization to the data synthesis unit 117.

  The vector quantization code book 121 is generated from an input image by a code book generating unit (not shown) built in the encoding unit 82. Hereinafter, it is also simply referred to as a code book 121.

  The data synthesizer 117 reads the motion vector from the motion estimator 114, the quantized code data from the residual vector quantizer 116, and the code book 121, and reads the motion vector, quantized code data, and code book 121, The encoded data Vcd is output to the subsequent recording unit 83 or decoding unit 84.

  As described above, the motion estimation block generation unit 113 generates a motion estimation block for motion estimation performed by the motion estimation unit 114 according to the intra-block frequency distribution of the input block of the digital image data Vdg1. The pixels used by the motion estimation unit 114 are selected, but since the phase shift is added to the digital image data Vdg1 input from the A / D conversion unit 81, the pixels used by the motion estimation unit 114 are When the original image data recorded on the recording medium is encoded, it differs from the pixels used in accordance with the intra-block frequency distribution, and it is difficult to accurately perform motion estimation.

  That is, the motion vector obtained by the motion estimation unit 114 is not necessarily accurate, and the codebook generated from the residual after motion estimation obtained using the motion vector and the input digital image data Vdg1 is The quantized code data used and quantized is not always accurate.

  Therefore, the image quality of the digital image data Vdg2 obtained by decoding using the encoded data Vcd by the decoding unit 84 deteriorates. Thereby, analog copy is suppressed.

  Furthermore, the encoded data Vcd supplied to the decoding unit 84 also includes a codebook generated from the input digital image data Vdg1 and used for quantization of the residual after motion estimation. That is, in the decoding unit 84, since the encoding unit 82 uses the code book generated from the input digital image data Vdg1 to perform the residual compensation, the residual compensation is not accurately performed. The image quality of the digital image data Vdg2 is degraded. This further suppresses analog copying.

  FIG. 6 shows a configuration example of the motion estimation block generation unit 113 in FIG.

  In the example of FIG. 6, the motion estimation block generation unit 113 includes a frequency determination unit 131, a motion estimation pixel generation unit 132, and a raster scan 133.

  The frequency determination unit 131 reads the input block supplied from the blocking unit 111 and generates an intra-block frequency distribution of pixels of the input block. The frequency determination unit 131 selects a target pixel in the input block, determines whether or not the pixel value of the selected target pixel is within the top three of the intra-block frequency distribution, and performs motion estimation according to the determination result. The pixel generation unit 132 is controlled to set the pixel value of the motion estimation block. In this case, the criterion for determination is within the top three, but it may be the top two or the top four. That is, the criterion for determination is not limited to the top three.

  The motion estimation pixel generation unit 132 generates a motion estimation block by setting the pixel value of the motion estimation block under the control of the frequency determination unit 131 and supplies the generated motion estimation block to the motion estimation unit 114. To do.

  That is, when the frequency determination unit 131 determines that the pixel value of the target pixel is within the top three in the intra-block frequency distribution, the motion estimation pixel generation unit 132 calculates the pixel value of the target pixel of the input block as the motion estimation. When the pixel value of the target pixel in the block is set and the frequency determination unit 131 determines that the pixel value of the target pixel is not within the top three in the intra-block frequency distribution, the pixel value of the target pixel in the motion estimation block is 0 Set.

  The raster scan unit 133 moves the pixels of the input block in the raster scan order so that the frequency determination unit 131 selects the next target pixel in the raster scan order.

  FIG. 7 shows a configuration example of the motion estimation unit 114 of FIG.

  In the example of FIG. 7, the motion estimation unit 114 is configured by a motion search unit 141.

  The motion search unit 141 reads the previous frame from the frame memory 112 and the motion estimation block from the motion estimation block generation unit 113. The motion search unit 141 uses a pixel value other than 0 of the motion estimation block to search for a motion from the previous frame by the pixel matching square sum minimum norm by block matching, and as a result of searching for the motion, a motion vector is obtained. calculate. Then, the motion search unit 141 supplies the calculated motion vector to the residual calculation unit 115 and the data synthesis unit 117.

  FIG. 8 shows a configuration example of the residual calculation unit 115 of FIG.

  In the example of FIG. 8, the residual calculation unit 115 includes a prediction block calculation unit 151 and a residual calculation unit 152.

  The prediction block calculation unit 151 reads the motion vector from the motion estimation unit 114 and the previous frame from the frame memory 112, generates a pixel value of the prediction block using the motion vector and the previous frame, obtains a prediction block, The obtained prediction block is supplied to the residual calculation unit 152.

  The residual calculation unit 152 reads the input block from the blocking unit 111 and the prediction block from the prediction block calculation unit 151, calculates the residual between the input block and the prediction block, and uses the calculated residual as the residual block. Is supplied to the residual vector quantization unit 116.

  FIG. 9 shows a configuration example of the residual vector quantization unit 116 of FIG.

  In the example of FIG. 9, the residual vector quantization unit 116 is configured by a vector quantization unit 161.

  The vector quantization unit 161 reads the residual block from the residual calculation unit 152, uses the code book 121, performs vector quantization on the residual block by the LBG algorithm, and obtains quantized code data obtained by vector quantization. Is supplied to the data composition unit 117.

  Next, the encoding process of the encoding unit 82 in FIG. 5 will be described with reference to the flowchart in FIG. 10. This encoding process is the encoding process of step S5 in the process of the encoding device 63 described above with reference to FIG.

  Digital image data Vdg1 from the A / D conversion unit 81 is input to the blocking unit 111 and the frame memory 112 of the encoding unit 82. The image data of the previous frame input and stored in the frame memory 112 is supplied to the motion estimation unit 114 and the residual calculation unit 115.

  When the digital image data Vdg1 from the A / D conversion unit 81 is input, the blocking unit 111 executes a blocking process in step S21. This blocking process will be described later in detail with reference to FIG.

  By the blocking process in step S21, the read input image is divided into designated block sizes, and the divided image data is supplied as input blocks to the motion estimation block generation unit 113 and the residual calculation unit 115 for each block. The process proceeds to step S22.

  When the input block is input from the blocking unit 111, the motion estimation block generation unit 113 executes a motion estimation block generation process for generating a motion estimation block for motion estimation performed by the motion estimation unit 114 in step S22. . This motion estimation block generation process will be described later in detail with reference to FIG.

  By the motion estimation block generation processing in step S22, the intra-block frequency distribution of the pixels of the input block is generated, and the pixels of the motion estimation block are generated based on the generated intra-block frequency distribution. Then, the generated motion estimation block is supplied to the motion estimation unit 114, and the process proceeds to step S23.

  When the motion estimation block is supplied from the motion estimation block generation unit 113, the motion estimation unit 114 uses the motion estimation block generated by the motion estimation block generation unit 113 in step S23 to perform motion estimation processing by the block matching method. Execute. This motion estimation process will be described later in detail with reference to FIG.

  By the motion estimation process in step S23, the motion search of the previous frame from the frame memory 112 is performed using the pixel value (other than 0) of the motion estimation block, and as a result, a motion vector is calculated. Then, the calculated motion vector is supplied to the residual calculation unit 115 and the data synthesis unit 117, and the process proceeds to step S24.

  When the motion vector is supplied from the motion estimation unit 114, the residual calculation unit 115 executes a residual calculation process in step S24. This residual calculation process will be described later in detail with reference to FIG.

  By the residual calculation process in step S24, the pixel value of the prediction block is generated using the motion vector from the motion estimation unit 114 and the previous frame from the frame memory 112, and the prediction block is obtained. Then, the obtained prediction block and the residual of the input block from the blocking unit 111 are supplied as a residual block to the residual vector quantization unit 116, and the process proceeds to step S25.

  When the residual block is supplied from the residual calculation unit 115, the residual vector quantization unit 116 performs a residual quantization process in step S25. This residual quantization process will be described later in detail with reference to FIG.

  By the residual quantization processing in step S25, the residual block from the residual calculation unit 115 is vector quantized using the vector quantization code book 121, and the quantized code data obtained by vector quantization is obtained. , And the process proceeds to step S26.

  When the quantized code data is input from the residual vector quantizing unit 116, the data synthesizing unit 117 executes a data synthesizing process in step S26. This data composition processing will be described later in detail with reference to FIG.

  By the data synthesis process in step S26, the motion vector from the motion estimation unit 114, the quantized code data from the residual vector quantization unit 116, and the vector quantization code book 121 are synthesized as encoded data Vcd, and the subsequent stage. Are output to the recording unit 83 or the decoding unit 84.

  As described above, the encoding process of the encoding unit 82 is terminated, and the process returns to step S5 in FIG. 4 and proceeds to step S6, where the decoding process is executed.

  Next, the blocking process of the blocking unit 111 in FIG. 5 in step S21 in FIG. 10 will be described with reference to the flowchart in FIG.

  In step S41, the block forming unit 111 reads the digital image data Vdg1 input from the A / D conversion unit 81 as an input image, proceeds to step S42, and converts the input image into a designated block (for example, 4 × 4 pixels or 8 X8 pixels, etc.), and the process proceeds to step S43.

  In step S43, the blocking unit 111 supplies the divided image data of the designated block size as an input block for each block to the motion estimation block generation unit 113 and the residual calculation unit 115, and ends the blocking process. Returning to step S21 of FIG. 10, the process proceeds to step S22.

  Next, the motion estimation block generation process of the motion estimation block generation unit 113 in FIG. 5 in step S22 in FIG. 10 will be described with reference to the flowchart in FIG.

  In step S61, the frequency determination unit 131 reads the input block supplied from the blocking unit 111, proceeds to step S62, generates an intra-block frequency distribution of pixel values of the input block, and proceeds to step S63.

  In step S63, the frequency determination unit 131 selects a target pixel from the input block, and proceeds to step S64. Note that the target pixel is selected in the raster scan order from the upper left pixel of the input block.

  In step S64, the frequency determination unit 131 determines whether or not the pixel value of the selected target pixel is within the top three of the intra-block frequency distribution, and the pixel value of the selected target pixel is the top of the intra-block frequency distribution. When it is determined that it is within the third place, the process proceeds to step S65, the motion estimation pixel generation unit 132 is controlled, and the pixel value of the target pixel in the input block is set as the pixel value of the target pixel in the motion estimation block. Proceed to

  On the other hand, if the frequency determination unit 131 determines in step S64 that the pixel value of the selected target pixel is not within the top three of the intra-block frequency distribution, the frequency determination unit 131 proceeds to step S66 and controls the motion estimation pixel generation unit 132. , 0 is set as the pixel value of the target pixel in the motion estimation block, and the process proceeds to step S67.

  That is, in the motion estimation block, a pixel value that is within the top three in the intra-block frequency distribution is set with the corresponding pixel value in the input block, and the pixel value does not enter the top three in the intra-block frequency distribution. 0 is set to the pixel.

  In step S67, the motion estimation pixel generation unit 132 determines whether or not the processing of all the pixels of the motion estimation block has been completed (that is, all the pixel values of the motion estimation block have been set). If it is determined that the pixel processing has not ended, the raster scan unit 133 moves the pixels of the input block in the raster scan order, and the process returns to step S63.

  Thereby, in step S63, the frequency determination unit 131 selects the next pixel of interest in the input block.

  If the motion estimation pixel generation unit 132 determines in step S67 that all the pixels of the motion estimation block have been processed, all the pixel values of the motion estimation block are set and the motion estimation block is generated. Then, the generated motion estimation block is supplied to the motion estimation unit 114. Then, the motion estimation pixel generation unit 132 ends the motion estimation block generation processing, returns to step S22 in FIG. 10, and proceeds to step S23.

  Next, the motion estimation process of the motion estimation unit 114 in FIG. 5 in step S23 in FIG. 10 will be described with reference to the flowchart in FIG.

  In step S81, the motion search unit 141 reads the motion estimation block from the motion estimation block generation unit 113, proceeds to step S82, reads the previous frame from the frame memory 112, and proceeds to step S83.

  In step S83, the motion search unit 141 uses the pixel value other than 0 of the motion estimation block to search for motion from the previous frame using the pixel difference square sum minimum norm, and proceeds to step S84 to search for motion. As a result, a motion vector is calculated, the calculated motion vector is supplied to the residual calculation unit 115 and the data synthesis unit 117, the motion estimation process is terminated, the process returns to step S23 in FIG. 10, and the process proceeds to step S24.

  Next, the residual calculation process of the residual calculation unit 115 in FIG. 5 in step S24 in FIG. 10 will be described with reference to the flowchart in FIG.

  In step S101, the residual calculation unit 152 reads the input block supplied from the blocking unit 111, and proceeds to step S102.

  In step S102, the prediction block calculation unit 151 reads the motion vector supplied from the motion estimation unit 114, proceeds to step S103, reads the previous frame supplied from the frame memory 112, and proceeds to step S104.

  In step S104, the prediction block calculation unit 151 generates a pixel value of the prediction block using the motion vector from the motion estimation unit 114 and the previous frame from the frame memory 112, obtains a prediction block, and obtains the prediction block Is supplied to the residual calculation unit 152, and the process proceeds to step S105.

  The residual calculation unit 152 calculates the residual of the input block from the blocking unit 111 and the prediction block from the prediction block calculation unit 151, and uses the calculated residual as a residual block as a residual vector quantization unit. 116 is supplied to the vector quantization unit 161, the residual calculation process is terminated, the process returns to step S24 in FIG. 10, and the process proceeds to step S25.

  Next, the residual quantization processing of the residual vector quantization unit 116 in FIG. 5 in step S25 in FIG. 10 will be described with reference to the flowchart in FIG.

  In step S121, the vector quantization unit 161 reads the residual block from the residual calculation unit 152, proceeds to step S122, reads the code book 121 from the code book generation unit (not shown), and proceeds to step S123.

  The code book 121 is generated by using an input image by a code book generation unit (not shown) built in the encoding unit 82.

  In step S123, the vector quantization unit 161 acquires the quantized code data by vector quantization using the LBG algorithm using the code book 121, and supplies the acquired quantized code data to the data synthesis unit 117. Then, the residual quantization process is terminated, the process returns to step S25 in FIG. 10, and the process proceeds to step S26.

  Next, the data composition processing of the data composition unit 117 in FIG. 5 in step S26 in FIG. 10 will be described with reference to the flowchart in FIG.

  In step S141, the data synthesis unit 117 reads the quantization code data supplied from the vector quantization unit 161, proceeds to step S142, reads the motion vector supplied from the motion estimation unit 114, and proceeds to step S143.

  In step S143, the data synthesis unit 117 reads the code book 121 from a code book generation unit (not shown), proceeds to step S144, synthesizes the quantized code data, the motion vector, and the code book 121, and generates the encoded data Vcd. To the recording unit 83 or the decoding unit 84 in the subsequent stage.

  Then, the data synthesis unit 117 ends the data synthesis process, returns to step S26 in FIG. 10, ends the encoding process in FIG. 10, returns to step S5 in FIG. 4, and proceeds to step S6.

  As described above, the encoding unit 82 calculates the motion vector searched using only the pixels that are within the top three in the intra-block frequency distribution and the prediction block generated based on the motion vector. The quantized code data obtained by quantizing the residual block and the code book 121 generated from the input image and used for the quantization processing are supplied to the subsequent stage as encoded data Vcd.

  That is, this motion vector is obtained by the motion estimation block 114 using the motion estimation block generated by the motion estimation block generation unit 113 according to the intra-block frequency distribution of the input block of the digital image data Vdg1. The digital image data Vdg1 input from the A / D conversion unit 81 is added with white noise and phase shift, and the pixels used by the motion estimation unit 114 and the recording medium are added due to the phase shift. When the original image data recorded in is encoded, the pixels used in accordance with the intra-block frequency distribution are different, which is not necessarily accurate.

  Therefore, the quantized code data obtained by quantizing the residual after motion estimation obtained using this motion vector using the code book generated from the input digital image data Vdg1 is not necessarily accurate.

  As a result, the image quality of the digital image data Vdg2 obtained by decoding using the encoded data Vcd by the decoding unit 84 deteriorates.

  Furthermore, the encoded data Vcd supplied to the decoding unit 84 also includes a codebook generated from the input digital image data Vdg1 and used for quantization of the residual after motion estimation. That is, since the decoding unit 84 uses this codebook to perform residual compensation, the residual compensation is not accurately performed, and the image quality of the digital image data Vdg2 deteriorates.

  As described above, the analog copy is suppressed by the encoding by the encoding unit 82.

  Next, details of the configuration of the decoding unit 84 of FIG. 2 will be described.

  FIG. 17 is a block diagram illustrating a configuration of the decoding unit 84. The decoding unit 84 receives the encoded data Vcd from the encoding unit 82 or the recording unit 83, decodes the encoded data Vcd, and supplies it as digital image data Vdg2 to the D / A conversion unit 85 at the subsequent stage. .

  The decoding unit 84 includes a data decomposition unit 211, a residual inverse vector quantization unit 212, a frame memory 213, a motion compensation unit 214, a residual addition unit 215, and a data combining unit 216.

  The data decomposing unit 211 receives the encoded data Vcd from the encoding unit 82 (or the recording unit 83), decomposes the motion vector, the quantized code data, and the code book 121 from the encoded data Vcd, and moves the motion vector. Is supplied to the motion compensation unit 214, and the quantized code data and the code book 121 are supplied to the residual inverse vector quantization unit 212.

  The residual inverse vector quantization unit 212 reads the quantized code data and the code book 121 from the data decomposition unit 211 and performs residual compensation. That is, the residual inverse vector quantization unit 212 performs inverse quantization using the quantized code data and the code book 121, obtains a residual block from the value obtained by the inverse quantization, and obtains the obtained residual block. Is supplied to the residual adder 215.

  Digital image data Vdg2 from the data combination unit 216 is accumulated in the frame memory 213, and the frame memory 213 supplies the image data of the previous frame to the motion compensation unit 214.

  The motion compensation unit 214 obtains a motion estimation destination block from the previous frame read from the frame memory 213 based on the motion vector from the data decomposition unit 211, obtains a prediction block from the obtained block, and obtains the obtained prediction The block is supplied to the residual adder 215.

  The residual addition unit 215 adds the residual block obtained by the residual inverse vector quantization unit 212 to the prediction block obtained by the motion compensation unit 214, obtains an output block, and obtains the obtained output block as data Supply to the coupling unit 216.

  The data combination unit 216 has an output image area in a built-in memory (not shown), writes the image data of the output block from the residual addition unit 215 to the output image area, and all the output blocks are written. At this time, the image data written in the output image area is supplied as digital image data Vdg2 to the D / A conversion unit 85 in the subsequent stage and written into the frame memory 213.

  As described above, in the decoding unit 84 in FIG. 17, the prediction block obtained by the motion compensation unit 214 is acquired based on the motion vector obtained by the encoding unit 82 which is not very certain. Further, the residual block obtained by the residual inverse vector quantization unit 212 is obtained by using the code book 121 generated from the input image by the encoding unit 82.

  That is, neither the motion compensation performed by the motion compensation unit 214 nor the residual compensation performed by the residual inverse vector quantization unit 212 is necessarily accurate. Therefore, the image quality of the digital image data Vdg2 including the output block generated by adding the prediction block and the residual block deteriorates. Thereby, analog copy is suppressed.

  FIG. 18 shows a configuration example of the residual inverse vector quantization unit 212 in FIG.

  In the example of FIG. 18, the residual inverse vector quantization unit 212 is configured by an inverse vector quantization unit 221.

  The inverse vector quantization unit 221 reads the quantized code data and the code book 121 from the data decomposing unit 211, and performs inverse quantization using the quantized code data and the code book 121. The inverse vector quantization unit 221 obtains a residual block from the value obtained by inverse quantization, and supplies the obtained residual block to the residual addition unit 215.

  The code book 121 is a code book generated by using the input image in the encoding unit 82. That is, in the inverse vector quantization unit 221, the residual quantization inverse quantization is executed by the code book using the image before motion estimation.

  FIG. 19 shows a configuration example of the motion compensation unit 214 of FIG.

  In the example of FIG. 19, the motion compensation unit 214 includes a motion compensation processing unit 231 and a prediction block acquisition unit 232.

  The motion compensation processing unit 231 reads the motion vector supplied from the data decomposition unit 211 and reads the previous frame from the frame memory 213. Then, the motion compensation processing unit 231 obtains a motion estimation destination block from the previous frame from the frame memory 213 based on the motion vector from the data decomposing unit 211.

  The prediction block acquisition unit 232 acquires a prediction block from the motion estimation destination block obtained by the motion compensation processing unit 231, and supplies the acquired prediction block to the residual addition unit 215.

  FIG. 20 shows a configuration example of the residual adder 215 of FIG.

  In the example of FIG. 20, the residual adding unit 215 includes a residual calculating unit 241.

  The residual calculation unit 241 reads the prediction block supplied from the motion compensation unit 214 and reads the residual block supplied from the residual inverse vector quantization unit 212. The residual calculation unit 241 obtains an output block by adding the residual block from the residual inverse vector quantization unit 212 to the prediction block from the motion compensation unit 214, and obtains the obtained output block as a data combining unit. 216.

  Next, the decoding process of the decoding unit 84 of FIG. 17 will be described with reference to the flowchart of FIG. This encoding process is the decoding process in step S6 in the process of the encoding device 63 described above with reference to FIG.

  The encoded data Vcd is supplied from the encoding unit 82 (or recording unit 83) to the data decomposition unit 211 of the decoding unit 84. When the encoded data Vcd is supplied, the data decomposing unit 211 performs a data decomposing process in step S211. This data decomposition process will be described later in detail with reference to FIG.

  The encoded data Vcd from the encoding unit 82 is decomposed by the data decomposition processing in step S211, the decomposed motion vector is supplied to the motion compensation unit 214, and the quantized code data and the code book 121 are converted into the inverse of the residual. The data is supplied to the vector quantization unit 212, and the process proceeds to step S212.

  When the quantized code data and the code book 121 are supplied from the data decomposition unit 211, the residual inverse vector quantization unit 212 performs a residual inverse quantization process in step S212. This residual inverse quantization process will be described later in detail with reference to FIG.

  By the residual inverse quantization process in step S212, vector inverse quantization is performed using the quantized code data and the code book 121, and a residual block is obtained from the value obtained by the vector inverse quantization. The data is supplied to the adding unit 215, and the process proceeds to step S213.

  When the motion vector is supplied from the data decomposing unit 211, the motion compensating unit 214 performs a motion compensation process in step S213. This motion compensation process will be described later in detail with reference to FIG.

  Based on the motion vector from the data decomposing unit 211, a motion estimation destination block is obtained from the previous frame read from the frame memory 213 by the motion compensation processing in step S213, and a predicted block is obtained from the obtained block. The obtained prediction block is supplied to the residual adding unit 215, and the process proceeds to step S214.

  When the prediction block is supplied from the motion compensation unit 214, the residual addition unit 215 performs residual addition processing in step S214. This residual addition process will be described later in detail with reference to FIG.

  By the residual addition processing in step S214, the residual block from the residual inverse vector quantization unit 212 is added to the prediction block from the motion compensation unit 214, and is supplied to the data combining unit 216 as an output block. Proceed to step S215.

  When the output block is supplied from the residual adding unit 215, the data combining unit 216 executes data combining processing in step S215. This data combination process will be described later in detail with reference to FIG.

  The image data of the output block from the residual adding unit 215 is written in the output image area by the data combination processing in step S215, and the image data written in the output image area when all the output blocks are written. Is supplied as digital image data Vdg2 to the D / A conversion unit 85 in the subsequent stage, the decoding process ends, the process returns to step S6 in FIG. 4, and proceeds to step S7.

  Next, the data decomposition process of the data decomposition unit 211 in FIG. 17 in step S211 in FIG. 21 will be described with reference to the flowchart in FIG.

  In step S231, the data decomposing unit 211 receives the encoded data Vcd supplied from the encoding unit 82, proceeds to step S232, and decomposes the input encoded data Vcd.

  That is, in step S232, the data decomposing unit 211 decomposes the motion vector, the quantized code data, and the code book 121 from the encoded data Vcd, and proceeds to step S233.

  In step S233, the data decomposition unit 211 supplies the decomposed motion vector to the motion compensation unit 214, supplies the decomposed quantized code data and the code book 121 to the residual inverse vector quantization unit 212, and performs data decomposition. The process ends, the process returns to step S211 in FIG. 21, and the process proceeds to step S212.

  Next, with reference to the flowchart of FIG. 23, the residual inverse quantization process of the residual inverse vector quantization unit 212 of FIG. 17 in step S212 of FIG. 21 will be described.

  In step S251, the inverse vector quantization unit 221 reads the quantized code data supplied from the data decomposition unit 211, proceeds to step S252, reads the code book 121 supplied from the data decomposition unit 211, and proceeds to step S253. .

  In step S253, the inverse vector quantization unit 221 performs inverse quantization using the quantized code data and the code book 121, proceeds to step S254, obtains a residual block from the value obtained by the inverse quantization, The obtained residual block is supplied to the residual adding unit 215, the residual inverse quantization process is terminated, the process returns to step S212 in FIG. 21, and the process proceeds to step S213.

  Next, the motion compensation processing of the motion compensation unit 214 in FIG. 17 in step S213 in FIG. 21 will be described with reference to the flowchart in FIG.

  In step S271, the motion compensation processing unit 231 reads the motion vector supplied from the data decomposition unit 211, proceeds to step S272, reads the previous frame from the frame memory 213, and proceeds to step S273.

  In step S273, the motion compensation processing unit 231 obtains a motion estimation destination block from the previous frame from the frame memory 213 based on the motion vector from the data decomposition unit 211, and proceeds to step S274.

  In step S274, the prediction block acquisition unit 232 acquires a prediction block from the motion estimation destination block obtained by the motion compensation processing unit 231, supplies the acquired prediction block to the residual calculation unit 241, and performs motion compensation processing. Is finished, the process returns to step S213 in FIG. 21, and proceeds to step S214.

  Next, the residual addition process of the residual addition unit 215 in FIG. 17 in step S214 in FIG. 21 will be described with reference to the flowchart in FIG.

  In step S291, the residual calculation unit 241 reads the residual block supplied from the residual inverse vector quantization unit 212, proceeds to step S292, reads the prediction block supplied from the motion compensation unit 214, and proceeds to step S293. move on.

  In step S293, the residual calculation unit 241 obtains an output block by adding the residual block from the residual inverse vector quantization unit 212 to the prediction block from the motion compensation unit 214, and obtains the obtained output block. The data is supplied to the data combination unit 216, the residual addition process is terminated, the process returns to step S214 in FIG. 21, and the process proceeds to step S215.

  Next, the data combining process of the data combining unit 216 in FIG. 17 in step S215 in FIG. 21 will be described with reference to the flowchart in FIG.

  In step S311, the data combination unit 216 inputs all output blocks supplied from the residual addition unit 215 (that is, all blocks corresponding to the input image input by the blocking unit 111 of the encoding unit 82). The process proceeds to step S312.

  In step S312, the data combination unit 216 writes the image data of the output block to the output image area, proceeds to step S313, determines whether writing of all the output blocks is completed, and writes all of the output blocks. If it is determined that the process has not ended, the process returns to step S312 to repeat the subsequent processes.

  If the data combination unit 216 determines in step S313 that writing of all output blocks has been completed, the process proceeds to step S314, in which the image data written in the output image area is converted into digital image data Vdg2, and the subsequent D The data is supplied to the / A converter 85 and written to the frame memory 213, and the process proceeds to step S215 in FIG. 21, the decoding process in FIG. 21 is terminated, the process returns to step S6 in FIG. 4, and the process proceeds to step S7.

  As described above, the decoding unit 84 uses the motion vector searched using only the pixels that are within the top three in the intra-block frequency distribution of the image data to which the phase shift is added by the encoding unit 82. Since the motion compensation is performed, the image quality of the image data generated by using the prediction block obtained by the motion compensation is deteriorated.

  Further, in the decoding unit 84, the code generated from the quantized code data obtained by quantizing the residual after motion estimation obtained by using the motion vector by the encoding unit 82 and the input digital image data Vdg1 Since the residual compensation is executed using the book, the image quality of the image data generated by using the residual block obtained by the residual compensation is deteriorated.

  Therefore, analog copying can be suppressed.

  As described above, in the image processing system 51 according to the present invention, the encoding process is executed using the digital image data Vdg1 to which white noise and phase shift are added. Accurate motion estimation and vector quantization at 82 are suppressed.

  In the image processing system 51 according to the present invention, since the decoding process is executed using the encoded data Vcd obtained by encoding the digital image data Vdg1 to which white noise and phase shift are added, motion compensation is performed. And the residual compensation is suppressed from being performed accurately.

  As described above, the encoded data Vcd obtained from the encoding unit 82 and the digital image data Vdg2 from the decoding unit 84 obtained by decoding the encoded data Vcd greatly deteriorate in image quality compared with the digital image data Vdg0 and the analog image data Van1. End up. This can contribute to prevention of analog copy.

  In the above description, the decoding unit 84 of the encoding device 63 has been described. However, the decoding unit 71 of the reproduction device 61 has the same configuration, and the same processing is executed. Therefore, the encoding and decoding according to the present invention may be repeatedly performed. In this case, the image quality of the resulting image data is further deteriorated every time it is repeated. It is effective in contributing to prevention.

  Further, in the present embodiment, the block for performing each process has been described as being configured by, for example, 8 pixels × 8 pixels, 4 pixels × 4 pixels, and the like. However, these are examples, and each process is performed. The number of pixels constituting the block to be performed is not limited to the number of pixels.

  The series of processes described above can be executed by hardware, but can also be executed by software. In this case, the playback device 61 and the encoding device 63 of FIG. 2 are configured by a personal computer 301 as shown in FIG. 27, for example.

  As illustrated in FIG. 27, the CPU 311 executes various processes according to a program recorded in a ROM (Read Only Memory) 312 or a program loaded from a storage unit 318 to a RAM (Random Access Memory) 313. The RAM 313 also appropriately stores data necessary for the CPU 311 to execute various processes.

  The CPU 311, the ROM 312, and the RAM 313 are connected to each other via the bus 314. An input / output interface 315 is also connected to the bus 314.

  The input / output interface 315 includes an input unit 316 including a keyboard and a mouse, a display such as a CRT and an LCD (for example, the display 62 and the display 86 in FIG. 2), an output unit 317 including a speaker, a hard disk, and the like. A communication unit 319 including a storage unit 318, a modem, a terminal adapter, and the like is connected. The communication unit 319 performs communication processing with other information processing apparatuses via a network (not shown) including the Internet.

  A drive 320 is connected to the input / output interface 315 as necessary, and a removable recording medium including a magnetic disk 321, an optical disk 322, a magneto-optical disk 323, a semiconductor memory 324, or the like is appropriately mounted and read from them. The computer program thus installed is installed in the storage unit 318 or the like as necessary.

  That is, the drive 320 corresponds to the recording unit 83 in FIG.

  When a series of processing is executed by software, a program constituting the software may execute various functions by installing a computer incorporated in dedicated hardware or various programs. For example, it is installed from a network or a recording medium into a general-purpose personal computer or the like.

  For example, software having functions such as the decoding unit 71 and the D / A conversion unit 72, the A / D conversion unit 81, the encoding unit 82, the decoding unit 84, and the D / A conversion unit 85 in FIG. The program to be configured is installed. Note that the form of the program is not particularly limited as long as it can execute the series of processes described above as a whole. For example, the module configuration may include a module corresponding to each of the blocks described above, a module in which some or all of the functions of several blocks are combined, or the functions of the blocks are divided. It may be a module configuration made up of modules. Or the program which has only one algorithm may be sufficient.

  As shown in FIG. 27, the recording medium including such a program is distributed to provide a program to the user separately from the apparatus main body, and includes a magnetic disk 321 (including a floppy disk) on which the program is recorded. ), Optical disk 322 (including compact disk-read only memory (CD-ROM), DVD (digital versatile disk)), magneto-optical disk 323 (including MD (mini-disk)), or semiconductor memory 324, etc. In addition to being configured by a recording medium (package medium), it is configured by a ROM 312 on which a program is recorded, a storage unit 318, and the like provided to the user in a state of being incorporated in the apparatus main body in advance.

  Here, in this specification, the processing steps for describing a program for causing a computer to perform various types of processing do not necessarily have to be processed in time series according to the order described in the flowchart, but in parallel or individually. This includes processing to be executed (for example, parallel processing or processing by an object).

  Further, the program may be processed by one computer or may be distributedly processed by a plurality of computers. Furthermore, the program may be transferred to a remote computer and executed.

  In the present specification, the term “system” represents the entire apparatus constituted by a plurality of apparatuses.

It is a block diagram which shows the structural example of the conventional image processing system. It is a block diagram which shows the structural example of the image processing system to which this invention is applied. It is a figure which shows the structure of the flame | frame of image data. It is a flowchart explaining the process of the image processing system of FIG. FIG. 3 is a block diagram illustrating a configuration example of an encoding unit of the encoding device in FIG. 2. It is a block diagram which shows the structural example of the motion estimation block production | generation part of FIG. It is a block diagram which shows the structural example of the motion estimation part of FIG. It is a block diagram which shows the structural example of the residual calculation part of FIG. It is a block diagram which shows the structural example of the residual vector quantization part of FIG. 5 is a flowchart for describing an encoding process of an encoding unit in FIG. 2 in step S5 in FIG. It is a flowchart explaining the blocking process of step S21 of FIG. It is a flowchart explaining the motion estimation block production | generation process of step S22 of FIG. It is a flowchart explaining the motion estimation process of step S23 of FIG. It is a flowchart explaining the residual calculation process of step S24 of FIG. It is a flowchart explaining the residual quantization process of step S25 of FIG. It is a flowchart explaining the data composition process of step S26 of FIG. FIG. 3 is a block diagram illustrating a configuration example of a decoding unit of the encoding device in FIG. 2. It is a block diagram which shows the structural example of the residual inverse vector quantization part of FIG. It is a block diagram which shows the structural example of the motion compensation part of FIG. It is a block diagram which shows the structural example of the residual addition part of FIG. 6 is a flowchart for explaining the decoding process of the decoding unit of FIG. 2 in step S6 of FIG. It is a flowchart explaining the data decomposition process of step S211 of FIG. It is a flowchart explaining the residual dequantization process of step S212 of FIG. It is a flowchart explaining the motion compensation process of step S213 of FIG. It is a flowchart explaining the residual addition process of step S214 of FIG. It is a flowchart explaining the data combination process of step S215 of FIG. It is a block diagram which shows the structural example of the personal computer to which this invention is applied.

Explanation of symbols

  51 image processing system, 61 playback device, 62 display, 63 encoding device, 81 A / D conversion unit, 82 encoding unit, 83 recording unit, 84 decoding unit, 85 D / A conversion unit, 86 display, 111 block Unit, 112 frame memory, 113 motion estimation block generation unit, 114 motion estimation unit, 115 residual calculation unit, 116 residual vector quantization unit, 117 data synthesis unit, 121 vector quantization codebook, 211 data decomposition unit, 212 Residual inverse vector quantization unit, 213 frame memory, 214 motion compensation unit, 215 residual addition unit, 216 data combination unit

Claims (32)

In an encoding device for encoding image data,
Blocking means for blocking the first frame in the input image data;
Pixel extraction means for extracting pixels representing the block based on a frequency distribution of pixel values of the blocks blocked by the blocking means;
Motion estimation means for estimating a motion vector of the block using the pixels extracted by the pixel extraction means;
Using the motion vector estimated by the motion estimation means, a difference calculation means for calculating a difference between the image data of the second frame corresponding to the block and the image data in the block;
An encoding apparatus comprising: vector quantization means for performing vector quantization on the difference calculated by the difference calculation means. Noise addition means for adding noise to the image data and outputting the image data to which the noise has been added;
The encoding apparatus according to claim 1, wherein the blocking unit blocks the first frame of the image data to which the noise is added by the noise adding unit. The motion estimation means estimates the motion vector by searching the second frame for a block having a minimum sum of squares of pixel differences using the pixels extracted by the pixel extraction means. The encoding device according to claim 1. The encoding apparatus according to claim 1, wherein a code book used for the vector quantization by the vector quantization means is generated based on the input image data. The codebook used for the vector quantization by the vector quantization means, the quantized data obtained by performing the vector quantization by the vector quantization means, and the motion vector estimated by the motion estimation means The encoding apparatus according to claim 4, further comprising data output means for outputting the data as encoded data to a subsequent stage. The encoding apparatus according to claim 1, wherein the pixel extracting unit extracts pixels having a frequency distribution of pixel values of the block within the top three as pixels representing the block. In an encoding method of an encoding device for encoding image data,
A blocking step of blocking the first frame in the input image data;
A pixel extraction step of extracting a pixel representing the block based on a frequency distribution of pixel values of the block blocked by the block step;
A motion estimation step of estimating a motion vector of the block using the pixels extracted by the processing of the pixel extraction step;
A difference calculating step of calculating a difference between the image data of the second frame corresponding to the block and the image data in the block using the motion vector estimated by the processing of the motion estimating step;
And a vector quantization step of performing vector quantization on the difference calculated by the difference calculating step. A noise adding step of adding noise to the image data and outputting the image data with the noise added;
8. The encoding method according to claim 7, wherein, in the block forming step, the first frame of the image data to which the noise is added by the noise adding step is blocked. In the process of the motion estimation step, the motion vector is searched by searching the second frame for a block having a minimum sum of squares of pixel differences using the pixels extracted by the process of the pixel extraction step. The encoding method according to claim 7, wherein the encoding method is estimated. The encoding method according to claim 7, wherein a codebook used for the vector quantization by the processing of the vector quantization step is generated based on the input image data. By the codebook used for the vector quantization by the processing of the vector quantization step, the quantized data obtained by performing the vector quantization by the processing of the vector quantization step, and the processing of the motion estimation step The encoding method according to claim 10, further comprising a data output step of outputting the estimated motion vector as encoded data to a subsequent stage. 8. The encoding method according to claim 7, wherein in the processing of the pixel extraction step, pixels having a frequency distribution of pixel values of the block that are within the top three are extracted as pixels representing the block. A recording medium on which a program for causing a computer to perform processing for encoding image data is recorded,
A blocking step of blocking the first frame in the input image data;
A pixel extraction step of extracting a pixel representing the block based on a frequency distribution of pixel values of the block blocked by the block step;
A motion estimation step of estimating a motion vector of the block using the pixels extracted by the processing of the pixel extraction step;
A difference calculating step of calculating a difference between the image data of the second frame corresponding to the block and the image data in the block using the motion vector estimated by the processing of the motion estimating step;
A recording medium on which a program is recorded, comprising: a vector quantization step for performing vector quantization on the difference calculated by the processing of the difference calculation step. A program for causing a computer to perform processing for encoding image data,
A blocking step of blocking the first frame in the input image data;
A pixel extraction step of extracting a pixel representing the block based on a frequency distribution of pixel values of the block blocked by the block step;
A motion estimation step of estimating a motion vector of the block using the pixels extracted by the processing of the pixel extraction step;
A difference calculating step of calculating a difference between the image data of the second frame corresponding to the block and the image data in the block using the motion vector estimated by the processing of the motion estimating step;
And a vector quantization step for performing vector quantization on the difference calculated by the difference calculation step. In a decoding device for decoding image data,
A motion vector obtained using a pixel representing the block extracted based on a frequency distribution of pixel values in a block in image data, and a difference value between a plurality of frames calculated using the motion vector. Data input means for inputting quantized data obtained by performing vector quantization on the data;
Difference value extraction means for extracting the difference value by inverse quantization of the quantized data input by the data input means;
By adding the difference value extracted by the difference value extracting means to the image data of the first frame in accordance with the motion vector input by the data input means, the image data of the second frame is obtained. And a data adding means for generating the decoding device. The decoding apparatus according to claim 15, further comprising noise adding means for adding noise to the image data generated by the data adding means and outputting the image data to which the noise is added to a subsequent stage. The data input means also inputs a codebook used for the vector quantization,
The decoding apparatus according to claim 15, wherein the difference value extraction unit performs inverse quantization on the quantized data using the codebook input by the data input unit. Further comprising motion compensation means for predicting image data of a prediction block from the image data of the first frame in accordance with the motion vector input by the data input means;
The data adding unit generates the image data of the second frame by adding the difference value extracted by the difference value extracting unit to the image data of the prediction block predicted by the motion compensation unit. The decoding device according to claim 15. In a decoding method of a decoding device for decoding image data,
A motion vector obtained using a pixel representing the block extracted based on a frequency distribution of pixel values in a block in image data, and a difference value between a plurality of frames calculated using the motion vector. A data input step for inputting quantized data obtained by performing vector quantization on the data;
A difference value extracting step of extracting the difference value by dequantizing the quantized data input by the data input step; and
By adding the difference value extracted by the difference value extraction step to the image data of the first frame in accordance with the motion vector input by the data input step process, the second frame And a data adding step for generating the image data. The decoding according to claim 19, further comprising a noise adding step of adding noise to the image data generated by the processing of the data adding step and outputting the image data to which the noise is added to a subsequent stage. Method. In the data input step, the codebook used for the vector quantization is also input,
The decoding method according to claim 19, wherein in the process of the difference value extraction step, the quantized data is inversely quantized using the codebook input by the process of the data input step. A motion compensation step of predicting image data of a prediction block from image data of the first frame according to the motion vector input by the processing of the data input step;
In the process of the data addition step, the second frame is added by adding the difference value extracted by the process of the difference value extraction step to the image data of the prediction block predicted by the process of the motion compensation step. The decoding device according to claim 19, wherein the image data is generated. A recording medium on which a program for causing a computer to perform processing for decoding image data is recorded,
A motion vector obtained using a pixel representing the block extracted based on a frequency distribution of pixel values in a block in image data, and a difference value between a plurality of frames calculated using the motion vector. A data input step for inputting quantized data obtained by performing vector quantization on the data;
A difference value extracting step of extracting the difference value by dequantizing the quantized data input by the data input step; and
By adding the difference value extracted by the difference value extraction step to the image data of the first frame in accordance with the motion vector input by the data input step process, the second frame And a data adding step for generating the image data. A recording medium on which a program is recorded. A program for causing a computer to perform a process of decoding image data,
A motion vector obtained using a pixel representing the block extracted based on a frequency distribution of pixel values in a block in image data, and a difference value between a plurality of frames calculated using the motion vector. A data input step for inputting quantized data obtained by performing vector quantization on the data;
A difference value extracting step of extracting the difference value by dequantizing the quantized data input by the data input step; and
By adding the difference value extracted by the difference value extraction step to the image data of the first frame in accordance with the motion vector input by the data input step process, the second frame And a data adding step for generating the image data. In an image processing system comprising an encoding device and a decoding device for encoding and decoding image data,
The encoding device includes:
Blocking means for blocking the first frame in the input image data;
Pixel extraction means for extracting pixels representing the block based on a frequency distribution of pixel values of the blocks blocked by the blocking means;
Motion estimation means for estimating a motion vector of the block using the pixels extracted by the pixel extraction means;
Using the motion vector estimated by the motion estimation means, a difference calculation means for calculating a difference between the image data of the second frame corresponding to the block and the image data in the block;
An image processing system comprising: vector quantization means for performing vector quantization on the difference calculated by the difference calculation means. The image according to claim 25, further comprising noise adding means for adding noise to the image data from the decoding device and inputting the image data to which the noise has been added to the encoding device. Processing system. In an image processing method of an image processing system that includes an encoding device and a decoding device and performs encoding and decoding on image data,
The encoding method of the encoding device is:
A blocking step of blocking the first frame in the input image data;
A pixel extraction step of extracting a pixel representing the block based on a frequency distribution of pixel values of the block blocked by the block step;
A motion estimation step of estimating a motion vector of the block using the pixels extracted by the processing of the pixel extraction step;
A difference calculating step of calculating a difference between the image data of the second frame corresponding to the block and the image data in the block using the motion vector estimated by the processing of the motion estimating step;
An image processing method comprising: a vector quantization step of performing vector quantization on the difference calculated by the difference calculation step. 28. The image according to claim 27, further comprising a noise addition step of adding noise to the image data from the decoding device and inputting the image data to which the noise has been added to the encoding device. Processing method. In an image processing system comprising an encoding device and a decoding device for encoding and decoding image data,
The decoding device
A motion vector obtained using a pixel representing the block extracted based on a frequency distribution of pixel values in a block in image data, and a difference value between a plurality of frames calculated using the motion vector. Data input means for inputting quantized data obtained by performing vector quantization on the data;
Difference value extraction means for extracting the difference value by inverse quantization of the quantized data input by the data input means;
In accordance with the motion vector input by the data input unit, the image data of the second frame is generated by adding the difference value extracted by the difference value extraction unit to the image data of the first frame. And an image processing system. 30. The image according to claim 29, further comprising noise adding means for adding noise to the image data from the decoding device and outputting the image data to which the noise has been added to the encoding device. Processing system. In an image processing method of an image processing system that includes an encoding device and a decoding device and performs encoding and decoding on image data,
The decoding method of the decoding device is:
A motion vector obtained using a pixel representing the block extracted based on a frequency distribution of pixel values in a block in image data, and a difference value between a plurality of frames calculated using the motion vector. A data input step for inputting quantized data obtained by performing vector quantization on the data;
A difference value extracting step of extracting the difference value by dequantizing the quantized data input by the data input step; and
In accordance with the motion vector input by the data input step, the difference value extracted by the difference value extraction step is added to the image data of the first frame, so that the second frame An image processing method comprising: a data addition step for generating image data. 32. The image according to claim 31, further comprising a noise adding step of adding noise to the image data from the decoding device and outputting the image data to which the noise has been added to the encoding device. Processing method.

Images