| 摘要 |
A field to be encoded is divided into blocks. Then, first, second, and third motion vectors relative to data of a block of the field to be encoded are calculated from data of an m-th preceding field (m: even), data of an n-th preceding field (n: odd), and data of an interpolated image produced from the n-th preceding field. First, second, and third prediction signals are produced from the first motion vector and the m-th preceding field data, the second motion vector and the n-th preceding field data, and the third motion vector and the interpolated image data, respectively. In succession, a first difference signal between the first prediction signal and the block data to be encoded, a second difference signal between the second prediction signal and the block data to be encoded, and a third difference signal between the third prediction signal and the block data to be encoded are calculated and compared each other. As a result of comparison, one of the first, second, and third difference signals is selected and encoded.
|
