| 摘要 |
PROBLEM TO BE SOLVED: To improve the quality of a video image after compression as a whole by adjusting a generated bit amount in response to complicatedness of a pattern for each part of video data while keeping the bit amount to be a prescribed value or over. SOLUTION: Real difficulty data D1 corresponding to a data amount after compression are calculated based on index data denoting complexity of a pattern of a video image (ME residue, flatness and intra AC and activity) or a global complexity used for calculation of a quantized value by the MPEG. Furthermore, a parameter Rj ' is adjusted depending on the occupancy of the VBV(video buffering verifier) buffer. In the case of the adjustment, a prescribed lower limit is set so as to prevent the quality of compressed video data from being remarkably deteriorated. A multiple calculated on the real difficulty data Dj is multiplied with the adjusted parameter Rj ' and an object data quantity Tj denoting the object data amount of a video image after compression is calculated and it is used for compression coding processing. |
