Abstract: An apparatus detects a motion vector using image data to be encoded, and a plurality of reference image data. The apparatus conducts a motion vector search for first reference image data using a large block obtained by dividing the image data to be encoded, and divides the large block into a smaller block size at the position of a smallest matching error. The apparatus determines a block size used in a motion vector search for another reference data based on matching errors in the small blocks.

Abstract: A motion vector detection apparatus is configured to calculate a temporal distance between a frame to be coded and each of a plurality of reference candidate frames referred to by the frame to be coded. The motion vector detection apparatus searches for a candidate motion vector between the frame to be coded and each the plurality of reference candidate frames and detects a motion vector for the frame to be coded from the candidate motion vectors. In searching for and detecting a candidate motion vector, the motion vector detection apparatus changes an amount of the calculation performed during the detection of a candidate motion vector according to the calculated temporal distance between the frame to be coded and the reference candidate frame, and a coding type of the reference candidate frame.

Abstract: A moving image encoding apparatus, comprising, a first image transformation unit adapted to transform image data into first image information, with reference to reference image data, a second image transformation unit adapted to inverse transform the first image information into reconstructed image data and a determination unit adapted to determine, based on the image data and the reconstructed image data, whether or not to use the reconstructed image data as the reference image data, wherein the determination unit calculates a peak signal-to-noise ratio (PSNR) from the reconstructed image data and the image data, and, if the PSNR is greater than or equal to a predetermined threshold, determines to use the reconstructed image data as the reference image data.

Abstract: A motion compensator detects a first shift amount between temporally adjacent first and second frames, wherein the first shift amount is detected based upon a differential value between pixels constituting the first frame and pixels constituting the second frame; stores in a storing unit of information of the detected first shift amount with regard to each reference frame candidate; calculates a second shift amount between the input frame and each reference frame candidate based upon the first shift amount or a sum of the first shift amount from each reference frame candidate to the input frame; selects a reference frame candidate for which the second shift amount is smallest as a reference frame; and calculates a motion vector, with regard to each macro block of the input frame, based upon each macro block of the reference frame.

Abstract: An image processing apparatus includes an encoding unit encoding an input motion image signal, a recording unit recording encoded motion image data on a recording medium, a complexity detection unit detecting complexity of a motion image according to the input motion image data, a remaining amount detection unit detecting a remaining recording capacity of the recording medium, and a control unit having a first mode for controlling the encoding unit based on information of the complexity detected by the complexity detection unit so as to adjust a code amount of the encoded motion image data and a second mode for controlling the encoding unit without using the information of the complexity detected by the complexity detection unit so as to adjust the code amount of the encoded motion image data, wherein the control unit switches over the first mode and the second mode on the basis of an output of the remaining amount detection unit.

Abstract: When detecting a motion vector, the present invention sets a search window not on a current macro block but in a reference image, by obtaining a global vector expressing the motion between a current image and a reference image and using that global vector to set the search window in the reference image, thus enabling detection of a motion vector even in images with much motion.