DYNAMIC IMAGE CODING METHOD AND DEVICE

摘要

PROBLEM TO BE SOLVED: To reduce transmission information quantity by selecting a motion vector with a minimum magnitude within a range where a coding object macro block is discriminated to be nonsignificant in the coding of a motion compensation prediction error method. SOLUTION: A frame memory 4 stores a video signal from an input section 2 and gives a video signal 5 of a coding object macro block of a current frame to a subtractor section 13 and gives a video signal 7 of each macro block of a preceding frame to a motion compensation section 8. The motion compensation section 8 gives a video signal 12 of a macro block of a preceding frame whose motion is compensated by motion vector information 9 to a subtractor section 13. A significance/non-significance discrimination section 15 discriminates an inter-macro block difference value 14 from the subtractor section 13 based on a threshold level to give a macro block significance signal 21 to a coding engine section. A motion vector generating section 11 generates motion vector information 9 whose picture element is increased by a prescribed number so as to increase the magnitude sequentially while being started from a zero vector (0,0).