Image encoding apparatus and method, and image encoding program

An image encoding apparatus that encodes an image so as to optimize estimated values of processing units which are a coding unit, a prediction unit, and a transform unit by selecting sizes of the individual processing units, the apparatus comprising:

• a division device that divides an image having a size of the coding unit into a plurality of blocks;
• a motion information computation device that computes motion information which indicates a motion state of each of the blocks by utilizing a numeric value:
• a transform unit size determination device that determines, based on the computed motion information, a size pattern of the transform unit within the image having the size of the coding unit,
• wherein the motion information computation device computes, for each of the blocks, motion information items of any two blocks adjacent to the relevant block, where the two blocks each have a size smaller than or equal to the size of the relevant block;
• and
• if a difference between the two motion information items of the two blocks adjacent to the relevant block is smaller than or equal to a predetermined threshold, the transform unit size determination device determines the size of the transform unit in the relevant block to be a block size of the relevant block, otherwise the relevant block is further divided into small blocks, and the transform unit size determination device determines the size pattern of the transform unit by repeatedly subjecting, a predetermined number of times, each of the small blocks to the motion information acquisition of the adjacent blocks and the comparison between the difference of the relevant motion information items and the threshold.

An image encoding method that encodes an image so as to optimize estimated values of processing units which are a coding unit, a prediction unit, and a transform unit by selecting sizes of the individual processing units, the method comprising:

• a division step that divides a coding unit included in the image into four divided blocks having a block size;
• an activity computation step that computes activities pertaining to the four divided blocks;
• and
• a transform unit size determination step that determines, based on the computed activities, a size pattern of the transform unit within the image having the size of the coding unit,
• wherein the transform unit size determination step including:
• determining whether any one of the four divided blocks has an adjacent block which produces an activity difference smaller than or equal to a predetermined threshold;
• when any one of the four divided blocks has the adjacent block, determining the block size of the four divided blocks as the size pattern of the transform unit;
• and
• when none of the four divided blocks has the adjacent block, and determining a smaller block size as the size pattern of the transform unit, the smaller block size being obtained by dividing the block size of the four divided blocks.

An image encoding method that encodes an image so as to optimize estimated values of processing units which are a coding unit, a prediction unit, and a transform unit by selecting sizes of the individual processing units, the method comprising:

• a division step that divides an image having a size of the coding unit into a plurality of blocks;
• a motion information computation step that computes motion information which indicates a motion state of each of the blocks by utilizing a numeric value:
• a transform unit size determination step that determines, based on the computed motion information, a size pattern of the transform unit within the image having the size of the coding unit,
• wherein the motion information computation step computes, for each of the blocks, motion information items of any two blocks adjacent to the relevant block, where the two blocks each have a size smaller than or equal to the size of the relevant block;
• and
• if a difference between the two motion information items of the two blocks adjacent to the relevant block is smaller than or equal to a predetermined threshold, the transform unit size determination step determines the size of the transform unit in the relevant block to be a block size of the relevant block, otherwise the relevant block is further divided into small blocks, and the transform unit size determination step determines the size pattern of the transform unit by repeatedly subjecting, a predetermined number of times, each of the small blocks to the motion information acquisition of the adjacent blocks and the comparison between the difference of the relevant motion information items and the threshold.