JP2014204347A - Image encoding method, image decoding method, image encoding device, image decoding device, program therefor and recording medium with program recorded thereon - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image encoding method and an image decoding method capable of improving encoding efficiency and accelerating encoding processing and decoding processing by adaptively applying color space conversion while utilizing a parameter being encoded or a parameter obtained in the color space conversion.SOLUTION: A color space conversion matrix for performing color space conversion is calculated from a decoded reference signal and in accordance with a result of determining whether to execute color space conversion on the basis of a magnitude relation of a predetermined threshold value and an intrinsic value which can be obtained in the matrix calculation, image encoding is implemented by executing the color space conversion based on a calculation result of the color space conversion matrix. An inverse color space conversion matrix for implementing inverse color space conversion is calculated from a decoded reference signal and in accordance with a result of determining whether to execute the inverse color space conversion on the basis of a magnitude relation of a predetermined threshold value and an intrinsic value which can be obtained in the matrix calculation, image decoding is implemented by executing the inverse color space conversion based on a calculation result of the inverse color space conversion matrix.

Description

  The present invention relates to encoding and decoding of an image signal, and more particularly to a technique for encoding or decoding an image signal using color space conversion.

  International standard compression method for moving images (also referred to as images or simply images). In H.264 / AVC and the next generation high efficiency video coding HEVC (High Efficiency Video Coding) currently being developed, two types of processing, prediction and conversion, are performed in order to efficiently compress a large amount of moving picture information. . The prediction is divided into two types: inter coding (also referred to as inter-frame prediction coding and inter-screen coding) and intra coding (also referred to as intra-frame prediction coding and intra-screen coding). Inter-frame predictive coding is a method for compressing information by using temporal correlation in a moving image. A typical example is inter-frame prediction using motion compensation.

  On the other hand, intra-frame predictive encoding is, for example, H.264. H.264 / AVC is a technique for generating a prediction signal using decoded pixels around a block to be encoded, that is, a technique for compressing information using correlation within a frame. Regarding the conversion, a method called Discrete Cosine Transform (DCT) is adopted in JPEG, which is an international standard compression system for still images, and MPEG-2, which is an international standard compression system for moving images. JPEG2000, which is a successor to JPEG, employs a technique called discrete wavelet transform (DWT). After prediction between the frames and within the frame, the prediction residual signal (also referred to as a prediction error signal) undergoes the transformation and quantization, and finally becomes a binary signal (bit stream) by entropy coding. .

  In HEVC, encoding is performed in units of blocks called coding units (Coding Units: CU), and the maximum unit is called Large Coding Unit (LCU), and HEVC reference software (HEVC test Model: In HM), a size of 64 × 64 is usually set. The 64 × 64 LCU is divided into 8 × 8 CUs as a minimum unit on a quadtree basis. Each CU is divided into a block that performs prediction called a prediction unit (PU) and a block that performs transformation called a transform unit (TU). These PU and TU are defined independently in the CU. PUs and TUs are basically divided on a quadtree basis like CUs, but there are tools that apply divisions other than squares. A tool that allows non-square partitioning of PU is called Asymmetric Motion Partition (AMP), and a tool that allows non-square partitioning of TU is called Non-Square Quadtree Transform (NSQT). The latter NSQT is not adopted in the Final Draft International Standard (FDIS) issued in January 2013, and is expected to be issued in the spring of 2014. 1 is not adopted. AMP is HEVC ver. 1 is adopted.

  FIG. 22 is a diagram showing the definition of each processing unit in the HEVC. H. In H.264 / AVC, encoding is performed in 16 × 16 blocks called macro blocks (MB), but in HEVC, encoding is performed in units of larger blocks such as 64 × 64 blocks. There are features to be performed. Encoding, prediction, and conversion processing at a larger size as described above greatly contributes to improvement of encoding efficiency particularly at high resolution.

  HEVC ver. Reference numeral 1 is composed of three types of profiles, which are Main Profile, Main 10 Profile (for 10-bit input signal), and Main Still Picture Profile (for still image), respectively. The input image signal of each profile has only a YUV 4: 2: 0 color format (hereinafter referred to as 4: 2: 0). 4: 2: 0 indicates a format in which the sampling of the color difference component is thinned out in half in the horizontal direction and the vertical direction with respect to the luminance component. Other than 4: 2: 0, YUV4: 4: 4 format (format with same sampling of luminance signal and chrominance signal, hereinafter 4: 4: 4) and YUV4: 2: 2 format (chrominance signal with respect to luminance signal) There is a format in which sampling is thinned in half only in the horizontal direction (hereinafter referred to as 4: 2: 2), and these are used for professional video editing and broadcasting station material transmission. As a color space, there is an RGB format in addition to the YUV format described above. The RGB format is widely used in video reproduction such as output from digital cameras and various displays.

  In order to support 4: 4: 4, 4: 2: 2, and RGB4: 4: 4 data (hereinafter referred to as RGB4: 4: 4), HEVC ver. 2 standardization is ver. Progressed in parallel with 1. In order to issue a final draft amendment (FDAM) in January 2014, HEVC ver. Standardization is progressing about 1 year later.

  HEVC ver. Regarding the standardization of No. 2, a highly efficient coding tool has been proposed particularly for compression of color difference signals. For example, for a residual signal after prediction in units of CU or PU, a color space conversion matrix is derived from the decoded signal, color space conversion is performed, and the color space converted signal is input to the encoder. Then, a method has been proposed in which inverse color space conversion is performed on the color space conversion decoded signal decoded by the decoder, and the signal is returned to the original color space and output.

  In addition to this, various coding tools such as an improved in-loop filter for color difference and an improved intra prediction for lossless compression have been proposed.

Benjamin Bross, Woo-Jin Han, Jens-Rainer Ohm, Gary J. Sullivan, Ye-Kui Wang, Thomas Wiegand, “High efficiency video coding (HEVC) text specification draft 10 (for FDIS & Consent)”, JCTVC-L1003_v20. doc, Joint Collaborative Team on Video Coding (JCT-VC) of ITU-T SG16 WP3 and ISO / IEC JTC1 / SC29 / WG11, pp.149-162, January 2013 Kei Kawamura, Tomonobu Yoshino, and Sei Naito, “AHG7: In-loop color-space transformation of residual signals for range extensions”, JCTVC-L0371.doc, Joint Collaborative Team on Video Coding (JCT-VC) of ITU-T SG16 WP3 and ISO / IEC JTC1 / SC29 / WG11, January 2013

  By the way, in the color space conversion for the residual signal in the loop described in Non-Patent Document 2, there is no particular restriction except when the prediction residual energy becomes 0 when performing the color space conversion. Always applied color space conversion. Depending on the block such as CU or PU, there is a case where it is better not to apply the color space conversion from the viewpoint of coding efficiency. Even in such a case, since the color space conversion is always applied, there is a possibility that the encoding efficiency is deteriorated. Actually, in the experimental data of Non-Patent Document 2, the encoding efficiency is degraded in some images. In an image in which deterioration occurs, large deterioration occurs particularly in the color difference component.

  The present invention has been made in view of such circumstances, and an image encoding method, an image decoding method, and an image encoding that can improve encoding efficiency by adapting application of color space conversion. An object of the present invention is to provide a device, an image decoding device, a program thereof, and a recording medium on which the program is recorded.

  The present invention is an image encoding method performed by an image encoding apparatus that encodes an image by applying color space conversion, and calculates a color space conversion matrix for performing the color space conversion from a decoded reference signal. A matrix calculation step, an eigenvalue acquisition step of acquiring an eigenvalue obtained in the process of calculating the color space conversion matrix, and whether or not to execute the color space conversion based on a magnitude relationship between a predetermined threshold and the eigenvalue A determination step for determining; and a color space conversion execution step for executing the color space conversion based on the calculation result of the color space conversion matrix according to the determination result.

  According to the present invention, in the eigenvalue acquisition step, eigenvalues of a plurality of components are acquired, and in the determination step, a weighted sum of the plurality of eigenvalues is calculated, and the magnitude relationship between the weighted sum and a predetermined threshold is calculated. Based on this, it is determined whether or not to execute the color space conversion.

  According to the present invention, in the eigenvalue acquisition step, eigenvalues of a plurality of components are acquired, and in the determination step, the color space conversion is performed based on a magnitude relationship between an eigenvalue corresponding to each component and a predetermined threshold corresponding to the eigenvalue. It is characterized by determining whether to execute.

  The present invention is an image encoding method performed by an image encoding apparatus that encodes an image by applying color space conversion, and calculates a color space conversion matrix for performing the color space conversion from a decoded reference signal. A matrix calculation step, a distance calculation step for calculating a distance between the color space conversion matrix and a predetermined reference matrix, and whether or not to execute the color space conversion based on a magnitude relationship between a predetermined threshold and the distance. A determination step for determining; and a color space conversion execution step for executing the color space conversion based on the calculation result of the color space conversion matrix according to the determination result.

  The present invention is an image encoding method performed by an image encoding apparatus that encodes an image by applying color space conversion, and calculates a color space conversion matrix for performing the color space conversion from a decoded reference signal. A matrix calculation step, an eigenvalue acquisition step of acquiring an eigenvalue obtained in the process of calculating the color space conversion matrix, a distance calculation step of calculating a distance between the color space conversion matrix and a predetermined reference matrix, and the eigenvalue A determination step for determining whether or not to execute the color space conversion using the distance and a predetermined threshold; and the color space conversion based on a calculation result of the color space conversion matrix according to the determination result And a color space conversion executing step for executing.

  The present invention is characterized in that a predetermined threshold is encoded and transmitted.

  The present invention is an image decoding method performed by an image decoding apparatus that decodes encoded data by applying an inverse color space conversion, wherein an inverse color space conversion matrix for performing the inverse color space conversion from a decoded reference signal is provided. The inverse color space conversion is executed based on a matrix calculation step for calculating, an eigenvalue acquisition step for acquiring an eigenvalue obtained in the process of calculating the inverse color space conversion matrix, and a magnitude relationship between a predetermined threshold and the eigenvalue. A determination step of determining whether or not and a reverse color space conversion execution step of executing the reverse color space conversion based on the calculation result of the reverse color space conversion matrix according to the determination result To do.

  According to the present invention, in the eigenvalue acquisition step, eigenvalues of a plurality of components are acquired, and in the determination step, a weighted sum of the plurality of eigenvalues is calculated, and the magnitude relationship between the weighted sum and a predetermined threshold is calculated. Based on this, it is determined whether or not to perform the inverse color space conversion.

  According to the present invention, in the eigenvalue acquisition step, eigenvalues of a plurality of components are acquired, and in the determination step, the inverse color space is based on a magnitude relationship between an eigenvalue corresponding to each component and a predetermined threshold corresponding to the eigenvalue. It is characterized by determining whether or not to execute conversion.

  The present invention is an image decoding method performed by an image decoding apparatus that decodes encoded data by applying an inverse color space conversion, wherein an inverse color space conversion matrix for performing the inverse color space conversion from a decoded reference signal is provided. The inverse color space conversion is executed based on a matrix calculation step for calculating, a distance calculation step for calculating a distance between the inverse color space conversion matrix and a predetermined reference matrix, and a magnitude relationship between a predetermined threshold and the distance. A determination step of determining whether or not and a reverse color space conversion execution step of executing the reverse color space conversion based on the calculation result of the reverse color space conversion matrix according to the determination result To do.

  The present invention is an image decoding method performed by an image decoding apparatus that decodes encoded data by applying an inverse color space conversion, wherein an inverse color space conversion matrix for performing the inverse color space conversion from a decoded reference signal is provided. A matrix calculation step for calculating, an eigenvalue acquisition step for acquiring an eigenvalue obtained in the process of calculating the inverse color space conversion matrix, and a distance calculation step for calculating a distance between the inverse color space conversion matrix and a predetermined reference matrix; A determination step of determining whether to perform the inverse color space conversion using the eigenvalue, the distance, and a predetermined threshold; and a calculation result of the inverse color space conversion matrix according to the determination result And a reverse color space conversion execution step for executing the reverse color space conversion based on the above.

  The present invention is characterized in that a predetermined threshold value is set by decoding from a bit stream.

  The present invention is an image encoding device that encodes an image by applying color space conversion, matrix calculation means for calculating a color space conversion matrix for performing the color space conversion from a decoded reference signal, and Eigenvalue acquisition means for acquiring eigenvalues obtained in the process of calculating a color space conversion matrix, and determination means for determining whether or not to execute the color space conversion based on a magnitude relationship between a predetermined threshold value and the eigenvalues; And color space conversion executing means for executing the color space conversion based on the calculation result of the color space conversion matrix according to the result of the determination.

  In the present invention, the eigenvalue acquisition unit acquires eigenvalues of a plurality of components, and the determination unit calculates a weighted sum of the plurality of eigenvalues, and determines a magnitude relationship between the weighted sum and a predetermined threshold value. Based on this, it is determined whether or not to execute the color space conversion.

  In the present invention, the eigenvalue acquisition unit acquires eigenvalues of a plurality of components, and the determination unit performs the color space conversion based on a magnitude relationship between an eigenvalue corresponding to each component and a predetermined threshold corresponding to the eigenvalue. It is characterized by determining whether to execute.

  The present invention is an image encoding device that encodes an image by applying color space conversion, matrix calculation means for calculating a color space conversion matrix for performing the color space conversion from a decoded reference signal, and Distance calculating means for calculating the distance between the color space conversion matrix and a predetermined reference matrix; and determining means for determining whether or not to execute the color space conversion based on a magnitude relationship between a predetermined threshold and the distance; And color space conversion executing means for executing the color space conversion based on the calculation result of the color space conversion matrix according to the result of the determination.

  The present invention is an image encoding device that encodes an image by applying color space conversion, matrix calculation means for calculating a color space conversion matrix for performing the color space conversion from a decoded reference signal, and Eigenvalue acquisition means for acquiring eigenvalues obtained in the process of calculating the color space conversion matrix, distance calculation means for calculating a distance between the color space conversion matrix and a predetermined reference matrix, the eigenvalue, the distance, and a predetermined distance Determination means for determining whether or not to execute the color space conversion using a threshold value of the color space, and color space conversion for executing the color space conversion based on the calculation result of the color space conversion matrix according to the determination result Execution means.

  The present invention is characterized in that a predetermined threshold is encoded and transmitted.

  The present invention is an image decoding apparatus for decoding encoded data by applying inverse color space conversion, and matrix calculation means for calculating an inverse color space conversion matrix for performing the inverse color space conversion from a decoded reference signal And eigenvalue acquisition means for acquiring eigenvalues obtained in the process of calculating the inverse color space conversion matrix, and determining whether or not to perform the inverse color space conversion based on a magnitude relationship between a predetermined threshold and the eigenvalue And a reverse color space conversion executing means for executing the reverse color space conversion based on the calculation result of the reverse color space conversion matrix according to the determination result.

  In the present invention, the eigenvalue acquisition unit acquires eigenvalues of a plurality of components, and the determination unit calculates a weighted sum of the plurality of eigenvalues, and determines a magnitude relationship between the weighted sum and a predetermined threshold value. Based on this, it is determined whether or not to perform the inverse color space conversion.

  In the present invention, the eigenvalue acquisition unit acquires eigenvalues of a plurality of components, and the determination unit determines the inverse color space based on a magnitude relationship between an eigenvalue corresponding to each component and a predetermined threshold corresponding to the eigenvalue. It is characterized by determining whether or not to execute conversion.

  The present invention is an image decoding apparatus for decoding encoded data by applying inverse color space conversion, and matrix calculation means for calculating an inverse color space conversion matrix for performing the inverse color space conversion from a decoded reference signal And a distance calculating means for calculating a distance between the inverse color space conversion matrix and a predetermined reference matrix, and determining whether to execute the inverse color space conversion based on a magnitude relationship between a predetermined threshold and the distance And a reverse color space conversion executing means for executing the reverse color space conversion based on the calculation result of the reverse color space conversion matrix according to the determination result.

  The present invention is an image decoding apparatus for decoding encoded data by applying inverse color space conversion, and matrix calculation means for calculating an inverse color space conversion matrix for performing the inverse color space conversion from a decoded reference signal Eigenvalue acquisition means for acquiring eigenvalues obtained in the process of calculating the inverse color space conversion matrix, distance calculation means for calculating a distance between the inverse color space conversion matrix and a predetermined reference matrix, and the eigenvalues; A determination unit configured to determine whether to perform the reverse color space conversion using the distance and a predetermined threshold; and the reverse color based on the calculation result of the reverse color space conversion matrix according to the determination result. Inverse color space conversion executing means for executing space conversion is provided.

  The present invention is characterized in that a predetermined threshold value is set by decoding from a bit stream.

  The present invention is an image encoding program for causing a computer to execute the image encoding method.

  The present invention is an image decoding program for causing a computer to execute the image decoding method.

  The present invention is a computer-readable recording medium on which the image encoding program is recorded.

  The present invention is a computer-readable recording medium on which the image decoding program is recorded.

  According to the present invention, in encoding and decoding using color space conversion, it is determined whether to use color space conversion by using a parameter being encoded or a parameter obtained in color space conversion processing. By performing color space conversion adaptively, it is possible to achieve the effect of improving the coding efficiency and speeding up the processing.

It is a block diagram which shows the structure of the image coding apparatus in the 1st Embodiment of this invention. It is a block diagram which shows the structure of the image decoding apparatus in the 1st Embodiment of this invention. FIG. 2 is a block diagram illustrating a configuration example of a color space conversion processing unit 104 illustrated in FIG. 1. 4 is a flowchart illustrating an operation of color space conversion processing performed by a color space conversion processing unit 104 illustrated in FIG. 3. FIG. 3 is a block diagram illustrating a configuration example of an inverse color space conversion processing unit 109 illustrated in FIG. 1 and an inverse color space conversion processing unit 206 illustrated in FIG. 2. It is a flowchart which shows the operation | movement of the reverse color space conversion process which the reverse color space conversion process part 109 shown in FIG. 5 performs. It is a block diagram which shows the structure of the color space conversion process part 104 'in 2nd Embodiment. It is explanatory drawing which shows the determination process regarding the presence or absence of color space conversion implementation. It is a flowchart which shows the operation | movement of the color space conversion process which the color space conversion process part 104 'shown in FIG. 7 performs. It is a block diagram which shows the structural example of the reverse color space conversion process part 109 'and reverse color space conversion process part 206' in 2nd Embodiment. 11 is a flowchart illustrating an operation of a reverse color space conversion process performed by a reverse color space conversion processing unit 109 ′ illustrated in FIG. 10. It is a block diagram which shows the structure of the color space conversion process part 104 '' in 3rd Embodiment. 13 is a flowchart illustrating an operation of color space conversion processing performed by a color space conversion processing unit 104 ″ illustrated in FIG. FIG. 10 is a block diagram illustrating a configuration example of an inverse color space conversion processing unit 109 ″ and an inverse color space conversion processing unit 206 ″ according to a third embodiment. 15 is a flowchart showing an operation of a reverse color space conversion process performed by a reverse color space conversion processing unit 109 '' shown in FIG. It is a block diagram which shows the structure of the color space conversion process part 104 "'in 4th Embodiment. FIG. 17 is a flowchart illustrating an operation of color space conversion processing performed by a color space conversion processing unit 104 ″ ″ illustrated in FIG. 16. It is a block diagram which shows the structural example of the reverse color space conversion process part 109 "" and reverse color space conversion process part 206 "" in 4th Embodiment. FIG. 19 is a flowchart illustrating an operation of a reverse color space conversion process performed by a reverse color space conversion processing unit 109 ″ ″ illustrated in FIG. 18. It is a figure which shows the hardware constitutions in the case of comprising a moving image encoder by a computer and a software program. It is a figure which shows the hardware constitutions in the case of comprising a moving image decoding apparatus with a computer and a software program. It is a figure which shows the definition of each processing unit in HEVC.

<First Embodiment>
Hereinafter, an image encoding device and an image decoding device according to a first embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a configuration of an image encoding device according to the embodiment. The image encoding device 100 receives an input video signal to be encoded, divides each frame of the input video signal into a plurality of blocks, encodes each block, and outputs a bit stream of the encoding result as an encoded stream To do. The image includes a still image, a moving image, and a video.

  As illustrated in FIG. 1, the image encoding device 100 includes an intra prediction processing unit 101, an inter prediction processing unit 102, a prediction residual signal generation unit 103, a color space conversion processing unit 104, an orthogonal transformation processing unit 105, and a quantization process. Unit 106, inverse quantization processing unit 107, inverse orthogonal transform processing unit 108, inverse color space transform processing unit 109, decoded signal generation unit 110, frame memory 111, in-loop filter processing unit 112, frame buffer 113, inter prediction information storage Section 114, intra prediction information storage section 115, color space conversion matrix calculation section 116, and entropy encoding processing section 117.

  In the image encoding device 100 shown in FIG. 1, a color space conversion processing unit 104 and an inverse color space conversion processing unit 109 are characteristic configurations. The other functional units are H.264. It is equivalent to a functional unit included in a general image encoding device used as an encoder such as H.264 / AVC and HEVC. Note that the in-loop filter processing unit 112 may include a plurality of in-loop filter processes such as a deblocking filter (DF) and a sample adaptive offset (SAO) employed in HEVC. An in-loop filter called adaptive loop filter (ALF), which is proposed during HEVC formulation, for removing coding noise may be included in the in-loop filter processing unit 112.

  The intra prediction processing unit 101 generates a prediction signal based on the decoded signal output from the frame memory 111. Also, the intra prediction processing unit 101 performs cost calculation based on the decoded signal and the input video signal, generates intra prediction information including a prediction mode, and stores the generated intra prediction information in the intra prediction information storage unit 115. Store. The inter prediction processing unit 102 receives the input video signal and the reference image (decoded signal after the in-loop filter) output from the frame buffer 113. The inter prediction processing unit 102 determines a motion vector based on the input video signal and the reference image, and generates a prediction signal. Also, the inter prediction processing unit 102 generates inter prediction information including the obtained motion vector, and stores the generated inter prediction information in the inter prediction information storage unit 114 for storage.

  The prediction residual signal generation unit 103 calculates a difference between the input video signal and a prediction signal output from the intra prediction processing unit 101 or the inter prediction processing unit 102. The prediction residual signal generation unit 103 outputs the calculated difference to the color space conversion processing unit 104 as a prediction residual signal. The color space conversion processing unit 104 performs color space conversion on the prediction residual signal input from the prediction residual signal generation unit 103 using the color space conversion matrix obtained by the color space conversion matrix calculation unit 116. To do. The color space conversion processing unit 104 outputs the coefficient value after color space conversion to the orthogonal transformation processing unit 105.

  The orthogonal transformation processing unit 105 performs orthogonal transformation such as DCT on the coefficient value after color space transformation input from the color space transformation processing unit 104. Not only DCT but also H.264. The integer conversion used in H.264 / AVC and HEVC may be used. The orthogonal transform processing unit 105 outputs the orthogonal transform coefficient value obtained by the orthogonal transform to the quantization processing unit 106. The quantization processing unit 106 quantizes the orthogonal transformation coefficient value input from the orthogonal transformation processing unit 105 and outputs the quantization coefficient value to the inverse quantization processing unit 107 and the entropy coding processing unit 117.

  The inverse quantization processing unit 107 performs inverse quantization on the quantization coefficient value input from the quantization processing unit 106 and outputs the result to the inverse orthogonal transform processing unit 108. The inverse orthogonal transform processing unit 108 performs inverse orthogonal transform on the quantization coefficient value input from the inverse quantization processing unit 107. The inverse orthogonal transform processing unit 108 outputs the decoded color space post-conversion coefficient value obtained by the inverse orthogonal transform to the inverse color space conversion processing unit 109.

  The inverse color space conversion processing unit 109 inputs the decoded color space post-conversion coefficient value input from the inverse orthogonal conversion processing unit 108 and the inverse color space conversion matrix obtained from the color space conversion matrix calculation unit 116, Perform spatial transformation. The prediction residual decoded signal obtained by the inverse color space conversion is output to the decoded signal generation unit 110. The decoded signal generation unit 110 adds the prediction residual decoded signal input from the inverse color space conversion processing unit 109 and the prediction signal output from the intra prediction processing unit 101 or the inter prediction processing unit 102. The decoded signal generation unit 110 stores the addition result in the frame memory 111 as a decoded signal. This decoded signal is used as a reference image in the intra prediction processing unit 101 or the inter prediction processing unit 102.

  The frame memory 111 stores the decoded signal generated and output by the decoded signal generation unit 110. The stored decoded signal is output to the in-loop filter processing unit 112. Similarly, in-loop filtered decoded signals are output to the color space conversion matrix calculator 116 in order to generate a color space conversion matrix necessary for color space conversion. The in-loop filter processing unit 112 reads the decoded signal stored in the frame memory 111 and performs an in-loop filter process on the read decoded signal. For example, DF and / or SAO are implemented. Another in-loop filter such as ALF may be separately applied. The decoded signal after in-loop filtering is output to the frame buffer 113.

  The frame buffer 113 outputs the intra-loop filtered decoded signal to the inter prediction processing unit 102 as a reference image. Similarly, in-loop filtered decoded signals are output to the color space conversion matrix calculator 116 in order to generate a color space conversion matrix necessary for color space conversion. The inter prediction information storage unit 114 stores the inter prediction information generated by the inter prediction processing unit 102. The intra prediction information storage unit 115 stores the intra prediction information generated by the intra prediction processing unit 101. The stored prediction information is output to the entropy encoding processing unit 117.

  The color space conversion matrix calculation unit 116 inputs a decoded signal from the frame memory 111 at the time of intra prediction, inputs a decoded signal from the frame buffer 113 at the time of inter prediction, and performs principal component analysis from the obtained decoded signal. The color space conversion matrix necessary for the color space conversion is calculated. The inverse color space conversion matrix is a relationship between a color space conversion matrix and a transposed matrix. The obtained color space conversion matrix and reverse color space conversion matrix are output to the color space conversion processing unit 104 and the reverse color space conversion processing unit 109, respectively.

  The entropy encoding processing unit 117 stores the quantization coefficient value output from the quantization processing unit 106, the inter prediction information stored in the inter prediction information storage unit 114, and the intra prediction information storage unit 115. The intra prediction information is entropy encoded and output as an encoded stream. The image encoding apparatus 100 includes the above-described functional units, thereby dividing each frame of the input video signal into a plurality of blocks, performing block-based predictive encoding with color space conversion, and converting the input video signal to An encoded stream obtained by encoding is output.

  FIG. 2 is a block diagram illustrating a configuration of the image decoding device 200 according to the first embodiment. The image decoding apparatus 200 receives an encoded stream that is encoded and output by the image encoding apparatus 100 shown in FIG. 1, and outputs a decoded video signal that is a decoded image by decoding the encoded stream. .

  As shown in FIG. 2, the image decoding apparatus 200 includes an entropy decoding processing unit 201, an intra prediction processing unit 202, an inter prediction processing unit 203, an inverse quantization processing unit 204, an inverse orthogonal transform processing unit 205, an inverse color space conversion process. Unit 206, decoded signal generation unit 207, frame memory 208, in-loop filter processing unit 209, frame buffer 210, inter prediction information storage unit 211, intra prediction information storage unit 212, and color space conversion matrix calculation unit 213. .

  In the image decoding apparatus 200 shown in FIG. 2, the inverse color space conversion processing unit 206 has a characteristic configuration. The other functional units are H.264. The configuration is the same as that of a general image decoding device used as a decoder such as H.264 / AVC and HEVC.

  The entropy decoding processing unit 201 receives an encoded stream, entropy-decodes the quantization coefficient value of the decoding target block from the input encoded stream, and entropy-decodes intra prediction information related to intra prediction and inter prediction information related to inter prediction. To do. Also, the entropy decoding processing unit 201 outputs the quantized coefficient value to the inverse quantization processing unit 204, stores the inter prediction information in the inter prediction information storage unit 211, and stores the intra prediction information in the intra prediction information storage unit. The data is stored in 212. The intra prediction processing unit 202 reads the decoded signal stored in the frame memory 208 as a reference image. Further, the intra prediction processing unit 202 reads the intra prediction information from the intra prediction information storage unit 212. Then, the intra prediction processing unit 202 generates a prediction signal based on the read reference image and the read intra prediction information.

  The inter prediction processing unit 203 reads inter prediction information from the inter prediction information storage unit 211. Then, the inter prediction processing unit 203 generates a prediction signal based on the inter prediction information and the reference image input from the frame buffer 210. The inverse quantization processing unit 204 inversely quantizes the quantized coefficient value input from the entropy decoding processing unit 201 to calculate a decoded orthogonal transform coefficient value, and the obtained decoded orthogonal transform coefficient value is sent to the inverse orthogonal transform processing unit 205. Output.

  The inverse orthogonal transform processing unit 205 performs inverse orthogonal transform on the decoded orthogonal transform coefficient value input from the inverse quantization processing unit 204, calculates a decoded color space converted coefficient value, and calculates the calculated decoded color space converted coefficient value. The result is output to the inverse color space conversion processing unit 206. The inverse color space conversion processing unit 206 uses the inverse color space conversion matrix obtained from the color space conversion matrix calculation unit 213 for the decoded color space post-conversion coefficient value input from the inverse orthogonal transformation processing unit 205, and the inverse color space. Apply conversion. The prediction residual decoded signal obtained thereby is output to decoded signal generation section 207.

  The decoded signal generation unit 207 adds the prediction residual decoded signal input from the inverse color space conversion processing unit 206 and the prediction signal output from the intra prediction processing unit 202 or the inter prediction processing unit 203. Then, the decoded signal generation unit 207 stores the addition result in the frame memory 208 as a decoded signal of the decoding target block and stores it. The frame memory 208 stores the decoded signal calculated by the decoded signal generation unit 207.

  The in-loop filter processing unit 209 reads the decoded signal from the frame memory 208 and performs a filter process for reducing the coding distortion on the image indicated by the read decoded signal. For example, DF and / or SAO are implemented. Another in-loop filter such as ALF may be separately applied. The in-loop filter processing unit 209 outputs the filtered image as a decoded video signal. Further, the in-loop filter processing unit 209 outputs the filtered image to the frame buffer 210 in order to use it as a reference image.

  The inter prediction information storage unit 211 stores the inter prediction information decoded by the entropy decoding processing unit 201. The intra prediction information storage unit 212 stores the intra prediction information decoded by the entropy decoding processing unit 201. The color space conversion matrix calculation unit 213 inputs a decoded signal from the frame memory 208 at the time of intra prediction, inputs a decoded signal from the frame buffer 210 at the time of inter prediction, and performs principal component analysis from the obtained decoded signal. The color space conversion matrix necessary for the color space conversion is calculated. The inverse color space conversion matrix is a relationship between a color space conversion matrix and a transposed matrix. The obtained inverse color space conversion matrix is output to the inverse color space conversion processing unit 206.

  With the above-described configuration, the image decoding apparatus 200 receives an encoded video stream encoded by block-based predictive encoding using color space conversion, and decodes the video from the input encoded stream. Then, the decoded video signal is output.

  Next, the detailed configuration of the color space conversion processing unit 104 shown in FIG. 1 will be described with reference to FIG. FIG. 3 is a block diagram showing a configuration of the color space conversion processing unit 104 shown in FIG. As shown in FIG. 3, the color space conversion processing unit 104 includes a prediction residual signal storage unit 301, a color space conversion matrix storage unit 302, a color space conversion execution unit 303, an eigenvalue storage unit 304, and a color space conversion execution determination unit 305. And a threshold value setting unit 306. Here, in FIG. 3, the eigenvalue storage unit 304, the color space conversion execution determination unit 305, and the threshold setting unit 306 are characteristic configurations.

  The prediction residual signal storage unit 301 inputs and stores the prediction residual signal of the block to be processed. Further, the color space conversion matrix storage unit 302 receives and stores a color space conversion matrix obtained from the decoded signal. The eigenvalue storage unit 304 receives and stores eigenvalues calculated in the process of principal component analysis for calculating a color space conversion matrix from the color space conversion matrix calculation unit 116.

  The color space conversion execution determination unit 305 uses the eigenvalue information read from the eigenvalue storage unit 304 and the threshold set by the threshold setting unit 306 to determine whether or not to perform color space conversion. For example, if the eigenvalue that is the first principal component (the eigenvalue that indicates the largest value) is equal to or greater than the set threshold, it can be determined that extreme color space conversion is not performed, and therefore it is determined that color space conversion is to be performed. . If it is smaller than the threshold value, it is determined that the color space conversion is not performed. Information regarding whether or not to implement these is defined as color space conversion execution information, and is output to the color space conversion execution unit 303. Although the case where the eigenvalue of the first principal component is compared with the threshold value is shown, for example, two types of eigenvalues of the first principal component and the second principal component are acquired from the eigenvalue storage unit 304, and the threshold value corresponding to each is obtained as the threshold value. A configuration may be adopted in which color space conversion is performed when both the first principal component and the second principal component are simultaneously equal to or greater than the respective threshold values, acquired from the setting unit 306.

  The threshold setting unit 306 reads a threshold from a predetermined table or the like, and sets the threshold to the color space conversion execution determination unit 305. The threshold value is in the range of 0 to 1 from the range that the eigenvalue can take, but is expressed as an integer value of 0 to 4096, for example, depending on the internal structure (0 indicates the eigenvalue 0, and 4096 is In the case where eigenvalue 1 is indicated), a value corresponding thereto may be set. If the threshold value is set to 0.5, the threshold value to be set may be set to 2048. A threshold conversion processing unit that performs such replacement may be provided between the threshold setting unit 306 and the color space conversion execution determination unit 305. The threshold value may be encoded and transmitted to the decoding side. In the case of encoding and transmitting, a threshold encoding unit is separately prepared, and the threshold output from the threshold setting unit 306 is entropy encoded. Switching can be performed in sequence units, GOP (Group of Pictures) units, slice units, frame units, block units, and the like. When switching, it is necessary to encode and transmit threshold values in the units. If transmission is performed with a fine granularity, the code amount (overhead) of threshold information increases. However, since adaptive color space conversion can be controlled accordingly, it is possible to improve coding efficiency and speed up processing.

  The color space conversion execution unit 303 includes a prediction residual signal output from the prediction residual signal storage unit 301, a color space conversion matrix output from the color space conversion matrix storage unit 302, and a color space conversion execution determination unit 305. Color space conversion is performed using the output color space conversion execution information. If the color space conversion execution information indicates that color space conversion is to be performed, color space conversion is performed. If the color space conversion execution information indicates that the color space conversion is not performed, the color space conversion is not performed.

  As described above, the color space conversion processing unit 104 includes the prediction residual signal storage unit 301, the color space conversion matrix storage unit 302, the color space conversion execution unit 303, the eigenvalue storage unit 304, the color space conversion execution determination unit 305, and the threshold value. The setting unit 306 performs adaptive color space conversion processing by determining the magnitude of the eigenvalue and the threshold value, and outputs a coefficient value after color space conversion.

  Next, the processing operation of the color space conversion processing unit 104 shown in FIG. 3 will be described with reference to FIG. FIG. 4 is a flowchart showing the operation of the color space conversion process performed by the color space conversion processing unit 104 shown in FIG. First, the prediction residual signal storage unit 301 inputs the prediction residual signal of the block to be processed from the prediction residual signal generation unit 103 and stores it. Then, in order to perform color space conversion, the data is output to the color space conversion execution unit 303 (step S1). Subsequently, the color space conversion matrix storage unit 302 inputs and stores the color space conversion matrix output from the color space conversion matrix calculation unit 116. Then, since it is necessary to execute the color space conversion, it is output to the color space conversion execution unit 303 (step S2).

  Next, the eigenvalue storage unit 304 inputs and stores eigenvalues obtained when the color space conversion matrix calculation unit 116 calculates the color space conversion matrix. Then, the eigenvalue information is output to the color space conversion execution determination unit 305 in order to determine execution of the color space conversion. Further, the threshold setting unit 306 inputs threshold information for determining execution of color space conversion from a predetermined table and outputs it to the color space conversion execution determining unit 305 (step S3). Next, the color space conversion execution determination unit 305 receives the eigenvalue information read from the eigenvalue storage unit 304 and the threshold value set by the threshold value setting unit 306, and performs color space determination execution based on the magnitude relationship between these values. The presence / absence is determined, and information on the determination result is output to the color space conversion execution unit 303 as color space conversion execution information (step S4).

  Next, the color space conversion execution unit 303 reads the prediction residual signal read from the prediction residual signal storage unit 301, the color space conversion matrix read from the color space conversion matrix storage unit 302, and the color space conversion execution determination unit 305. The color space conversion execution information output by the user is input, and the color space execution determination result is determined based on the color space conversion execution information (step S5). If it is determined that the color space conversion is to be executed as a result of this determination, the color space conversion is executed (step S6). On the other hand, if it is determined not to execute, the color space conversion is not executed and the prediction residual signal is output as it is.

  Next, the configuration of the reverse color space conversion processing unit 109 shown in FIG. 1 and the reverse color space conversion processing unit 206 shown in FIG. 2 will be described with reference to FIG. FIG. 5 is a block diagram illustrating a configuration example of the reverse color space conversion processing unit 109 illustrated in FIG. 1 and the reverse color space conversion processing unit 206 illustrated in FIG. Since the reverse color space conversion processing unit 109 and the reverse color space conversion processing unit 206 perform common processing and have the same configuration, the reverse color space conversion processing unit 109 will be described as reverse color space conversion processing. The description of the inverse color space conversion processing unit 206 is omitted.

  As shown in FIG. 5, the inverse color space conversion processing unit 109 includes a decoded color space converted coefficient value storage unit 401, an inverse color space conversion matrix storage unit 402, an inverse color space conversion execution unit 403, an eigenvalue storage unit 404, an inverse A color space conversion execution determination unit 405 and a threshold setting unit 406 are provided. Here, in FIG. 5, the eigenvalue storage unit 404, the inverse color space conversion execution determination unit 405, and the threshold setting unit 406 are characteristic configurations.

  The decoded color space converted coefficient value storage unit 401 inputs and stores the decoded color space converted coefficient value of the block to be processed. Further, the inverse color space conversion matrix storage unit 402 inputs and stores an inverse color space conversion matrix obtained from the decoded signal. The eigenvalue storage unit 404 receives and stores eigenvalues calculated in the process of principal component analysis for calculating the inverse color space conversion matrix from the color space conversion matrix calculation unit 116.

  The reverse color space conversion execution determination unit 405 determines whether or not reverse color space conversion is performed using the eigenvalue information read from the eigenvalue storage unit 404 and the threshold value set from the threshold value setting unit 406. For example, if the eigenvalue that is the first principal component (the eigenvalue that indicates the largest value) is equal to or greater than the set threshold value, it is determined that extreme inverse color space conversion is not performed, and it is determined that inverse color space conversion is performed. To do. If it is smaller than the threshold value, it is determined that reverse color space conversion is not performed. Information regarding whether or not to perform these operations is defined as reverse color space conversion execution information, and is output to the reverse color space conversion execution unit 403. Although the case where the eigenvalue of the first principal component is compared with the threshold value is shown, for example, two types of eigenvalues of the first principal component and the second principal component are acquired from the eigenvalue storage unit 404, and the threshold value corresponding to each is obtained as the threshold value. A configuration may be adopted in which reverse color space conversion is performed when both the first principal component and the second principal component are simultaneously equal to or greater than the respective threshold values, acquired from the setting unit 406.

  The threshold setting unit 406 reads a threshold from a predetermined table or the like, and sets the threshold to the inverse color space conversion execution determination unit 405. The threshold value is in the range of 0 to 1 from the range that the eigenvalue can take, but is expressed as an integer value of 0 to 4096, for example, depending on the internal structure (0 indicates the eigenvalue 0, and 4096 is In the case where eigenvalue 1 is indicated), a value corresponding thereto may be set. If the threshold value is set to 0.5, the threshold value to be set may be set to 2048. A threshold conversion processing unit that performs such replacement may be provided between the threshold setting unit 406 and the color space conversion execution determination unit 405. When the threshold is configured to be transmitted on the encoding side, a threshold decoding unit is separately prepared, and the entropy decoded threshold is set in the threshold setting unit 406. When switching by sequence unit, GOP unit, slice unit, frame unit, block unit, etc., it is necessary to decode the threshold value in the unit according to the encoding side.

  The inverse color space conversion execution unit 403 includes the decoded color space converted coefficient value output from the decoded color space converted coefficient value storage unit 401 and the inverse color space conversion output from the inverse color space conversion matrix storage unit 402. Reverse color space conversion is performed using the matrix and reverse color space conversion execution information output from the reverse color space conversion execution determination unit 405. If the reverse color space conversion execution information indicates that reverse color space conversion is to be performed, reverse color space conversion is performed. If the reverse color space conversion execution information indicates that reverse color space conversion is not performed, reverse color space conversion is not performed.

  As described above, the inverse color space conversion processing unit 109 includes the decoded color space post-conversion coefficient value storage unit 401, the inverse color space conversion matrix storage unit 402, the inverse color space conversion execution unit 403, the eigenvalue storage unit 404, the inverse color space. The conversion execution determination unit 405 and the threshold setting unit 406 perform an adaptive inverse color space conversion process by determining the magnitude of the eigenvalue and the threshold, and output a prediction residual decoded signal.

  Next, the processing operation of the inverse color space conversion processing unit 109 shown in FIG. 5 will be described with reference to FIG. FIG. 6 is a flowchart showing the operation of the reverse color space conversion process performed by the reverse color space conversion processing unit 109 shown in FIG. The decoded color space converted coefficient value storage unit 401 inputs and stores the decoded color space converted coefficient value of the block to be processed. Then, in order to perform inverse color space conversion, the coefficient value after the decoded color space conversion is output to the inverse color space conversion execution unit 403 (step S11).

  Next, the inverse color space conversion matrix storage unit 402 receives and stores the inverse color space conversion matrix output from the color space conversion matrix calculation unit 116. Then, since it is necessary for executing the reverse color space conversion, it is output to the reverse color space conversion execution unit 403 (step S12). Next, the eigenvalue storage unit 404 receives and stores eigenvalues obtained when the color space conversion matrix calculation unit 116 calculates the inverse color space conversion matrix. Actually, the color space conversion matrix and the inverse color space conversion matrix have a transposed matrix relationship. In order to determine execution of the reverse color space conversion, the eigenvalue information is output to the reverse color space conversion execution determination unit 405. Further, the threshold setting unit 406 inputs threshold information for determining execution of the reverse color space conversion from a predetermined table and outputs the threshold information to the reverse color space conversion execution determination unit 405 (step S13).

  Next, the inverse color space conversion execution determination unit 405 receives the eigenvalue information output from the eigenvalue storage unit 404 and the threshold value output from the threshold setting unit 406 as input, and determines the inverse color space determination based on the magnitude relationship between these values. The presence / absence of execution is determined, and the information of the determination result is output to the reverse color space conversion execution unit 403 as reverse color space conversion execution information (step S14).

  Next, the inverse color space conversion execution unit 403 outputs the decoded color space converted coefficient value output from the decoded color space converted coefficient value storage unit 401 and the inverse color space output from the inverse color space conversion matrix storage unit 402. Using the conversion matrix and the reverse color space conversion execution information output by the reverse color space conversion execution determination unit 405 as input, the reverse color space execution determination result is determined based on the reverse color space conversion execution information (step S15). If it is determined that color space conversion is to be executed, reverse color space conversion is executed (step S16). If it is determined not to execute, the inverse color space conversion is not executed and the prediction residual decoded signal is output as it is.

  As described above, the color space conversion processing unit 104 and the inverse color space conversion processing unit 109 determine whether or not to perform color space conversion using the eigenvalue information obtained when generating the color space conversion matrix. As a result, an adaptive color space conversion process can be performed in units of blocks. Depending on the threshold value to be set, it is possible to greatly omit the processing related to color space conversion, and an increase in speed can be expected. In addition, unnecessary conversion that causes deterioration in encoding efficiency can be omitted, which can contribute to improvement in encoding efficiency.

<Second Embodiment>
Next, an image encoding device and an image decoding device according to the second embodiment of the present invention will be described. The second embodiment differs from the configuration of the color space conversion processing unit 104 shown in FIG. 3 only in the inverse color space conversion processing units 109 and 206 shown in FIG. Therefore, only the configuration of the color space conversion processing unit 104 ′ and the inverse color space conversion processing unit 109 ′ will be described here. Note that the reverse color space conversion processing unit 209 ′ is common to the reverse color space conversion processing unit 109 ′, and thus description thereof is omitted.

  FIG. 7 is a block diagram showing the configuration of the color space conversion processing unit 104 '. As shown in FIG. 7, the color space conversion processing unit 104 ′ includes a prediction residual signal storage unit 501, a color space conversion matrix storage unit 502, a color space conversion execution unit 503, a reference matrix setting unit 504, and a color space conversion execution determination unit. 505 and a threshold setting unit 506 are provided.

  The prediction residual signal storage unit 501, the color space conversion matrix storage unit 502, and the color space conversion execution unit 503 are respectively the prediction residual signal storage unit 301, the color space conversion matrix storage unit 302, and the color space conversion execution unit illustrated in FIG. This is the same configuration as 303.

  The color space conversion execution determination unit 505 is set from the color space conversion matrix information read from the color space conversion matrix storage unit 502, the reference matrix information set from the reference matrix setting unit 504, and the threshold setting unit 506. Whether or not color space conversion is performed is determined using the threshold. For example, as shown in FIG. 8, in the YUV format, the distance between the input color space conversion matrix and the reference matrix is calculated, and the threshold value input from the threshold value setting unit 506 is compared. It is determined that color space conversion is performed by determining that the matrix is close to the reference matrix. If the distance is larger than the threshold value, it is determined that the distance from the reference matrix is determined, and it is determined that the color space conversion is not performed. FIG. 8 is an explanatory diagram illustrating a determination process regarding whether or not color space conversion is performed.

  The reference matrix setting unit 504 outputs a reference matrix predetermined for each format to the color space conversion execution determination unit 505 according to the format of the input image (RGB format or YUV format). The threshold setting unit 506 inputs and sets threshold information from a predetermined table, and outputs the threshold information to the color space conversion execution determination unit 505.

  The color space conversion execution unit 503 receives the prediction residual signal output from the prediction residual signal storage unit 501, the color space conversion matrix output from the color space conversion matrix storage unit 502, and the color space conversion execution determination unit 505. Color space conversion is performed using the output color space conversion execution information. If the color space conversion execution information indicates that color space conversion is to be performed, color space conversion is performed. If the color space conversion execution information indicates that the color space conversion is not performed, the color space conversion is not performed.

  As described above, the color space conversion processing unit 104 ′ includes the prediction residual signal storage unit 501, the color space conversion matrix storage unit 502, the color space conversion execution unit 503, the reference matrix setting unit 504, and the color space conversion execution determination unit 505. The threshold value setting unit 506 performs adaptive color space conversion processing by determining the distance between the color space conversion matrix and the reference matrix and the threshold value, and outputs a coefficient value after color space conversion.

  Next, the processing operation of the color space conversion processing unit 104 'shown in FIG. 7 will be described with reference to FIG. FIG. 9 is a flowchart showing the operation of the color space conversion process performed by the color space conversion processing unit 104 'shown in FIG. The prediction residual signal storage unit 501 inputs and stores the prediction residual signal of the block to be processed. Then, the data is output to the color space conversion execution unit 503 for color space conversion (step S21).

  Next, the color space conversion matrix storage unit 502 inputs and stores the color space conversion matrix output from the color space conversion matrix calculation unit 116. Then, since it is necessary for executing the color space conversion, it is output to the color space conversion executing unit 503 (step S22). Subsequently, the reference matrix setting unit 504 reads the reference matrix from the table according to the format of the input image, and outputs the information to the color space conversion execution determination unit 505. Similarly, the threshold setting unit 506 inputs threshold information for determining the execution of color space conversion from a predetermined table and outputs it to the color space conversion execution determining unit 505 (step S23).

  Next, the color space conversion execution determination unit 505 outputs the color space conversion matrix output from the color space conversion matrix storage unit 502, the reference matrix output from the reference matrix setting unit 504, and the threshold output from the threshold setting unit 506. As input. Then, an inter-vector distance (also referred to as an inter-matrix distance) between the input color space conversion matrix and the reference matrix is calculated (step S24). Based on the calculated relationship between the vector distance and the threshold value, the presence / absence of color space determination execution is determined, and the information is output to the color space conversion execution unit 503 as color space conversion execution information (step S25).

  Next, the color space conversion execution unit 503 includes a prediction residual signal output from the prediction residual signal storage unit 501, a color space conversion matrix output from the color space conversion matrix storage unit 502, and a color space conversion execution determination unit 505. The color space conversion execution information output by the user is input, and the color space execution determination result is determined based on the color space conversion execution information (step S26). If it is determined that the color space conversion is to be executed, the color space conversion is executed. (Step S27). On the other hand, if it is determined not to execute, the color space conversion is not executed and the prediction residual signal is output as it is.

  Next, the configurations of the inverse color space conversion processing unit 109 ′ and the inverse color space conversion processing unit 206 ′ according to the second embodiment will be described with reference to FIG. 10. FIG. 10 is a block diagram illustrating a configuration example of the reverse color space conversion processing unit 109 ′ and the reverse color space conversion processing unit 206 ′ illustrated in FIG. 2 according to the second embodiment. Since the reverse color space conversion processing unit 109 ′ and the reverse color space conversion processing unit 206 ′ perform the same processing and have the same configuration, as the reverse color space conversion processing, the reverse color space conversion processing unit 109 ′ is A description of the inverse color space conversion processing unit 206 ′ will be omitted.

  As shown in FIG. 10, the inverse color space conversion processing unit 109 ′ includes a decoded color space converted coefficient value storage unit 601, an inverse color space conversion matrix storage unit 602, an inverse color space conversion execution unit 603, and a reference matrix setting unit 604. , An inverse color space conversion execution determination unit 605 and a threshold setting unit 606 are provided. Here, in FIG. 10, a reference matrix setting unit 604, an inverse color space conversion execution determination unit 605, and a threshold setting unit 606 are characteristic configurations.

  The decoded color space converted coefficient value storage unit 601 receives and stores the decoded color space converted coefficient value of the block to be processed. The inverse color space conversion matrix storage unit 602 inputs and stores an inverse color space conversion matrix obtained from the decoded signal.

  The reference matrix setting unit 604 reads a reference matrix necessary for performing reverse color space conversion execution determination from a predetermined table and outputs the matrix to the reverse color space conversion execution determination unit 605. The table is a table that can be switched according to the format of the input image. For example, in the case of the YUV format, the matrix shown in FIG. 8 is output as a reference matrix. In the case of the RGB format, an RGB-YUV conversion matrix is used as a reference matrix. In the case of the RGB format, in addition to the matrix, for example, the color space conversion matrix obtained for the entire sequence of each image, the average value of the color space conversion matrix obtained for each frame, or the color space obtained for the previous slice / frame / block A conversion matrix may be used.

  The reverse color space conversion execution determination unit 605 receives the reverse color space conversion matrix input from the reverse color space conversion matrix storage unit 602, the reference matrix input from the reference matrix setting unit 604, and the threshold setting unit 606. Whether or not reverse color space conversion is performed is determined using the threshold. For example, if the distance between the inverse color space conversion matrix and the reference matrix is larger than the set threshold value in the processing procedure shown in FIG. 8, it can be determined that the vector is significantly larger than the reference matrix. It is determined that no spatial transformation is performed. If it is less than or equal to the threshold, it is determined that reverse color space conversion is to be performed. Since the color space conversion matrix and the inverse color space conversion matrix have a transposed matrix relationship, the reference matrix for the inverse color space conversion matrix may be considered to have a relationship between the color space conversion reference matrix and the transposed matrix. Based on this determination, information indicating whether to perform reverse color space conversion is defined as reverse color space conversion execution information, and is output to the reverse color space conversion execution unit 603.

  The threshold setting unit 606 inputs and sets threshold information from a predetermined table, and outputs the threshold information to the inverse color space conversion execution determination unit 605. The inverse color space conversion execution unit 603 includes the decoded color space converted coefficient value output from the decoded color space converted coefficient value storage unit 601 and the inverse color space conversion output from the inverse color space conversion matrix storage unit 402. Reverse color space conversion is performed using the matrix and reverse color space conversion execution information output from the reverse color space conversion execution determination unit 605. If the reverse color space conversion execution information indicates that reverse color space conversion is to be performed, reverse color space conversion is performed. If the reverse color space conversion execution information indicates that reverse color space conversion is not performed, reverse color space conversion is not performed.

  As described above, the inverse color space conversion processing unit 109 ′ includes the decoded color space post-conversion coefficient value storage unit 601, the inverse color space conversion matrix storage unit 602, the inverse color space conversion execution unit 603, the reference matrix setting unit 604, and the inverse. The color space conversion execution determination unit 605 and the threshold setting unit 606 perform adaptive inverse color space conversion processing by determining the size of the inter-matrix distance and the threshold, and output a prediction residual decoded signal.

  Next, the processing operation of the inverse color space conversion processing unit 109 'shown in FIG. 10 will be described with reference to FIG. FIG. 11 is a flowchart showing the operation of the reverse color space conversion process performed by the reverse color space conversion processing unit 109 'shown in FIG. The decoded color space converted coefficient value storage unit 601 receives and stores the decoded color space converted coefficient value of the block to be processed. In order to perform inverse color space conversion, the coefficient value after the decoded color space conversion is output to the inverse color space conversion execution unit 603 (step S31). Subsequently, the reverse color space conversion matrix storage unit 602 receives and stores the reverse color space conversion matrix output from the color space conversion matrix calculation unit 116. And since it is required in order to perform reverse color space conversion, it outputs to the reverse color space conversion execution part 603 (step S32).

  Next, the reference matrix setting unit 604 selects a reference matrix according to the input image format, and outputs the matrix information to the inverse color space conversion execution determination unit 605. Further, the threshold setting unit 606 inputs threshold information for determining execution of reverse color space conversion from a predetermined table and outputs the threshold information to the reverse color space conversion execution determination unit 605 (step S33).

  Next, the reverse color space conversion execution determination unit 605 receives the reverse color space conversion matrix information output from the reverse color space conversion matrix storage unit 602 and the reference matrix output from the reference matrix setting unit 604 as shown in FIG. Then, the distance between the two matrices is calculated (step S34). Thereafter, based on the magnitude relationship between the calculated inter-matrix distance and the threshold value output from the threshold setting unit 606, it is determined whether or not the reverse color space determination is performed, and the information is used as the reverse color space conversion execution information. It outputs to the space conversion execution part 603 (step S35).

  Next, the inverse color space conversion execution unit 603 outputs the decoded color space converted coefficient value output from the decoded color space converted coefficient value storage unit 601 and the inverse color space output from the inverse color space conversion matrix storage unit 602. Using the conversion matrix and the reverse color space conversion execution information output by the reverse color space conversion execution determination unit 605 as input, the reverse color space execution determination is performed based on the reverse color space conversion execution information (step S36). If it is determined that conversion is to be performed, reverse color space conversion is performed (step S37). On the other hand, if it is determined not to execute, the inverse color space conversion is not executed, and the prediction residual decoded signal is output as it is.

<Third Embodiment>
Next, an image encoding device and an image decoding device according to the third embodiment of the present invention will be described. The third embodiment differs from the configuration of the color space conversion processing unit 104 shown in FIG. 3 only in the inverse color space conversion processing units 109 and 206 shown in FIG. Therefore, only the configuration of the color space conversion processing unit 104 ″ and the inverse color space conversion processing unit 109 ″ in the third embodiment will be described. Since the reverse color space conversion processing unit 209 ″ is common to the reverse color space conversion processing unit 109 ″, description thereof is omitted.

  FIG. 12 is a block diagram illustrating a configuration of the color space conversion processing unit 104 ″ according to the third embodiment. As shown in FIG. 12, the color space conversion processing unit 104 ″ includes a prediction residual signal storage unit 701, a color space conversion matrix storage unit 702, a color space conversion execution unit 703, an eigenvalue storage unit 704, a reference matrix setting unit 705, A color space conversion execution determination unit 706 and a threshold setting unit 707 are provided. The prediction residual signal storage unit 701, the color space conversion matrix storage unit 702, and the color space conversion execution unit 703 are respectively the prediction residual signal storage unit 301, the color space conversion matrix storage unit 302, and the color space conversion execution unit illustrated in FIG. This is the same configuration as 303.

  The color space conversion execution determination unit 706 includes information on the color space conversion matrix read from the color space conversion matrix storage unit 702, eigenvalues input from the eigenvalue storage unit 704, and reference matrix input from the reference matrix setting unit 705. Using information and two types of threshold values set by the threshold value setting unit 707 (threshold value for eigenvalues and threshold value for inter-vector distance), the presence / absence of color space conversion is determined. For example, if the eigenvalue that is the first principal component (the eigenvalue that indicates the largest value) is equal to or greater than the set threshold for the eigenvalue, it can be determined that extreme color space conversion is not performed. judge. This determination result is held as the first determination result.

  Next, the distance between the input color space conversion matrix and the reference matrix is calculated, and the obtained distance is compared with a threshold related to the distance between vectors. If the distance is equal to or less than the threshold, it is determined that the distance is close to the reference matrix. It is determined that color space conversion is performed. This determination result is held as the second determination result. If both the first and second determination results are satisfied at the same time, it is determined that color space conversion is to be performed, and if not, the color space conversion is determined not to be performed. Here, when it is not satisfied with the result of the first determination, the second determination may be skipped. Moreover, you may replace the definition of a 1st determination and a 2nd determination. Based on the determination, information indicating whether or not to perform color space conversion is defined as color space conversion execution information, and is output to the color space conversion execution unit 703.

  The eigenvalue storage unit 704 inputs the eigenvalues calculated in the principal component analysis process for calculating the color space conversion matrix from the color space conversion matrix calculation unit 116 and stores them. The reference matrix setting unit 705 inputs a reference matrix predetermined for each format to the color space conversion execution determination unit 706 in accordance with the format of the input image (RGB format or YUV format). The threshold setting unit 707 inputs and sets the threshold for the eigenvalue and the threshold for the distance between vectors from a predetermined table, and outputs the threshold information to the color space conversion execution determination unit 706.

  The color space conversion execution unit 703 includes a prediction residual signal output from the prediction residual signal storage unit 701, a color space conversion matrix output from the color space conversion matrix storage unit 702, and a color space conversion execution determination unit 706. Color space conversion is performed using the output color space conversion execution information. If the color space conversion execution information indicates that color space conversion is to be performed, color space conversion is performed. If the color space conversion execution information indicates that the color space conversion is not performed, the color space conversion is not performed.

  As described above, the color space conversion processing unit 104 ″ includes the prediction residual signal storage unit 701, the color space conversion matrix storage unit 702, the color space conversion execution unit 703, the eigenvalue storage unit 704, the reference matrix setting unit 705, the color The space conversion execution determination unit 706 and the threshold setting unit 707 determine the magnitude relationship between the eigenvalue and the threshold value for the eigenvalue, and further, the adaptive color is determined by the distance between the color space conversion matrix and the reference matrix and the magnitude relationship of the threshold value with respect to the distance. A space conversion process is performed and coefficient values after color space conversion are output.

  Next, the processing operation of the color space conversion processing unit 104 '' shown in FIG. 12 will be described with reference to FIG. FIG. 13 is a flowchart illustrating the operation of the color space conversion process performed by the color space conversion processing unit 104 ″ illustrated in FIG. 12. First, the prediction residual signal storage unit 701 receives and stores the prediction residual signal of the block to be processed. Then, the data is output to the color space conversion execution unit 703 for color space conversion (step S41). Subsequently, the color space conversion matrix storage unit 702 inputs and stores the color space conversion matrix output from the color space conversion matrix calculation unit 116. Then, since it is necessary to execute the color space conversion, it is output to the color space conversion execution unit 703 (step S42).

  Next, the eigenvalue storage unit 704 inputs and stores eigenvalues obtained when the color space conversion matrix calculation unit 116 calculates the color space conversion matrix. In order to determine execution of the color space conversion, the eigenvalue information is output to the color space conversion execution determination unit 706. Further, the reference matrix setting unit 705 reads the reference matrix from the table according to the format of the input image, and outputs the information to the color space conversion execution determination unit 706. Similarly, the threshold setting unit 707 inputs color space conversion execution determination by inputting two types of threshold information for determining execution of color space conversion, that is, a threshold for an eigenvalue and a threshold for an inter-vector distance from a predetermined table. It outputs to the part 706 (step S43).

  Next, the color space conversion execution determination unit 706 includes a color space conversion matrix output from the color space conversion matrix storage unit 702, an eigenvalue output from the eigenvalue storage unit 704, a reference matrix output from the reference matrix setting unit 705, The threshold value output by the threshold value setting unit 707 is input. Then, the input eigenvalue information is compared with the threshold value for the eigenvalue, and the result of the magnitude relationship is held as the first determination condition (step S44). Further, the distance between the input color space conversion matrix and the reference matrix is calculated, the distance is compared with a threshold value for the distance between vectors, and the result of the magnitude relationship is held as the second determination condition (step S45). The presence / absence of color space determination execution is determined from the first determination condition and the second determination condition, and the information is output to the color space conversion execution unit 703 as color space conversion execution information (step S46). For example, it is determined that the color space conversion is performed for a block that satisfies both the first determination condition and the second determination condition at the same time, and it is determined that the color space conversion is not performed when the block is not satisfied at the same time.

  Next, the color space conversion execution unit 703 includes a prediction residual signal output from the prediction residual signal storage unit 701, a color space conversion matrix output from the color space conversion matrix storage unit 702, and a color space conversion execution determination unit 706. The color space conversion execution information output by the user is input, the color space execution determination result is determined based on the color space conversion execution information (step S47), and if it is determined that the color space conversion is to be executed, the color space conversion is executed. (Step S48). On the other hand, if it is determined not to execute, the color space conversion is not executed and the prediction residual signal is output as it is.

  Next, the configurations of the inverse color space conversion processing unit 109 ″ and the inverse color space conversion processing unit 206 ″ according to the third embodiment will be described with reference to FIG. FIG. 14 is a block diagram illustrating a configuration example of the reverse color space conversion processing unit 109 ″ and the reverse color space conversion processing unit 206 ″ illustrated in FIG. 2 according to the third embodiment. The reverse color space conversion processing unit 109 ″ and the reverse color space conversion processing unit 206 ″ perform common processing and have the same configuration. Therefore, as the reverse color space conversion processing, the reverse color space conversion processing unit 109 is used. '' Will be described, and the description of the inverse color space conversion processing unit 206 ″ will be omitted.

  As illustrated in FIG. 14, the inverse color space conversion processing unit 109 ″ includes a decoded color space converted coefficient value storage unit 801, an inverse color space conversion matrix storage unit 802, an inverse color space conversion execution unit 803, and an eigenvalue storage unit 804. A reference matrix setting unit 805, an inverse color space conversion execution determination unit 806, and a threshold setting unit 807. In FIG. 14, the eigenvalue storage unit 804, the reference matrix setting unit 805, the inverse color space conversion execution determination unit 806, and the threshold setting unit 807 are characteristic configurations.

  The decoded color space converted coefficient value storage unit 801 receives and stores the decoded color space converted coefficient value of the block to be processed. The inverse color space conversion matrix storage unit 802 receives and stores an inverse color space conversion matrix obtained from the decoded signal. The eigenvalue storage unit 804 inputs and stores eigenvalues calculated in the process of principal component analysis for calculating the inverse color space conversion matrix from the color space conversion matrix calculation unit 116.

  The reference matrix setting unit 805 reads a reference matrix necessary for performing the reverse color space conversion execution determination from a predetermined table and outputs the matrix to the reverse color space conversion execution determination unit 806. The table is a table that can be switched according to the format of the input image. For example, in the case of the YUV format, the reference matrix shown in FIG. 8 is output. In the case of the RGB format, an RGB-YUV conversion matrix is used as a reference matrix. In the case of the RGB format, in addition to the matrix, for example, the color space conversion matrix obtained for the entire sequence of each image, the average value of the color space conversion matrix obtained for each frame, or the color space obtained for the previous slice / frame / block A conversion matrix may be used.

  The inverse color space conversion execution determination unit 806 includes an inverse color space conversion matrix read from the inverse color space conversion matrix storage unit 802, an eigenvalue read from the eigenvalue storage unit 804, and a reference matrix set from the reference matrix setting unit 805. Using the two types of threshold values set from the threshold value setting unit 807, the presence / absence of reverse color space conversion is determined. For example, if the eigenvalue that is the first principal component (the eigenvalue that indicates the largest value) is equal to or greater than the set threshold for the eigenvalue, it can be determined that extreme color space conversion is not performed. judge. This determination result is held as the first determination result. Subsequently, the distance between the input color space conversion matrix and the reference matrix is calculated, and the thresholds related to the distance between vectors are compared. If the distance is equal to or less than the threshold, it is determined that the color space conversion is performed by determining that the distance is close to the reference matrix. To do. This determination result is held as the second determination result. If both the first and second determination results are satisfied at the same time, it is determined that color space conversion is to be performed, and if not, the color space conversion is determined not to be performed. Here, when it is not satisfied with the result of the first determination, the second determination may be skipped. Moreover, you may replace the definition of a 1st determination and a 2nd determination. Based on the determination, information indicating whether to perform reverse color space conversion is defined as reverse color space conversion execution information, and is output to the reverse color space conversion execution unit 803.

  The threshold setting unit 807 inputs and sets a threshold related to eigenvalues and a threshold related to the distance between vectors from a predetermined table, and outputs the threshold information to the inverse color space conversion execution determination unit 806. The inverse color space conversion execution unit 803 includes the decoded color space converted coefficient value output from the decoded color space converted coefficient value storage unit 801 and the inverse color space conversion output from the inverse color space conversion matrix storage unit 802. Reverse color space conversion is performed using the matrix and reverse color space conversion execution information output from the reverse color space conversion execution determination unit 806. If the reverse color space conversion execution information indicates that reverse color space conversion is to be performed, reverse color space conversion is performed. If the reverse color space conversion execution information indicates that reverse color space conversion is not performed, reverse color space conversion is not performed.

  As described above, the inverse color space conversion processing unit 109 ″ includes the decoded color space post-conversion coefficient value storage unit 801, the inverse color space conversion matrix storage unit 802, the inverse color space conversion execution unit 803, the eigenvalue storage unit 804, the reference The matrix setting unit 805, the inverse color space conversion execution determination unit 806, and the threshold setting unit 807 determine the magnitude relationship between the eigenvalue and the threshold for the eigenvalue, and further determine the distance between the color space conversion matrix and the reference matrix and the threshold for the distance. An adaptive color space conversion process is performed according to the magnitude relationship, and a coefficient value after color space conversion is output.

  Next, the processing operation of the inverse color space conversion processing unit 109 ″ shown in FIG. 14 will be described with reference to FIG. FIG. 15 is a flowchart illustrating the operation of the reverse color space conversion process performed by the reverse color space conversion processing unit 109 ″ illustrated in FIG. 14. The decoded color space converted coefficient value storage unit 801 receives and stores the decoded color space converted coefficient value of the block to be processed. Then, in order to perform inverse color space conversion, the coefficient value after the decoded color space conversion is output to the inverse color space conversion execution unit 803 (step S51). Subsequently, the reverse color space conversion matrix storage unit 802 inputs and stores the reverse color space conversion matrix output from the color space conversion matrix calculation unit 116. Since it is necessary for executing the inverse color space conversion, it is output to the inverse color space conversion execution unit 803 (step S52).

  Next, the eigenvalue storage unit 804 inputs and stores eigenvalues obtained when the color space conversion matrix calculation unit 116 calculates the color space conversion matrix. In order to determine execution of color space conversion, eigenvalue information is output to the inverse color space conversion execution determination unit 806. Further, the reference matrix setting unit 805 reads the reference matrix from the table according to the format of the input image, and outputs the information to the inverse color space conversion execution determination unit 806. Similarly, the threshold setting unit 807 inputs two types of threshold information for determining execution of the inverse color space conversion, that is, the threshold for the eigenvalue and the threshold for the inter-vector distance from the predetermined table, respectively, and performs the inverse color space conversion. It outputs to the execution determination part 806 (step S53).

  Next, the inverse color space conversion execution determination unit 806 is the inverse color space conversion matrix output from the inverse color space conversion matrix storage unit 802, the eigenvalue output from the eigenvalue storage unit 804, and the reference output from the reference matrix setting unit 805. The matrix and the threshold value output by the threshold value setting unit 807 are input. The input eigenvalue information is compared with the threshold value for the eigenvalue, and the result of the magnitude relationship is held as the first determination condition (step S54). Further, the distance between the input inverse color space conversion matrix and the reference matrix is calculated, the distance is compared with a threshold for the distance between vectors, and the result of the magnitude relationship is held as a second determination condition (step S55). Based on the first determination condition and the second determination condition, it is determined whether or not reverse color space determination is performed, and the information is output to the reverse color space conversion execution unit 803 as reverse color space conversion execution information (step S56). For example, it is determined that reverse color space conversion is performed for a block that satisfies both the first determination condition and the second determination condition at the same time, and it is determined that reverse color space conversion is not performed when the block does not satisfy both simultaneously.

  Next, the inverse color space conversion execution unit 803 outputs the decoded color space converted coefficient value output from the decoded color space converted coefficient value storage unit 801 and the inverse color space output from the inverse color space conversion matrix storage unit 802. Using the conversion matrix and the reverse color space conversion execution information output by the reverse color space conversion execution determination unit 806 as input, the reverse color space execution determination is performed based on the reverse color space conversion execution information (step S57). If it is determined that conversion is to be performed, reverse color space conversion is performed (step S58). On the other hand, if it is determined not to execute, the inverse color space conversion is not executed, and the prediction residual decoded signal is output as it is.

<Fourth Embodiment>
Next, an image encoding device and an image decoding device according to a fourth embodiment of the present invention will be described. The fourth embodiment differs from the configuration of the color space conversion processing unit 104 shown in FIG. 3 only in the inverse color space conversion processing units 109 and 206 shown in FIG. Therefore, only the configuration of the color space conversion processing unit 104 ′ ″ and the inverse color space conversion processing unit 109 ′ ″ in the fourth embodiment will be described. Since the reverse color space conversion processing unit 209 ′ ″ is common to the reverse color space conversion processing unit 109 ′ ″, description thereof is omitted.

  FIG. 16 is a block diagram illustrating a configuration of the color space conversion processing unit 104 ″ ″ according to the fourth embodiment. As shown in FIG. 16, the color space conversion processing unit 104 ′ ″ includes a prediction residual signal storage unit 901, a color space conversion matrix storage unit 902, a color space conversion execution unit 903, an eigenvalue storage unit 904, and an eigenvalue weighted sum calculation. A unit 905, a color space conversion execution determination unit 906, and a threshold setting unit 907. The prediction residual signal storage unit 901, the color space conversion matrix storage unit 902, and the color space conversion execution unit 903 are respectively the prediction residual signal storage unit 301, the color space conversion matrix storage unit 302, and the color space conversion execution unit shown in FIG. This is the same configuration as 303.

  The eigenvalue storage unit 904 receives a plurality of eigenvalues (for example, the first principal component and the second principal component) calculated in the process of principal component analysis for calculating the color space conversion matrix from the color space conversion matrix calculation unit 116. Enter and remember.

The eigenvalue weighted sum calculation unit 905 calculates a weighted sum from a plurality of eigenvalues input from the eigenvalue storage unit 904. For example, the eigenvalue of the first principal component is P1, and the eigenvalue of the second principal component is P2. At this time, the weighted sum is P, and P = α × P1 + β × P2 (1)
The weighted sum is calculated as follows. Here, α and β are coefficients, and if β = 0, this corresponds to the determination performed with one kind of eigenvalue, and therefore coincides with the encoding result of the first embodiment. The weighted sum P is output to the color space conversion execution determination unit 906. α and β can be read from a predetermined table, and fixed values can be used during the encoding process. For example, α and β can be changed in units of slices or frames. When changing in units, it is necessary to encode and transmit to the decoding side in the same manner as the threshold value.

  The color space conversion execution determination unit 906 determines whether or not to perform color space conversion using the weighted eigenvalue obtained from the eigenvalue weighted sum calculation unit 905 and the threshold set from the threshold setting unit 907. If the weighted eigenvalue is equal to or greater than the set threshold value, it can be determined that extreme color space conversion is not performed, and therefore it is determined that color space conversion is to be performed. If it is smaller than the threshold value, it is determined that the color space conversion is not performed. Information regarding whether or not to perform these operations is defined as color space conversion execution information, and is output to the color space conversion execution unit 903.

  The threshold setting unit 907 reads a threshold from a predetermined table or the like, and sets the threshold in the color space conversion execution determination unit 906. The threshold value is in the range of 0 to 1 from the range that the eigenvalue can take, but is expressed as an integer value of 0 to 4096, for example, depending on the internal structure (0 indicates the eigenvalue 0, and 4096 is In the case where eigenvalue 1 is indicated), a value corresponding thereto may be set. If the threshold value is set to 0.5, the threshold value to be set may be set to 2048. A threshold conversion processing unit that performs such replacement may be provided between the threshold setting unit 907 and the color space conversion execution determination unit 906.

  The color space conversion execution unit 903 receives the prediction residual signal output from the prediction residual signal storage unit 901, the color space conversion matrix output from the color space conversion matrix storage unit 902, and the color space conversion execution determination unit 906. Color space conversion is performed using the output color space conversion execution information. If the color space conversion execution information indicates that color space conversion is to be performed, color space conversion is performed. If the color space conversion execution information indicates that the color space conversion is not performed, the color space conversion is not performed.

  As described above, the color space conversion processing unit 104 ′ ″ includes the prediction residual signal storage unit 901, the color space conversion matrix storage unit 902, the color space conversion execution unit 903, the eigenvalue storage unit 904, and the eigenvalue weighted sum calculation unit. 905, the color space conversion execution determination unit 906 and the threshold setting unit 907 perform adaptive color space conversion processing by performing weighted eigenvalue determination and threshold size determination, and output coefficient values after color space conversion To do.

  Next, the processing operation of the color space conversion processing unit 104 ″ ″ shown in FIG. 16 will be described with reference to FIG. FIG. 17 is a flowchart showing the operation of the color space conversion process performed by the color space conversion processing unit 104 ″ ″ shown in FIG. First, the prediction residual signal storage unit 901 receives the prediction residual signal of the block to be processed from the prediction residual signal generation unit 103 and stores it. Then, the data is output to the color space conversion execution unit 903 to perform color space conversion (step S61). Subsequently, the color space conversion matrix storage unit 902 inputs and stores the color space conversion matrix output from the color space conversion matrix calculation unit 116. Then, since it is necessary for executing the color space conversion, it is output to the color space conversion executing unit 903 (step S62).

  Next, the eigenvalue storage unit 904 inputs and stores a plurality of eigenvalues (for example, the first principal component and the second principal component) obtained when the color space conversion matrix calculation unit 116 calculates the color space conversion matrix. . Then, the eigenvalue information of each component is output to the eigenvalue weighted sum calculation unit 905 in order to determine execution of color space conversion. Further, the threshold setting unit 906 inputs threshold information for determining execution of color space conversion from a predetermined table and outputs the threshold information to the color space conversion execution determination unit 906 (step S63). From the eigenvalues of each component input to the eigenvalue weighted sum calculation unit 905, a weighted sum as shown in Expression (1) is calculated, and the weighted sum of eigenvalues is output to the color space conversion execution determination unit 906. (Step S64). Next, the color space conversion execution determination unit 906 receives the weighted eigenvalue obtained from the eigenvalue weighted sum calculation unit 905 and the threshold set by the threshold setting unit 907 as input, and based on the magnitude relationship between these values, The presence / absence of space determination execution is determined, and information on the determination result is output to the color space conversion execution unit 903 as color space conversion execution information (step S65).

  Next, the color space conversion execution unit 903 has a prediction residual signal read from the prediction residual signal storage unit 901, a color space conversion matrix read from the color space conversion matrix storage unit 902, and a color space conversion execution determination unit 906. The color space conversion execution information output by the user is input, and the color space execution determination result is determined based on the color space conversion execution information (step S66). If it is determined that the color space conversion is to be executed as a result of the determination, the color space conversion is executed (step S67). On the other hand, if it is determined not to execute, the color space conversion is not executed and the prediction residual signal is output as it is.

  Next, the configuration of the inverse color space conversion processing unit 109 ″ ″ and the inverse color space conversion processing unit 206 ″ ″ according to the fourth embodiment will be described with reference to FIG. FIG. 18 is a block diagram illustrating a configuration example of the inverse color space conversion processing unit 109 ″ ″ according to the fourth embodiment and the inverse color space conversion processing unit 206 ″ ″ illustrated in FIG. 2. Since the reverse color space conversion processing unit 109 ′ ″ and the reverse color space conversion processing unit 206 ′ ″ perform the same processing and have the same configuration, the reverse color space conversion processing includes reverse color space conversion processing. The unit 109 ′ ″ will be described, and the description of the inverse color space conversion processing unit 206 ′ ″ will be omitted.

  As shown in FIG. 18, the inverse color space conversion processing unit 109 ′ ″ includes a decoded color space converted coefficient value storage unit 1001, an inverse color space conversion matrix storage unit 1002, an inverse color space conversion execution unit 1003, and an eigenvalue storage unit. 1004, an eigenvalue weighted sum calculation unit 1005, an inverse color space conversion execution determination unit 1006, and a threshold setting unit 1007. Here, in FIG. 18, an eigenvalue storage unit 1004, an eigenvalue weighted sum calculation unit 1005, an inverse color space conversion execution determination unit 1006, and a threshold setting unit 1007 are characteristic configurations.

  The decoded color space converted coefficient value storage unit 1001 inputs and stores the decoded color space converted coefficient value of the block to be processed. The inverse color space conversion matrix storage unit 1002 receives and stores an inverse color space conversion matrix obtained from the decoded signal. The eigenvalue storage unit 1004 stores a plurality of eigenvalues (for example, the first principal component and the second principal component) calculated in the process of principal component analysis for calculating the inverse color space conversion matrix, and the color space conversion matrix calculation unit 116. Input and memorize more.

The eigenvalue weighted sum calculation unit 1005 calculates a weighted sum from a plurality of eigenvalues input from the eigenvalue storage unit 1004. For example, the eigenvalue of the first principal component is P1, and the eigenvalue of the second principal component is P2. At this time, the weighted sum is P, and P = α × P1 + β × P2 (2)
The weighted sum is calculated as follows. Here, α and β are coefficients, and if β = 0, this corresponds to the determination performed with one kind of eigenvalue, and therefore coincides with the encoding result of the first embodiment. The weighted sum P is output to the inverse color space conversion execution determination unit 1006. α and β may be read from a predetermined table, and fixed values during decoding processing may be used. For example, α and β may be changed in units of slices or frames. When changing in units, it is necessary to decode and use α and β encoded on the encoding side from the bit stream.

  The inverse color space conversion execution determination unit 1006 determines whether to perform reverse color space conversion using the weighted eigenvalue read from the eigenvalue weighted sum calculation unit 1005 and the threshold set from the threshold setting unit 1007. If the weighted eigenvalue is equal to or greater than the set threshold value, it can be determined that extreme inverse color space conversion is not performed, and therefore it is determined that inverse color space conversion is to be performed. If it is smaller than the threshold value, it is determined that reverse color space conversion is not performed. Information indicating whether or not to perform these operations is defined as reverse color space conversion execution information, and is output to the color space conversion execution unit 1003.

  The inverse color space conversion execution unit 1003 outputs the decoded color space converted coefficient value output from the decoded color space converted coefficient value storage unit 1001 and the inverse color space conversion output from the inverse color space conversion matrix storage unit 1002. Reverse color space conversion is performed using the matrix and reverse color space conversion execution information output from the reverse color space conversion execution determination unit 1006. If the reverse color space conversion execution information indicates that reverse color space conversion is to be performed, reverse color space conversion is performed. If the reverse color space conversion execution information indicates that reverse color space conversion is not performed, reverse color space conversion is not performed.

  The threshold setting unit 1007 reads the threshold from a predetermined table or the like, and sets the threshold for the inverse color space conversion execution determination unit 1006. The threshold value is in the range of 0 to 1 from the range that the eigenvalue can take, but is expressed as an integer value of 0 to 4096, for example, depending on the internal structure (0 indicates the eigenvalue 0, and 4096 is In the case where eigenvalue 1 is indicated), a value corresponding thereto may be set. For example, when the threshold value is 0.5, the threshold value to be set may be 2048.

  As described above, the inverse color space conversion processing unit 109 ′ ″ includes the decoded color space post-conversion coefficient value storage unit 1001, the inverse color space conversion matrix storage unit 1002, the inverse color space conversion execution unit 1003, the eigenvalue storage unit 1004, An eigenvalue weighted sum calculation unit 1005, an inverse color space conversion execution determination unit 1006, and a threshold value setting unit 1007 perform adaptive inverse color space conversion processing by determining the magnitude of the weighted eigenvalue and threshold value, A prediction residual decoding signal is output.

  Next, the processing operation of the inverse color space conversion processing unit 109 '' 'shown in FIG. 18 will be described with reference to FIG. FIG. 19 is a flowchart showing the operation of the reverse color space conversion process performed by the reverse color space conversion processing unit 109 ″ ″ shown in FIG. The decoded color space converted coefficient value storage unit 1001 inputs and stores the decoded color space converted coefficient value of the block to be processed. Then, in order to perform inverse color space conversion, the coefficient value after the decoded color space conversion is output to the inverse color space conversion execution unit 1003 (step S71).

  Next, the reverse color space conversion matrix storage unit 1002 inputs and stores the reverse color space conversion matrix output from the color space conversion matrix calculation unit 116. Then, since it is necessary for executing the reverse color space conversion, it is output to the reverse color space conversion execution unit 1003 (step S72). Next, the eigenvalue storage unit 1004 inputs and stores a plurality of eigenvalues (for example, the first principal component and the second principal component) obtained when the color space conversion matrix calculation unit 116 calculates the inverse color space conversion matrix. To do. Further, the threshold setting unit 1007 inputs threshold information for determining execution of reverse color space conversion from a predetermined table and outputs the threshold information to the reverse color space conversion execution determination unit 1006 (step S73). Next, a weighted sum as shown in Equation (2) is calculated from the eigenvalues of each component input to the eigenvalue weighted sum calculation unit 1005, and the weighted sum of eigenvalues is determined to be the inverse color space conversion execution determination unit 1006. (Step S74).

  Next, the inverse color space conversion execution determination unit 1005 receives the weighted eigenvalue output from the eigenvalue weighted sum calculation unit 1005 and the threshold output from the threshold setting unit 1007 as input, and based on the magnitude relationship between these values. The presence / absence of reverse color space determination execution is determined, and information on the determination result is output to the reverse color space conversion execution unit 1003 as reverse color space conversion execution information (step S75).

  Next, the inverse color space conversion execution unit 1003 outputs the decoded color space converted coefficient value output from the decoded color space converted coefficient value storage unit 1001 and the inverse color space output from the inverse color space conversion matrix storage unit 1002. Using the conversion matrix and the reverse color space conversion execution information output by the reverse color space conversion execution determination unit 1006 as input, the reverse color space execution determination result is determined based on the reverse color space conversion execution information (step S76). If it is determined that color space conversion is to be performed, reverse color space conversion is performed (step S77). If it is determined not to execute, the inverse color space conversion is not executed and the prediction residual decoded signal is output as it is.

  As described above, the color space conversion processing unit 104 ′ ″ and the inverse color space conversion processing unit 109 ′ ″ use the weighted sum of eigenvalues obtained when the color space conversion matrix is generated, It is possible to perform adaptive color space conversion processing on a block basis by determining whether or not to perform conversion. Depending on the threshold value to be set, it is possible to greatly omit the processing related to color space conversion, and an increase in speed can be expected. In addition, unnecessary conversion that causes deterioration in encoding efficiency can be omitted, which can contribute to improvement in encoding efficiency.

  As described above, the color space conversion processing unit 104 described in the first, second, third, and fourth embodiments, the image encoding device 100 including the reverse color space conversion processing unit 109, and the reverse color space conversion processing unit 206. The image decoding apparatus 200 including the above can be realized using a computer and a software program.

  FIG. 20 is a diagram illustrating a hardware configuration example of a computer when the image encoding device 100 including the color space conversion processing unit 104 is configured by a computer and a software program. The computer includes a CPU 1100, a memory 1101, a video signal input unit 1102, a program storage device 1103, and an encoded stream output unit 1105.

  The CPU 1100 executes a program. The memory 1101 is a RAM that temporarily stores programs and data accessed by the CPU 1100. The video signal input unit 1102 inputs an input video signal to be encoded from a device such as a camera that generates a video signal. The video signal input unit 1102 may be a storage device that stores an input video signal from a hard disk device or the like. The program storage device 1103 stores an image encoding program 1104 that is a software program that causes the CPU 1100 to execute the encoding process described in the above embodiments. The encoded stream output unit 1105 outputs an encoded stream generated when the CPU 1100 executes the image encoding program 1104 loaded in the memory 1101. The encoded stream output unit 1105 may output an encoded stream via a network. The encoded stream output unit 1105 may be a storage device that stores an encoded stream such as a hard disk device. The CPU 1100, the memory 1101, the video signal input unit 1102, the program storage device 1103, and the encoded stream output unit 1105 are connected to each other via a bus.

  FIG. 21 is a diagram illustrating a hardware configuration example of a computer when the image decoding apparatus 200 including the inverse color space conversion processing unit 109 and the inverse color space conversion processing unit 206 is configured by a computer and a software program. The computer includes a CPU 1200, a memory 1201, an encoded stream input unit 1202, a program storage device 1203, and a decoded video output unit 1205.

  CPU 1200 executes a program. The memory 1201 is a RAM or the like that temporarily stores programs and data accessed by the CPU 1200. The encoded stream input unit 1202 inputs, for example, an encoded stream obtained by encoding the input video signal by the image encoding apparatus 100 through the above-described processing. The encoded stream input unit 1202 may be a storage device that stores an encoded stream by a hard disk device or the like. The program storage device 1203 stores an image decoding program 1204 that is a software program that causes the CPU 1200 to execute the decoding processing described in each of the above-described embodiments. The decoded video output unit 1205 outputs the decoded video obtained by decoding the encoded stream to a playback device or the like by the CPU 1200 executing the image decoding program 1204 loaded in the memory 1201. Note that the decoded video output unit 1205 may be a storage device that stores a decoded video signal from a hard disk device or the like. The CPU 1200, the memory 1201, the encoded stream input unit 1202, the program storage device 1203, and the decoded video output unit 1205 are connected to each other via a bus.

  The functions of the color space conversion processing unit and the inverse color space conversion processing unit in the above-described embodiment may be realized by a computer. In that case, a program for realizing this function may be recorded on a computer-readable recording medium, and the program recorded on this recording medium may be read into a computer system and executed. Here, the “computer system” includes an OS and hardware such as peripheral devices. The “computer-readable recording medium” refers to a storage device such as a flexible medium, a magneto-optical disk, a portable medium such as a ROM and a CD-ROM, and a hard disk incorporated in a computer system. Furthermore, the “computer-readable recording medium” dynamically holds a program for a short time like a communication line when transmitting a program via a network such as the Internet or a communication line such as a telephone line. In this case, a volatile memory inside a computer system serving as a server or a client in that case may be included and a program held for a certain period of time. Further, the program may be for realizing a part of the functions described above, and may be a program capable of realizing the functions described above in combination with a program already recorded in the computer system. It may be realized using hardware such as PLD (Programmable Logic Device) or FPGA (Field Programmable Gate Array).

  As mentioned above, although embodiment of this invention has been described with reference to drawings, the said embodiment is only the illustration of this invention, and it is clear that this invention is not limited to the said embodiment. is there. Therefore, additions, omissions, substitutions, and other modifications of the components may be made without departing from the technical idea and scope of the present invention.

  In color space conversion processing, it is determined whether to perform color space conversion from eigenvalue information, and the coding space is improved by performing color space conversion adaptively in units of blocks, and encoding processing and decoding with color space conversion. It can be applied to applications where speeding up of processing is essential.

  DESCRIPTION OF SYMBOLS 100 ... Image coding apparatus 101 ... Intra prediction processing part 102 ... Inter prediction processing part 103 ... Prediction residual signal generation part 104 ... Color space conversion processing part 105 ... Orthogonal transformation processing part 106 ... Quantization processing part , 107 ... Inverse quantization processing unit, 108 ... Inverse orthogonal transformation processing unit, 109 ... Inverse color space transformation processing unit, 110 ... Decoded signal generation unit, 111 ... Frame memory, 112 ... In-loop filter processing unit, 113 ... Frame buffer , 114 ... inter prediction information storage section, 115 ... intra prediction information storage section, 116 ... color space conversion matrix calculation section, 117 ... entropy encoding processing section, 200 ... image decoding apparatus, 201 ... entropy decoding processing section, 202 ... intra Prediction processing unit, 203 ... inter prediction processing unit, 204 ... inverse quantization processing unit, 205 ... inverse orthogonal transform processing unit, 206 ... inverse Spatial conversion processing unit, 207 ... decoded signal generation unit, 208 ... frame memory, 209 ... in-loop filter processing unit, 210 ... frame buffer, 211 ... inter prediction information storage unit, 212 ... intra prediction information storage unit, 213 ... color space Conversion matrix calculation unit, 301, 501, 701, 901 ... Prediction residual signal storage unit, 302, 502, 702, 902 ... Color space conversion matrix storage unit, 303, 503, 703, 903 ... Color space conversion execution unit, 304 404, 704, 804, 904004 ... Eigen value storage unit, 504, 604, 705, 805 ... Reference matrix setting unit, 305, 505, 706, 906 ... Color space conversion execution determination unit, 306, 406, 506, 606, 707 , 807, 907, 1007... Threshold setting unit, 401, 601, 801, 1001 Decoded color space converted coefficient value storage unit, 402, 602, 802, 1002 ... Inverse color space conversion matrix storage unit, 403, 603, 803, 1003 ... Inverse color space conversion execution unit, 405, 605, 806, 1006 ... Inverse Color space conversion execution determination unit, 905, 1005 ... Eigen value weighted sum calculation unit, 1100, 1200 ... CPU, 1101, 1201 ... Memory, 1102 ... Video signal input unit, 1103, 1203 ... Program storage device, 1104 ... Image encoding Program, 1105 ... encoded stream output unit, 1202 ... encoded stream input unit, 1204 ... image decoding program, 1205 ... decoded video output unit

Claims (28)

An image encoding method performed by an image encoding device that encodes an image by applying color space conversion,
A matrix calculation step of calculating a color space conversion matrix for performing the color space conversion from the decoded reference signal;
An eigenvalue acquisition step of acquiring an eigenvalue obtained in the process of calculating the color space conversion matrix;
A determination step of determining whether to perform the color space conversion based on a magnitude relationship between a predetermined threshold and the eigenvalue;
And a color space conversion execution step of executing the color space conversion based on the calculation result of the color space conversion matrix according to the determination result. In the eigenvalue acquisition step,
Get the eigenvalues of multiple components,
In the determination step,
The weighted sum of the plurality of eigenvalues is calculated, and it is determined whether or not to perform the color space conversion based on a magnitude relationship between the weighted sum and a predetermined threshold. 2. The image encoding method according to 1. In the eigenvalue acquisition step,
Get the eigenvalues of multiple components,
In the determination step,
2. The image coding according to claim 1, wherein whether or not to perform the color space conversion is determined based on a magnitude relationship between an eigenvalue corresponding to each component and a predetermined threshold corresponding to the eigenvalue. Method. An image encoding method performed by an image encoding device that encodes an image by applying color space conversion,
A matrix calculation step of calculating a color space conversion matrix for performing the color space conversion from the decoded reference signal;
A distance calculating step of calculating a distance between the color space conversion matrix and a predetermined reference matrix;
A determination step of determining whether or not to perform the color space conversion based on a magnitude relationship between a predetermined threshold and the distance;
And a color space conversion execution step of executing the color space conversion based on the calculation result of the color space conversion matrix according to the determination result. An image encoding method performed by an image encoding device that encodes an image by applying color space conversion,
A matrix calculation step of calculating a color space conversion matrix for performing the color space conversion from the decoded reference signal;
An eigenvalue acquisition step of acquiring an eigenvalue obtained in the process of calculating the color space conversion matrix;
A distance calculating step of calculating a distance between the color space conversion matrix and a predetermined reference matrix;
A determination step of determining whether to perform the color space conversion using the eigenvalue, the distance, and a predetermined threshold;
And a color space conversion execution step of executing the color space conversion based on the calculation result of the color space conversion matrix according to the determination result.   The image encoding method according to claim 1, wherein a predetermined threshold is encoded and transmitted. An image decoding method performed by an image decoding apparatus that decodes encoded data by applying inverse color space conversion,
A matrix calculation step of calculating an inverse color space conversion matrix for performing the inverse color space conversion from the decoded reference signal;
An eigenvalue acquisition step of acquiring an eigenvalue obtained in the process of calculating the inverse color space conversion matrix;
A determination step of determining whether or not to perform the inverse color space conversion based on a magnitude relationship between a predetermined threshold and the eigenvalue;
An image decoding method comprising: an inverse color space conversion execution step of executing the inverse color space conversion based on the calculation result of the inverse color space conversion matrix according to the determination result. In the eigenvalue acquisition step,
Get the eigenvalues of multiple components,
In the determination step,
The weighted sum of the plurality of eigenvalues is calculated, and it is determined whether or not to perform the inverse color space conversion based on a magnitude relationship between the weighted sum and a predetermined threshold. Item 8. The image decoding method according to Item 7. In the eigenvalue acquisition step,
Get the eigenvalues of multiple components,
In the determination step,
8. The image decoding according to claim 7, wherein whether or not to perform the inverse color space conversion is determined based on a magnitude relationship between an eigenvalue corresponding to each component and a predetermined threshold corresponding to the eigenvalue. Method. An image decoding method performed by an image decoding apparatus that decodes encoded data by applying inverse color space conversion,
A matrix calculation step of calculating an inverse color space conversion matrix for performing the inverse color space conversion from the decoded reference signal;
A distance calculating step of calculating a distance between the inverse color space conversion matrix and a predetermined reference matrix;
A determination step of determining whether to perform the inverse color space conversion based on a magnitude relationship between a predetermined threshold and the distance;
An image decoding method comprising: an inverse color space conversion execution step of executing the inverse color space conversion based on the calculation result of the inverse color space conversion matrix according to the determination result. An image decoding method performed by an image decoding apparatus that decodes encoded data by applying inverse color space conversion,
A matrix calculation step of calculating an inverse color space conversion matrix for performing the inverse color space conversion from the decoded reference signal;
An eigenvalue acquisition step of acquiring an eigenvalue obtained in the process of calculating the inverse color space conversion matrix;
A distance calculating step of calculating a distance between the inverse color space conversion matrix and a predetermined reference matrix;
A determination step of determining whether to perform the inverse color space conversion using the eigenvalue, the distance, and a predetermined threshold;
An image decoding method comprising: an inverse color space conversion execution step of executing the inverse color space conversion based on the calculation result of the inverse color space conversion matrix according to the determination result.   The image decoding method according to any one of claims 7 to 11, wherein the predetermined threshold is set by decoding from a bitstream. An image encoding device that encodes an image by applying color space conversion,
Matrix calculation means for calculating a color space conversion matrix for performing the color space conversion from the decoded reference signal;
Eigenvalue acquisition means for acquiring eigenvalues obtained in the process of calculating the color space conversion matrix;
Determination means for determining whether or not to execute the color space conversion based on a magnitude relationship between a predetermined threshold and the eigenvalue;
An image coding apparatus comprising: color space conversion executing means for executing the color space conversion based on the calculation result of the color space conversion matrix according to the determination result. The eigenvalue acquisition means includes
Get the eigenvalues of multiple components,
The determination means includes
The weighted sum of the plurality of eigenvalues is calculated, and it is determined whether or not to perform the color space conversion based on a magnitude relationship between the weighted sum and a predetermined threshold. 14. The image encoding device according to 13. The eigenvalue acquisition means includes
Get the eigenvalues of multiple components,
The determination means includes
14. The image coding according to claim 13, wherein whether or not to perform the color space conversion is determined based on a magnitude relationship between an eigenvalue corresponding to each component and a predetermined threshold corresponding to the eigenvalue. apparatus. An image encoding device that encodes an image by applying color space conversion,
Matrix calculation means for calculating a color space conversion matrix for performing the color space conversion from the decoded reference signal;
A distance calculating means for calculating a distance between the color space conversion matrix and a predetermined reference matrix;
Determination means for determining whether or not to execute the color space conversion based on a magnitude relationship between a predetermined threshold and the distance;
An image coding apparatus comprising: color space conversion executing means for executing the color space conversion based on the calculation result of the color space conversion matrix according to the determination result. An image encoding device that encodes an image by applying color space conversion,
Matrix calculation means for calculating a color space conversion matrix for performing the color space conversion from the decoded reference signal;
Eigenvalue acquisition means for acquiring eigenvalues obtained in the process of calculating the color space conversion matrix;
A distance calculating means for calculating a distance between the color space conversion matrix and a predetermined reference matrix;
Determination means for determining whether to perform the color space conversion using the eigenvalue, the distance, and a predetermined threshold;
An image coding apparatus comprising: color space conversion executing means for executing the color space conversion based on the calculation result of the color space conversion matrix according to the determination result.   The image encoding device according to any one of claims 13 to 17, wherein a predetermined threshold is encoded and transmitted. An image decoding device that decodes encoded data by applying inverse color space conversion,
Matrix calculation means for calculating an inverse color space conversion matrix for performing the inverse color space conversion from the decoded reference signal;
Eigenvalue acquisition means for acquiring eigenvalues obtained in the process of calculating the inverse color space conversion matrix;
Determination means for determining whether to perform the inverse color space conversion based on a magnitude relationship between a predetermined threshold and the eigenvalue;
An image decoding apparatus comprising: an inverse color space conversion executing unit configured to execute the inverse color space conversion based on the calculation result of the inverse color space conversion matrix according to the determination result. The eigenvalue acquisition means includes
Get the eigenvalues of multiple components,
The determination means includes
The weighted sum of the plurality of eigenvalues is calculated, and it is determined whether or not to perform the inverse color space conversion based on a magnitude relationship between the weighted sum and a predetermined threshold. Item 20. The image decoding device according to Item 19. The eigenvalue acquisition means includes
Get the eigenvalues of multiple components,
The determination means includes
20. The image decoding according to claim 19, wherein whether or not to perform the inverse color space conversion is determined based on a magnitude relationship between an eigenvalue corresponding to each component and a predetermined threshold corresponding to the eigenvalue. apparatus. An image decoding device that decodes encoded data by applying inverse color space conversion,
Matrix calculation means for calculating an inverse color space conversion matrix for performing the inverse color space conversion from the decoded reference signal;
Distance calculating means for calculating a distance between the inverse color space conversion matrix and a predetermined reference matrix;
Determination means for determining whether or not to perform the inverse color space conversion based on a magnitude relationship between a predetermined threshold and the distance;
An image decoding apparatus comprising: an inverse color space conversion executing unit configured to execute the inverse color space conversion based on the calculation result of the inverse color space conversion matrix according to the determination result. An image decoding device that decodes encoded data by applying inverse color space conversion,
Matrix calculation means for calculating an inverse color space conversion matrix for performing the inverse color space conversion from the decoded reference signal;
Eigenvalue acquisition means for acquiring eigenvalues obtained in the process of calculating the inverse color space conversion matrix;
Distance calculating means for calculating a distance between the inverse color space conversion matrix and a predetermined reference matrix;
Determination means for determining whether to perform the inverse color space conversion using the eigenvalue, the distance, and a predetermined threshold;
An image decoding apparatus comprising: an inverse color space conversion executing unit configured to execute the inverse color space conversion based on the calculation result of the inverse color space conversion matrix according to the determination result.   The image decoding apparatus according to any one of claims 19 to 23, wherein the predetermined threshold is set by decoding from a bit stream.   An image encoding program for causing a computer to execute the image encoding method according to any one of claims 1 to 6.   An image decoding program for causing a computer to execute the image decoding method according to any one of claims 7 to 12.   A computer-readable recording medium on which the image encoding program according to claim 25 is recorded.   A computer-readable recording medium on which the image decoding program according to claim 26 is recorded.

Images