Image processing device and method
View Patent ↗The present invention relates to an image processing device and method, which realize improvement in encoding efficiency for color difference signals and reduction in address calculations for memory access. In a case where a block size of orthogonal transform is 4×4, and a macroblock of luminance signals is configured of four 4×4 pixel blocks appended with 0 through 1, the four luminance signal blocks are corresponded with one color difference signal 4×4 block appended with C. At this time, there exist four motion vector information of mv 0 , mv 1 , mv 2 , and mv 3 , as to the four luminance signal blocks. The motion vector information mv c of the one color difference signal 4×4 block is calculated by averaging processing using these four motion vector information. The present invention can be applied to an image encoding device which performed encoding based on the H.264/AVC format, for example.
1. An image processing device comprising:
motion prediction means configured to perform motion prediction of a luminance block to be encoded, and generate a motion vector;
block determining means configured to classify a color difference block, corresponding to said luminance block regarding which said motion vector has been generated by said motion prediction means, into a category according to a block size of orthogonal transform; and
color difference motion vector generating means configured to, in the event that said color difference block has been classified by said block determining means into a category of a size smaller than a block size of orthogonal transform, correspond a single color difference block as to a plurality of said luminance blocks, so as to generate a motion vector for said single color difference block.
2. The image processing device according to claim 1 , wherein said color difference motion vector generating means generate the motion vector for said single color difference block using at least one of motion vectors of a plurality of said luminance blocks generated by said motion prediction means.
3. The image processing device according to claim 2 , wherein said color difference motion vector generating means generate the motion vector for said single color difference block by a function using at least one of motion vectors of a plurality of said luminance blocks generated by said motion prediction means.
4. The image processing device according to claim 2 , further comprising:
encoding means configured to encode information of motion vectors of said luminance blocks generated by said motion prediction means;
wherein said color difference motion vector generating means generate a prediction motion vector for said single color difference block by a function using at least one of motion vectors of a plurality of said luminance blocks obtained by said motion prediction means, and obtain a difference between said prediction motion vector that has been generated and the motion vector of said single color difference block searched from a predetermined range centered on said prediction motion vector;
and wherein said encoding means encode information of said difference obtained by said color difference motion vector generating means.
5. The image processing device according to claim 2 , wherein, in the event that said color difference block is classified into a category of a size equal to or greater than said orthogonal transform block size by said block determining means, said motion prediction means generate the motion vector of said color difference block by a method according to an encoding standard.
6. An image processing method comprising the steps of:
an image processing device
performing motion prediction of a luminance block to be encoded, and generating a motion vector,
classifying a color difference block, corresponding to said luminance block regarding which said motion vector has been generated, into a category according to a block size of orthogonal transform, and
in the event that said color difference block has been classified into a category of a size smaller than a block size of orthogonal transform, corresponding a single color difference block as to a plurality of said luminance blocks, so as to generate a motion vector for said single color difference block.
7. An image processing device comprising:
decoding means configured to decode information of a prediction mode for a luminance block to be decoded;
block determining means configured to classify a color difference block corresponding to said luminance block into a category corresponding to a block size of orthogonal transform, based on a size of said luminance block which information of said prediction mode decoded by said decoding means indicates; and
color difference motion vector generating means configured to, in the event that said color difference block has been classified into a category by said block determining means of a size smaller than a block size of orthogonal transform, correspond a single color difference block as to a plurality of said luminance blocks, so as to generate a motion vector for said single color difference block.
8. The image processing device according to claim 7 , wherein said decoding means decode information of a motion vector of said luminance block;
and wherein said color difference motion vector generating means generate the motion vector for said single color difference block using at least one of motion vectors of a plurality of said luminance blocks decoded by said decoding means.
9. The image processing device according to claim 8 , wherein said color difference motion vector generating means generate the motion vector for said single color difference block by a function using at least one of motion vectors of a plurality of said luminance blocks decoded by said decoding means.
10. The image processing device according to claim 8 , wherein said decoding means decode information of difference between a prediction motion vector for said single color difference block generated by calculating an average value using motion vectors of a plurality of said luminance blocks encoded at an encoding side, and the motion vector of said single color difference block searched from a predetermined range centered on said prediction motion vector;
and wherein said color difference motion vector generating means generate the motion vector for said single color difference block by generating a prediction motion vector for said single color difference block by a function using at least one of motion vectors of a plurality of said luminance blocks obtained by said motion prediction means, and adds said prediction motion vector that has been generated to the information of said difference decoded by said decoding means.
11. The image processing device according to claim 8 , further comprising:
prediction image generating means configured to generate a prediction image of said luminance blocks using motion vectors of said luminance blocks;
wherein, in the event that said color difference block is classified into a category of a size equal to or greater than said orthogonal transform block size by said block determining means, said prediction image generating means generate the motion vector of said color difference block by a method according to an encoding standard, and generate a prediction image for said color difference block.
12. An image processing method comprising the steps of:
an image processing device
decoding information of a prediction mode for a luminance block to be decoded,
classifying a color difference block corresponding to said luminance block into a category corresponding to a block size of orthogonal transform, based on a size of said luminance block which information of said prediction mode that has been decoded indicates, and
in the event that said color difference block has been classified into a category of a size smaller than a block size of orthogonal transform, corresponding a single color difference block as to a plurality of said luminance blocks, so as to generate a motion vector for said single color difference block.