JP5753062B2 - Color component prediction type image encoding apparatus and decoding apparatus - Google Patents

Description

  The present invention relates to an inter-color component predictive image encoding apparatus and decoding apparatus, and more particularly to an efficient compression encoding for high-definition video and an inter-color component predictive image encoding apparatus for decoding the same. Relates to the device.

  Compression encoding systems for storage applications such as hard disk recorders and VTRs are likely to support a 4: 4: 4 color space represented by RGB and a 4: 2: 2 color space. As a conventional technique for improving the encoding efficiency in such a compression encoding system, there are the following.

  In Patent Document 1 below, for 4: 4: 4 video input, prediction reference between color components is performed prior to encoding, and only the residual signal is handled in the subsequent encoding process. To improve performance.

  Further, in Patent Document 2, an input image signal itself or a residual signal obtained by intra-frame prediction is determined as an application target of inter-color component prediction by analyzing an input image. Propose that.

  Patent Document 3 is an invention of the present application (unknown) by the applicant of the present application. However, the invention of the present application mainly applied the 4: 4: 4 video to the above-described existing technology, whereas the 4: 4: 4 video was applied. : In addition to 4 images, luminance signals such as 4: 2: 2 signals and 4: 2: 0 signals can be applied to independent color spaces.

JP 2006-310941 A JP 2010-239423 A Japanese Patent Application No. 2010-240721

  According to the above-described prior art, the encoding efficiency is improved by removing the correlation between the color components for all the component video formats such as 4: 4: 4 format, 4: 2: 2 format, and 4: 2: 0 format. You can expect. In the prior art, the reference signal is fixed to any one of the color components, and the encoding of the other color components (non-reference signal) is a method in which only the reference signal is predicted and referenced.

  By the way, in an actual encoding process, an encoding process and a decoding process of a certain frame are allowed to be sequentially processed in units of color components. Therefore, although it is possible in principle to refer to the encoding result of the color component signal processed with reference to the reference signal in the encoding processing of the remaining color component signals, this is actively used. Conventionally, no consideration has been given to further improving the encoding performance.

  The present invention has been made in view of the above-described problems of the prior art, and an object of the present invention is to provide an inter-component prediction type image encoding apparatus and decoding apparatus capable of improving the encoding performance over the conventional one. .

In order to achieve the above-described object, the present invention provides a video encoding apparatus capable of adaptively setting a reference signal for inter-color component prediction, and a reference signal and a first signal from an input image signal composed of a plurality of color components. A reference signal selecting means for selecting a non-reference signal and second and subsequent non-reference signals; and a first means for performing prediction with reference to only the reference signal in encoding of the first non-reference signal; The encoding of the second and subsequent non-reference signals includes second means for performing prediction with reference to the reference signal and the first non-reference signal, and the second means includes the color The signal to be subjected to the inter-component prediction is corrected using the inter-color component prediction residual signal of the first non-reference signal, and then the inter-color component prediction with reference to the reference signal is performed. First special There is.

According to the present invention, in the decoding device for the signal encoded by the moving image encoding device, a reference signal, a first non-reference signal, and second and subsequent non-reference signals are discriminated from the encoded signal. In the decoding of the first non-reference signal, a determination unit, a first unit that performs prediction with reference to only the reference signal, and in the decoding of the second and subsequent non-reference signals, the reference signal and the And second means for performing prediction with reference to the first non-reference signal, and in decoding of the second and subsequent non-reference signals, the residual signal of inter-color component prediction is used as the first non-reference signal. A second feature is that inter-color component prediction with reference to the reference signal is performed after reverse correction using the inter-color component prediction residual signal of the signal .

  According to the present invention, when encoding another color component (non-reference signal) under the condition that the reference signal is fixed to any one of the color components, the reference signal is not predicted and referenced. Refer to the encoding result of the color component signal (first non-reference signal) processed with reference to the reference signal in the encoding process of the remaining color component signals (second, third and subsequent non-reference signals). Make it possible.

  As a result, the prediction performance of the second and subsequent non-reference signals can be improved. As a result, the code amount of the prediction residual signal can be reduced, and the encoding performance can be improved.

  Also, when the number of color components is 4 or more, the third and subsequent non-reference signals are processed in the same manner as the second non-reference signal, thereby improving the prediction performance.

  According to the decoding device of the present invention, the image signal encoded as described above can be adaptively or efficiently decoded. As a result, an encoding / decoding device with improved encoding performance can be provided.

It is a block diagram which shows the structure of the outline of the encoding apparatus of 1st Embodiment of this invention. It is a flowchart which shows the operation example of the pre-processing analysis part of FIG. It is a block diagram which shows the structure of the outline of the decoding apparatus of 1st Embodiment. It is a block diagram which shows the schematic structure of the encoding apparatus of 2nd Embodiment of this invention. It is a flowchart which shows the operation example of the pre-processing analysis part of FIG. It is a block diagram which shows the structure of the outline of the decoding apparatus of 2nd Embodiment. It is a block diagram which shows the schematic structure of the encoding apparatus of 3rd Embodiment of this invention. It is a block diagram which shows the structure of the outline of the decoding apparatus of 3rd Embodiment.

  Hereinafter, the present invention will be described in detail with reference to the drawings. In video coding, as a method or method for performing prediction between color components, it can be divided into the following three types depending on what the color component prediction target is. That is, (1) when the inter-color component prediction target is an intra-frame prediction residual signal, (2) when the inter-color component prediction target is a frame image signal (no intra-frame prediction is used), (3) the color component This can be divided into cases where the target of inter prediction is a frame image signal (intra-frame prediction is used in combination).

  The (2) and (3) will be described later, and the embodiment (first embodiment) in the case of (1) will be described below. FIG. 1 is a block diagram showing a schematic configuration of the first embodiment.

  As shown in the figure, when the input image 11 is input, the intra prediction unit 12 applies intra prediction for each macroblock, and intra or intra frame (hereinafter referred to as intra) prediction residual signal a. Is output to the preprocessing analysis unit 13 and an intra predicted image b is created. The preprocessing analysis unit 13 processes the intra prediction residual signal a to determine a reference signal for inter-color component prediction, a first non-reference signal, and second and subsequent non-reference signals.

  As the reference signal for inter-color component prediction, for example, the color component having the highest resolution is selected according to the resolution of the color component that can be obtained by the preprocessing analysis unit 13. In the case of R (red), G (green), and B (blue) signals in 4: 4: 4 format, the G signal is used, and Y (luminance), Cb (color difference), Cr (in 4: 2: 2 format). In the case of a (color difference) signal, the Y signal is often a reference signal, and the remaining signals are often non-reference signals.

  An example of processing of which of the remaining signals is used as the first non-reference signal and the second non-reference signal will be described with reference to the flowchart of FIG.

  In step S1, an intra-frame prediction residual signal of the non-reference signal is acquired. These prediction residual signals are defined as Diff1 (i, j) and Diff2 (i, j), respectively. In step S2, an intraframe prediction residual signal of the reference signal is acquired. Let this prediction residual signal be Diff0 (i, j). Next, in step S3, cross-correlation coefficients of Diff0 (i, j) and Diff1 (i, j) and Diff0 (i, j) and Diff2 (i, j) are obtained. The cross-correlation coefficients are assumed to be Corr1 and Corr2, respectively. In step S4, the cross-correlation coefficients Corr1 and Corr2 are compared, and if Corr1> Corr2, the process proceeds to step S5, and the non-reference signal of the prediction residual signal Diff1 (i, j) is set as the first prediction residual signal Diff2. The non-reference signal of (i, j) is the second. On the other hand, if Corr1 <Corr2, the process proceeds to step S6, where the non-reference signal of the prediction residual signal Diff2 (i, j) is the first and the non-reference signal of the prediction residual signal Diff1 (i, j) is the second. . That is, the first non-reference signal having an intra-frame prediction residual having a large correlation with the intra-frame prediction residual of the reference signal is set as the first.

  Next, the reference signal selection unit 14 selects a color component of the reference signal and outputs the prediction residual signal c1. The prediction residual signal c1 of the reference signal is encoded by the DCT unit 15, the quantization unit 16, and the variable length encoding unit 17, and is output. Further, the output of the quantization unit 16 is locally decoded to obtain a reference residual signal 18. The reference residual signal 18 is sent to the inter-color component prediction units 21 and 32.

  Next, the reference signal selection unit 14 selects the first non-reference signal color component and outputs the prediction residual signal c2. The prediction residual signal c2 is input to the inter-color component prediction unit 21 and the subtraction unit 23. The inter-color component prediction unit 21 performs prediction by linear conversion with reference to the reference residual signal 18, outputs a prediction coefficient d2, and generates a residual prediction image 22. The subtracting unit 23 subtracts the residual prediction image 22 from the prediction residual signal c2 to obtain a color component prediction error component e2 of the prediction residual signal. The inter-color component prediction error component e2 is then sent to the DCT unit 24, the quantization unit 25, and the variable length coding unit 26, and is encoded together with the prediction coefficient d2 and output. Further, the output of the quantization unit 25 is locally decoded to obtain an inter-color component prediction error component 27.

  Next, the reference signal selection unit 14 selects the second non-reference signal color component and outputs the prediction residual signal c3. The subtractor 31 obtains a corrected residual signal d3 by subtracting the inter-color component prediction error component 27 from the predicted residual signal c3. The inter-color component prediction unit 32 performs prediction by linear conversion with reference to the reference residual signal 18, outputs a prediction coefficient e3, and generates a residual prediction image 33. Next, the subtracting unit 34 subtracts the corrected residual prediction image 33 from the corrected residual signal d3 to obtain a color component prediction error component f3 of the corrected residual signal d3. Subsequently, the inter-color component prediction error component f3 is sent to the DCT unit 35, the quantization unit 36, and the variable length encoding unit 37, and is encoded and output together with the prediction coefficient e3.

  Identification information such as a flag representing each signal is added to the encoded output of the reference signal and the first and second non-reference signals and output.

  According to the above encoding apparatus, since the second non-reference signal obtains the corrected residual signal d3 using the inter-color component prediction error component 27 of the first non-reference signal, the color component The prediction performance of the second non-reference signal can be improved compared to the case where the inter prediction error component 27 is not used.

  The above is the description of the configuration and operation of the encoding apparatus. Next, a decoding apparatus corresponding to the encoding apparatus will be described with reference to FIG. FIG. 3 is a block diagram showing a schematic configuration of the decoding apparatus.

  When the stream compressed by the encoding device, that is, the input stream 41 is input, the reference signal determination unit 42 uses the identification information added to the stream to generate the reference signal, the first and second non-references. Perform signal discrimination.

  The encoded data c4 of the reference signal is sent to the variable length decoding unit 43 and partially decoded, the intraframe prediction residual signal d4 is sent to the inverse quantization unit 44, and the intra prediction method e4 is sent to the intra prediction unit 48. Sent. The intra-frame prediction residual signal d4 is locally decoded by an inverse quantization unit 44 and an inverse DCT unit 45 to become a reference residual signal 46. Further, the intra prediction unit 48 generates an intra prediction image 49 using the intra prediction method e4, and the intra prediction image 49 is added to the reference residual signal 46 by the adding unit 47 to become a decoded image 50. The decoded image 50 is output and sent to the intra prediction unit 48. The reference residual signal 46 is also input to the inter-color component prediction units 55 and 75.

  Next, the variable length decoding unit 51 partially decodes the encoded data c5 of the first non-reference signal, and encodes the inter-coded color component prediction error component d5, the prediction coefficient e5, and the intra of the intraframe prediction residual signal. The prediction method f5 is output. The encoded inter-color component prediction error component d5 is decoded by the inverse quantization unit 52 and the inverse DCT unit 53 to become an inter-color component prediction error component 54. The inter-color component prediction error component 54 is sent to the adders 57 and 79. The inter-color component prediction unit 55 linearly transforms the reference residual signal 46 according to the prediction coefficient e5 to generate a residual prediction image 56. The adder 57 adds the residual prediction image 56 to the inter-color component prediction error component 54 to obtain a decoded residual image 58. Next, the intra prediction unit 59 performs an intra prediction process with reference to a neighboring decoded image based on the intra prediction method f5 obtained from the variable length decoding unit 51, and generates an intra predicted image 60. The adding unit 61 obtains a decoded image 62 by adding the intra predicted image 60 to the decoded residual image 58. The decoded image 62 is output and sent to the intra prediction unit 59.

  Next, the variable length decoding unit 71 partially decodes the encoded data c6 of the second non-reference signal, and encodes an inter-color prediction error component d6, a prediction coefficient e6, and an intra of the intraframe prediction residual signal. The prediction method f6 is output. The encoded color component prediction error component d6 is decoded by the inverse quantization unit 72 and the inverse DCT unit 73 to become an inter color component prediction error component 74. The inter-color component prediction error component 74 is sent to the adding unit 77. The inter-color component prediction unit 75 linearly transforms the reference residual signal 46 according to the prediction coefficient e6 to generate a pre-correction residual prediction image 76. This “before correction” means before correction by the first non-reference signal. The adding unit 77 adds the uncorrected residual prediction image 76 to the inter-color component prediction error component 74 to obtain a decoded uncorrected residual image 78. Next, the adding unit 79 obtains a decoded residual image 80 by adding the inter-color component prediction error component 54 of the first non-reference signal to the residual image 78 before decoding correction. Next, the intra prediction unit 81 performs an intra prediction process with reference to a neighboring decoded image based on the intra prediction method f6 obtained from the variable length decoding unit 71, and generates an intra predicted image 82. The adding unit 83 obtains a decoded image 84 by adding the intra predicted image 82 to the decoded residual signal 80. The decoded image 84 is output and sent to the intra prediction unit 81.

  As described above, the image signal encoded by the encoding device in FIG. 1 can be adaptively or efficiently decoded.

  In the encoding and decoding apparatus, the image signal of one reference signal and two non-reference signals has been described. However, the present embodiment is not limited to this, and the image signal has three or more non-reference signals. The same applies to the case. In addition, when the number of samples to be referred to and the prediction method are different, the inter-color component prediction may be applied after using the means of Patent Document 3. The same applies to the following embodiments.

  Next, a second embodiment in the case where the target of inter-color component prediction (2) is a frame image signal (not using intra-frame prediction) will be described with reference to FIGS. FIG. 4 is a block diagram of the encoding device, and FIG. 6 is a block diagram of the decoding device. In this embodiment, the “intraframe prediction residual signal” in the first embodiment may be read as “image signal”, and therefore only the differences from the first embodiment will be described. Moreover, in FIGS. 4-6, the function part respectively corresponding to FIGS. 1-3 shall be represented with a dash (').

  The preprocessing analysis unit 13 ′ processes the input image 11 ′ to determine a reference signal for inter-color component prediction, a first non-reference signal, and second and subsequent non-reference signals. As the reference signal for inter-color component prediction, for example, the color component having the highest resolution is selected according to the resolution of the color component that can be obtained by the preprocessing analyzer 13 ′.

  The reference signal selection unit 14 ′ determines the first non-reference signal and the second non-reference signal by the processing of steps S1 ′ to S6 ′ in FIG. The processing in FIG. 5 is simply replaced with “image signal” in “intra-frame prediction residual signal” in the processing in FIG.

  The image signal c1 ′ of the reference signal is sent to the intra prediction unit 12 ′ and subjected to intra prediction. The intra prediction residual signal d1 ′ output from the intra prediction unit 12 ′ is transmitted to the DCT unit 15 ′, the quantization unit 16 ′, and the variable length encoding unit 17 ′, and is encoded and output. The intra prediction image 19 obtained by the intra prediction unit 12 ′ is added by the addition unit 20 with the locally decoded prediction residual signal d1 ′ to obtain a reference image signal 18 ′. The reference image signal 18 ′ is sent to the inter-color component prediction units 21 ′ and 32 ′.

  The inter-color component prediction unit 21 ′ performs inter-color component prediction between the image signal c2 ′ of the first non-reference signal and the reference image signal 18 ′, and outputs a predicted image 22 ′ and a prediction coefficient d2 ′. The subtractor 23 ′ subtracts the predicted image 22 ′ from the first non-reference image signal c2 ′ to obtain an inter-color component prediction error component e2 ′ of the image signal. The prediction error component e2 ′ is then sent to the DCT unit 24 ′, the quantization unit 25 ′, and the variable length coding unit 26 ′, encoded together with the prediction coefficient d2 ′, and output. Further, the output of the quantization unit 25 ′ is locally decoded to obtain an inter-color component prediction error component 27 ′.

  Next, the image signal c3 ′ of the second non-reference signal is subtracted from the inter-color component prediction error component 27 ′ by the subtractor 31 ′ to obtain a corrected image error signal d3 ′. The inter-color component prediction unit 32 ′ performs prediction by linear conversion with reference to the reference image signal 18 ′, outputs a prediction coefficient e3 ′, and generates a residual prediction image 33 ′. Next, the subtracting unit 34 ′ subtracts the corrected predicted image 33 ′ from the corrected image error signal d3 ′ to obtain an inter-color component predicted image component f3 ′ of the corrected image error signal d3 ′. Subsequently, the inter-color component prediction error component f3 ′ is sent to the DCT unit 35 ′, the quantization unit 36 ′, and the variable length coding unit 37 ′, and is encoded and output together with the prediction coefficient e3 ′.

  As described above, the encoding process is performed when the inter-color component prediction target is a frame image signal. In this encoding apparatus, since the second non-reference signal uses the inter-color component prediction error component 27 ′ of the first non-reference signal to obtain a corrected residual signal d3 ′. The prediction performance of the second non-reference signal can be improved compared to the case where the inter-color component prediction error component 27 ′ is not used.

  Next, in the decoding apparatus of FIG. 6, the input stream 41 ′ discriminates the reference signal and the first and second non-reference signals based on the identification information added to the stream by the reference signal discrimination unit 42 ′. . The reference signal stream (encoded signal) c4 ′ is partially decoded by the variable length decoding unit 43 ′, the partial decoding residual signal d4 ′ is sent to the inverse quantization unit 44 ′, and the intra prediction method e4 ′ is called intra. It is sent to the prediction unit 48 ′. The partial decoded residual signal d4 ′ is inversely quantized by an inverse quantization unit 44 ′, and is inversely DCTed by an inverse DCT unit 45 ′ to be a residual signal 46 ′. Also, the intra prediction unit 48 ′ generates an intra prediction image 49 ′ using the intra prediction method e4 ′, and the intra prediction image 49 ′ is added to the residual signal 46 ′ by the adding unit 47 ′ and decoded. Become an image. The decoded image is output and sent to the intra prediction unit 48 ′. The decoded image is sent to the inter-color component prediction units 55 ′ and 75 ′ as a reference decoded image 50 ′.

  Next, the first non-reference signal stream (encoded signal) c5 ′ is processed by the processing units 52 ′ to 57 ′ following the variable length decoding unit 51 ′ and the second non-reference signal stream (encoded signal). Although c6 ′ is decoded by the processing units 72 ′ to 79 ′ following the variable length decoding unit 71 ′, detailed description thereof is omitted.

  In the above decoding apparatus, the pre-decoding image 78 ′ of the second non-reference signal is added to the inter-color component prediction error component 54 ′ of the first non-reference signal and finally decoded. For this reason, decoding suitable for the image signal encoded by the encoding device of FIG. 4 can be performed efficiently.

  Next, a third embodiment in the case where the target of inter-color component prediction (3) is a frame image signal (using intra-frame prediction) will be described with reference to FIGS. FIG. 7 is a block diagram of the encoding device, and FIG. 8 is a block diagram of the decoding device. In this embodiment, in the encoding processing stage of the second embodiment, intra-frame prediction in the color component is applied to the encoding target, and the error signal is set as the encoding target.

  The main difference between the encoding device of FIG. 7 and the encoding device of FIG. 4 is that the error signal is obtained by applying the intra-frame prediction in the color component to the encoding target. Therefore, only the functional units surrounded by the dotted lines will be described below.

  The inter-color component prediction error component e2 ′ and the inter-color component prediction residual 105 of the image signal c2 ′ of the first non-reference signal are input to the intra prediction unit 100, and the intra prediction unit 100 generates an intra prediction image 102. . In addition, the subtraction unit 101 subtracts the intra prediction image 102 from the prediction error component e2 ′ to generate a color component inter-component prediction error component f2 ′. The adding unit 104 adds the intra prediction image 102 and the intra prediction residual 103 to generate the inter-color component prediction residual 105. The inter-color component prediction error component f2 ′ is encoded and output via the DCT unit 24 ′, the quantization unit 25 ′, and the variable length encoding unit 26 ′.

  The inter-color component prediction residual 105 is also sent to the subtraction signal 31 ′ for processing the image signal c3 ′ of the second non-reference signal, and is used for the subtraction processing. The corrected image error signal d3 ′ obtained by the subtraction process is then encoded and output in the same manner as in FIG.

  According to the encoding apparatus of this embodiment, since the second non-reference signal uses the inter-color component prediction residual component 105 of the first non-reference signal, the corrected residual signal d3 ′ is obtained. As a result, the prediction performance of the second non-reference signal can be improved as compared with the case where the inter-color component prediction residual component 105 is not used, and the code amount of the prediction residual signal can be reduced.

  FIG. 8 is a block diagram showing a configuration of a decoding apparatus corresponding to the encoding apparatus of FIG. Since the function of this decoding apparatus is similar to that of the decoding apparatus of FIGS. 3 and 6, detailed description of the function is omitted, but according to this decoding apparatus, the second non-reference signal is decoded before correction. The inter-color component prediction residual 120 obtained from the first non-reference signal is added to the image 130 to decode the second non-reference signal, so that the image 130 can be decoded adaptively. .

  The present invention has been described above with reference to the embodiments. However, the present invention is not limited to the above-described embodiments, and modifications within the scope of the present invention can be made by those skilled in the art. It is clear that it is included in

  13 ... Pre-processing analysis unit, 14 ... Reference signal selection unit, 18 ... Reference residual signal, 21, 32 ... Inter-color component prediction unit, 22 ... Residual prediction image, 27 ... color component prediction error component, 33 ... corrected residual prediction image, c1, c2, c3 ... residual signal, e2 ... color component prediction error component of residual signal, d3 .. Corrected residual signal, f3... Inter-color component prediction error component, 42... Reference signal determination unit, 43, 51, 71... Variable length decoding unit, 46. 54, 74 ... prediction error component between color components, 55, 75 ... prediction part between color components, 56 ... residual prediction image, 76 ... residual prediction image before correction, 78 ... decoding Uncorrected residual image, 58, 80... Decoded residual image, 50, 62 84 ... decoded image, 120 ... color components between residuals, 130 ... uncorrected decoded image.

Claims (7)

In a video encoding device capable of adaptively setting a reference signal for prediction between color components,
Reference signal selection means for selecting a reference signal, a first non-reference signal, and second and subsequent non-reference signals from an input image signal composed of a plurality of color components;
In the encoding of the first non-reference signal, first means for performing prediction with reference to only the reference signal;
The encoding of the second and subsequent non-reference signals comprises second means for performing a prediction with reference to the reference signal and the first non-reference signal ;
The second means corrects a signal to be subjected to the inter-color component prediction using an inter-color component prediction residual signal of the first non-reference signal, and then refers to the reference signal. A video encoding apparatus characterized by performing inter prediction . The moving image encoding device according to claim 1,
The first and second means perform inter-color component prediction, and a target of the inter-color component prediction is an intra-frame prediction residual signal in the component or an image signal itself. . The moving image encoding device according to claim 2,
The reference signal selection unit distinguishes the first, second and subsequent discrimination of the non-reference signal based on a correlation coefficient between the signal that is the target of inter-color component prediction and the signal that is the prediction reference destination. A moving picture coding apparatus, wherein a color component having high correlation is used as a first non-reference signal. The moving image encoding device according to claim 1 ,
The video encoding apparatus according to claim 1, wherein the correction of the signal to be subjected to the inter-color component prediction is performed by subtraction of the inter-color component prediction residual signal of the first non-reference signal. In the decoding apparatus of the signal encoded with the moving image encoding apparatus of Claim 1,
A discriminator for discriminating a reference signal, a first non-reference signal and a second and subsequent non-reference signals from the encoded signal;
In the decoding of the first non-reference signal, first means for performing prediction with reference to only the reference signal;
The decoding of the second and subsequent non-reference signals comprises second means for performing a prediction with reference to the reference signal and the first non-reference signal ;
In decoding of the second and subsequent non-reference signals, the residual signal of inter-color component prediction is inversely corrected using the inter-color component prediction residual signal of the first non-reference signal, and then the reference signal is A decoding apparatus characterized by performing inter-color component prediction . The decoding device according to claim 5 , wherein
The decoding apparatus characterized in that the first and second means perform inter-color component prediction, and the inter-color component prediction target is an intra-frame prediction residual signal in the component or an image signal itself. The decoding device according to claim 5 , wherein
The decoding apparatus according to claim 1, wherein reverse correction of the inter-color component prediction residual signal is performed by adding an inter-color component prediction residual signal of the first non-reference signal.