Video coding device, video coding procedure, video coding program, video decoding device, video decoding procedure and video decoding program

摘要

A motion image coding device that encodes moving images into units of a block by dividing a first block, which is obtained by dividing each image of the moving images into first blocks of predetermined sizes, at least a second block, comprising the device: a quantification parameter calculation unit configured to calculate a quantification parameter of at least a second block; a derivation unit of prediction quantification parameters configured to derive a prediction quantification parameter of at least one second block using quantization parameters of a third block adjacent to the left of at least a second block and a fourth block adjacent to the part superior of at least a second block; a differential quantification parameter generation unit configured to generate a differential quantification parameter of at least a second block from a difference between the quantization parameter of at least a second block and the prediction quantification parameter of at least one Second block; and an encoding unit configured to encode the differential quantification parameter of the at least a second block, in which the derivation of prediction quantization parameters sets the quantification parameter of the third block as the first quantification parameter when the third block is in a position that is not beyond a left boundary of the first block, it establishes a quantification parameter of a fifth block of the same encoded image immediately before at least a second block as the first quantification parameter when the third block is in a position beyond the left limit of the first block, set the quantification parameter of the fourth block as a second quantification parameter when the fourth block is in a position that is not beyond the upper limit of the first block, set the quantification parameter of the fifth block com or the second quantification parameter when the fourth block is in a position that is beyond the upper limit of the first block, and derives the prediction quantification parameter of at least a second block using the first and second quantization parameters.