A method of encoding pixel values in a merged block after merging an object boundary block when encoding image information

摘要

According to the present invention, when a merged block exists after object boundary block merging is applied, the merged after object boundary block merging when DCT-encoded image information encoding is determined as an average value of pixel values applied before the merge. The present invention relates to a method of encoding pixel values in a block, wherein the prediction error correction or original image information is input from the motion compensator 12 and the VOP forming unit 10, and the original image shape information is received from the shape information encoder 17. Merging of the object boundary block to merge the object boundary block from the signal and the restored shape information of the out-of-object area padding unit 21 and the outer object padding unit 21 to receive the restored shape information. Determining the pixel value for the extra-object pixel in the merged block that receives the signal from the unit 22 and the object boundary block merging unit 22 and the original image shape information (23), the pixel value determination unit 23 for the out-of-object pixel in the merged block, and the pixel value determination unit 23 for the out-of-object pixel in the merged block are DCT-coded and coded for the texture signal. A DCT and encoder 24 for outputting information to the multiplexer, an IDCT and encoder 25 for IDCT and encoding signals of the DCT and encoder 24, and the restored shape information and IDCT and encoder ( And a merged separator 26 that separates the merged block from the signal of 25) and outputs the decoded information to the previous VOP detector.