Motion picture encoding device and motion picture encoding processing program
The minimum Sum of Absolute Differences obtained by a motion vector search roughly judges the magnitude of quantization error by whether or not exceeding a predetermined threshold value. When the quantization error is lower, whether or not visually noticeable noise exists in some of the pixels of the current macroblock is judged based on the amount of flatness and noise detected in each of the 4×4 pixel blocks of the current macroblock partitioned into 16 sub-macroblocks. If there is visually noticeable noise, intra-frame coding is selected. When the quantization error is higher, whether or not visually noticeable noise exists in the current macroblock is judged while considering the magnitude of the motion vector. If there is visually noticeable noise, intra-frame coding is selected.
1. A motion picture encoding device which selects either interframe coding or intra-frame coding to perform video compression, comprising:
a motion compensation error detection unit which detects motion compensation error upon interframe prediction;
a detection unit which detects flatness of a macroblock image; and
an encoding determination unit which compares the motion compensation error with a product acquired by multiplying the flatness of the macroblock image by a coefficient which increases as a magnitude of a motion vector becomes larger, when the motion compensation error detected by the motion compensation error detection unit is greater than a first value during interframe coding by a macroblock image unit, and selects intra-frame coding if the motion compensation error is greater;
wherein the encoding determination unit selects intra-frame coding if a formula [1] is satisfied:
minSAD>(1+ a×f ( mv ))× k ×Flat [1]
where minSAD is a minimum Sum of Absolute Differences obtained by a motion vector search, coefficient a is constant to adjust a contributive factor of the motion vector, coefficient k is constant, function f (mv) is a magnitude of a current macroblock motion vector, and Flat is an amount of flatness.
2. The motion picture encoding device according to claim 1 , further comprising:
a partitioning unit which obtains block images by partitioning the macroblock image; and
a block image determination unit which detects flatness and quantization error of the block images and determines whether any of the block images has visually noticeable noise based on comparison results of: (i) the detected quantization error and a second value, and (ii) the detected flatness and a third value;
wherein the encoding determination unit selects intra-frame coding when the motion compensation error detected by the motion compensation error detection unit is less than the first value, and when the block image determination unit determines that one of the block images within the macroblock image has visually noticeable noise.
3. The motion picture encoding device according to claim 1 , wherein the encoding determination unit selects intra-frame coding by comparison of a fourth value and the flatness of the macroblock image detected by the detection unit, when the motion compensation error detected by the motion compensation error detection unit is less than the first value.
4. A motion picture encoding method which selects either inter-frame coding or intra-frame coding to perform video compression, the method comprising:
detecting motion compensation error upon inter-frame prediction;
detecting flatness of a macroblock image; and
comparing the motion compensation error with a product acquired by multiplying the flatness of the macroblock image by a coefficient which increases as a magnitude of a motion vector becomes larger, when the detected motion compensation error is greater than a first value during inter-frame coding by a macroblock image unit, and selecting intra-frame coding if the motion compensation error is greater;
wherein intra-frame coding is selected if a formula [1] is satisfied:
minSAD>(1+ a×f ( mv ))× k ×Flat [1]
where minSAD is a minimum Sum of Absolute Differences obtained by a motion vector search, coefficient a is constant to adjust a contributive factor of the motion vector, coefficient k is constant, function f (mv) is a magnitude of a current macroblock motion vector, and Flat is an amount of flatness.
5. The motion picture encoding method according to claim 4 , further comprising:
obtaining block images by partitioning the macroblock image; detecting flatness and quantization error of the block images and determining whether any of the block images has visually noticeable noise based on comparison results of: (i) the detected quantization error and a second value, and (ii) the detected flatness and a third value; and
selecting intra-frame coding when the detected motion compensation error is less than the first value during inter-frame coding by the macroblock image unit, and when it is determined that one of the block images within the macroblock image has visually noticeable noise.
6. The motion picture encoding method according to claim 4 , further comprising:
selecting intra-frame coding by comparison of a fourth value and the flatness of the macroblock image, when the detected motion compensation error is less than the first value during inter-frame coding by the macroblock image unit.
7. A non-transitory computer-readable storage medium having a program stored thereon which is executable by a computer controlling a motion picture encoding device which selects either inter-frame coding or intra-frame coding to perform video compression, wherein the program causes the computer to perform functions comprising:
detecting motion compensation error upon inter-frame prediction;
detecting flatness of a macroblock image; and
comparing the motion compensation error with a product acquired by multiplying the flatness of the macroblock image by a coefficient which increases as a magnitude of a motion vector becomes larger, when the detected motion compensation error is greater than a first value during inter-frame coding by a macroblock image unit, and selecting intra-frame coding if the motion compensation error is greater;
wherein intra-frame coding is selected if a formula [1] is satisfied:
minSAD>(1+ a×f ( mv ))× k ×Flat [1]
where minSAD is a minimum Sum of Absolute Differences obtained by a motion vector search, coefficient a is constant to adjust a contributive factor of the motion vector, coefficient k is constant, function f (mv) is a magnitude of a current macroblock motion vector, and Flat is an amount of flatness.
8. The non-transitory computer-readable storage medium according to claim 7 , wherein the program causes the computer to perform further functions comprising:
obtaining block images by partitioning the macroblock image;
detecting flatness and quantization error of the block images and determining whether any of the block images has visually noticeable noise based on comparison results of: (i) the detected quantization error and a second value, and (ii) the detected flatness and a third value; and
selecting intra-frame coding when the detected motion compensation error is less than the first value during inter-frame coding by the macroblock image unit, and when it is determined that one of the block images within the macroblock image has visually noticeable noise.
9. The non-transitory computer-readable storage medium according to claim 7 , wherein the program causes the computer to perform a further function comprising:
selecting intra-frame coding by comparison of a fourth value and the flatness of the macroblock image, when the detected motion compensation error is less than the first value during inter-frame coding by the macroblock image unit.