摘要 |
PROBLEM TO BE SOLVED: To encode a high quality image with a high compression rate by a simple operation without requiring special data. SOLUTION: An inputting part 1 receives an image signal, and an 8×8 block input circuit 2a divides the image into blocks of 8×8 pixels. An average calculation circuit 3a calculates the average of pixel values in a divided block, and a predictive error quantizer 4a calculates a predictive error from the calculated average and quantizes the predictive error. A predictive error inverse quantizer and a new average calculation circuit 5a dequantize the quantized error to calculate an error and add or subtract the error to/from the calculated error to calculate a new average. An error calculation circuit 6a compares the new average with all pixel values in the block to thereby calculate an error. An error comparison circuit 7a compares the calculated error with a threshold. COPYRIGHT: (C)2004,JPO |