| 摘要 |
<p>Apparatus for compression encoding video signals, comprising block segmenting means (3A, 3B, 3C) for segmenting a vertical interval, such as a field or frame, of video picture elements into a plurality of blocks of picture elements. An orthogonal transformation, such as discrete cosine transformation, of respective blocks is obtained (6), resulting in a two-dimensional array of transform coefficients of different values, the array being partitioned into respective areas. Each area is quantized (8) with a respective quantizing value that differs for different areas. Preferably, a transform coefficient is quantized by dividing the value of the transform coefficients in an area by a divisor 2<n/2>, where n is an integer that differs in different areas. The areas of the two-dimensional array of transform coefficients are formed by partitioning the array in a direction substantially perpendicular to the oblique frequency axis of that array. <IMAGE></p> |
