| 摘要 |
PURPOSE:To enable reducing whole search operation quantity for detecting a motion vector by detecting the motion vector for a part area which is constituted of an input screen field and permitting a position on a reference screen, which is indicated by means of an obtained field vector, as the reference point of search for a frame vector. CONSTITUTION:An input frame and a reference frame are generated through the use of a read pair of input field and reference field. That, is, the input frame 101 is constituted by the input fields 103, 104 and the reference frame 102 is constituted by the reference fields 105 and 106. The part area to be read is constituted of the part area 201 in the input frame 101, the part area 203 in the input field 103 and the part area 204 in the field 104. Then, the part area of the two input fields which constitutes the part area in one input frame is detected and the frame vector of the reference part area and the input part area is selected from the obtained two field vectors of one picture element precision. |
