Apparatus for motion vector estimation in a television image uses a block matching technique with successive refinement of the motion vector estimate. The apparatus comprises a vector filter and a vector calculator.

A receiver for a high-definition television signal, in which the signal prior to transmission is sub-sampled on a segmented basis according to the degree of movement, has an input terminal (1) which receives the sub-sampled signal which is applied to an inverse shuffler (2) which shuffles the pixels of each block in a manner which is the inverse to that performed prior to transmission. The output of shuffler (2) is applied to a first delay arrangement (4) which provides one or more delays of an integral number of fields periods and whose delayed and undelayed outputs are applied to an input field selector (6) to produce a changed sequence of fields to that received. The selector output is applied to a processing arrangement (8) in which the signal of changed field sequence is subjected to spatial interpolation and/or temporal interpolation and the resulting interpolated signal applied to a second delay arrangement (10) providing one or more delays of an integral number of fields periods. The delayed and undelayed outputs of delay arrangement (10) are applied to an output field selector (12) to produce an output in which the original field sequence is restored.

A correlation surface is derived by phase correlating two pictures selectively displaced in the X and Y directions. The illustrative surface shows a large peak at zero displacement, corresponding to a stationary background, and a fairly large peak corresponding to a moving object, the X, Y position of the peak indicating the magnitude (pixels per field period) and direction of the motion vector. A set of motion vectors is thus determined and testing is then carried out, on a pixel by pixel basis or pixel block by pixel block basis, to determine which of the motion vectors gives the best match in deriving the second picture from the first. The motion vector thus assigned may be used in temporal interpolation of the pictures.

For a video-signal-coding facility using adaptive DPCM in which switching is effected block by block between different predictors, a basic circuit arrangement is given for the case where switching is to be effected between a two-dimensional intraframe predictor and a pure interframe predictor (FIG. 1). Uniform processor elements (PE.sub.1, PE.sub.0) cooperate with predictor loops (Pr.sub.1, Pr.sub.2, 10, 11) and circuits are provided which ensure that the appropriate reconstructed sample values (X.sub.0 ', X.sub.1 ') are used whenever switching between coders occurs. In a preferred embodiment (FIG. 3), the decision on the best suited coder is delayed by one block, whereby the maximum processing speed is substantially increased. Circuits (PE.sub.2, 12, 13, S.sub.3 ', S.sub.4 ', S.sub.4 ') are provided which ensure that even in case of such a delayed decision, the correct reconstructed sample values (x.sub.0 ', x.sub.1 ', x.sub.2 ') are used.

The present invention relates to an encoding apparatus for digitizing and encoding an input image signal, the input image signal being sequentially sampled to sequentially form sample signals at respective sample points. The encoding data which respectively correspond to the sample signals at respective sample points are formed, and the plurality of resultant encoding data are sequentially output on a group unit basis of k (k being an integer equal to or greater than 2) encoding data in accordance with the order different from the forming order of the encoding data. In accordance with a combination of the k encoding data which have been output, the k encoding data are converted into the data train having less DC and low frequency components and output, so that the image signal can be highly efficiently encoded.