| 摘要 |
<p><P>PROBLEM TO BE SOLVED: To attain to reduce the amount of generated code without degrading subjective image quality and to speed up coding processing, in moving image coding in which a visual distortion model in consideration of contrast sensitivity is used for mode selection. <P>SOLUTION: Coding cost calculation and processing portions 20, 25 calculate the amount of visual distortion used in a cost function which calculates a coding cost for selecting a prediction mode, wherein in the case that this amount of visual distortion is equal to or below a critical point threshold value at predetermined image quality degradation detection, the amount of visual distortion is updated to zero, and otherwise the critical point threshold value is subtracted from the amount of visual distortion and its result is used as the amount of visual distortion after updating. Moreover, in the case that the amount of visual distortion after updating to a prediction residual signal is zero, the prediction residual signal is updated to all zero. The selection of prediction mode is made by using the amount of visual distortion after updating. <P>COPYRIGHT: (C)2009,JPO&INPIT</p> |
