Method and apparatus for encoding video signal

摘要

Input picture data represents a succession of pictures each composed of plural blocks. The input picture data is quantized into first quantization-resultant data in response to a fixed quantization width. The first quantization-resultant data is encoded into first code data of a variable length code for every block. Information of an accumulation transition in an amount of the first code data is generated for every block. A code amount accumulation transition value is calculated from the accumulation-transition information and a predetermined target code amount for every block. The predetermined target code amount is defined with respect to every picture. The input picture data is quantized into second quantization-resultant data in response to a variable quantization width. The second quantization-resultant data is encoded into second code data of the variable length code for every block. An actual amount of the second code data is detected for every block. Calculation is made as to an error between the code amount accumulation transition value and the detected actual amount of the second code data for every block. The variable quantization width is controlled in response to the calculated error for every block to maintain an actual amount of the second code data substantially at the target code amount for every picture.