| 摘要 |
<p>PURPOSE:To prevent encoding accuracy from being lowered in the case of encoding the data of edited images and storing them in a storage means by correcting the parameter of a quantizing matrix used for a coefficient quantizing means according to the editing work processing contents of image data. CONSTITUTION:The edited image data are read from an image memory 1 to a block read part 20, transformed to a DCT transformation coefficient value by a DCT transformation part 21 and quantized by a quantizing part 22. A quantizing coefficient value outputted from the quantizing part 22 is allocated with a Huffman code by a Huffman encoding part 23 and stored in a code storage part 3. Further, when writing the code data of edited images in the code storage part 3, the quantizing matrix read from a quantizing matrix storage part 8 and corrected is written as a header and used for the next decoding. Thus, accurate encoding is enabled.</p> |
