VIDEO SIGNAL CODING METHOD

摘要

PROBLEM TO BE SOLVED: To efficiently encode video signals by detecting an edge point from the preceding frame and selecting feature points based on a prescribed grid and the edge point. SOLUTION: A subtractor substrates the preceding frame signal on a line L12 connected to a frame memory 124 from present frame signal on a line L10 connected to a frame memory. Then the frame differential data indicating the differential pixel value between the pixels existing at the same position in the present and preceding frames are supplied to a comparing block 312. The block 312 compares the data with a preset threshold TH one pixel by one pixel, sets a converted value at '0' or '1', and supplies the value to a frame memory 314. A region selecting block 316 fetches the converted value of each pixel in the memory 314 and selects a region having a differential pixel value larger than a threshold TH1 as a first region. Then a plurality of pixel in the preceding frame is selected from the area as feature points based on a prescribed grid and an edge point detected from the preceding frame.