IMAGE COMPRESSION DEVICE, COMPRESSION METHOD, RECORDING MEDIUM HAVING PROGRAM RECORDED, AND IMAGE DECOMPRESSION DEVICE, DECOMPRESSION METHOD, AND RECORDING MEDIUM HAVING PROGRAM RECORDED

摘要

When image data is compressed, compressed pixel level values existing in a line to which a compressing target pixel belongs and in the preceding line are stored. It is determined whether quantization error is present by checking how the same value appears in the level difference between the two lines according to the compressed pixel level values so as to select one of quantization tables having different quantizing steps in the quantizer. Alternatively, a sum of absolute values of level differences of pixels around the compressing target pixel is calculated. Then, it is determined whether the compressing target pixel is a flat image or a highly-active image by checking the sum of the peripheral pixel level differences so as to select one of quantization tables having different quantizing steps in the quantizer. Thus, by switching the quantization table from one to another, it is possible to suppress image degradation between the lines caused by a quantization error.