High definition television transmission and reception system with an improved motion estimator and with reduced data throughput

Claim 2

A high definition television transmission system in which an image is processed as a sequence of frames each comprising first and second fields, each of said fields being divided into a plurality of image parts, the system comprising a motion estimator for determining motion vectors for a number of said image parts, said system further comprising: means for determining an optimum motion vector for a first image part of a second field of a subsequent frame by examining a plurality of respective prior motion vectors determined for each of a group of said number of image parts, said group comprising a second image part of a second field of a preceding frame and a plurality of image parts spatially adjacent to said second image part, said second image part having the same respective spatial position as said first image part within their respective fields; means coupled to said motion estimator, for selecting a video data transmission mode for each of said image parts for which motion vector data has been determined; and means coupled to said selecting means, for transmitting data representing said motion vectors on a digital transmission channel.

Claim 3

The high definition television transmission system of claim 1, wherein said determining means further comprises: means for decoding motion vectors determined for said subsequent and preceding frames; and means for comparing the decoded motion vectors.

Claim 4

The high definition television system of claim 1, wherein said determining means further comprises: means for determining a selection motion vector from said prior motion vectors which best represents said optimum motion vector; and wherein data representing said optimum motion vector comprises a signal which indicates said selected motion vector.

Claim 5

A high definition television transmission system, in which a first image and a preceding image are divided into a plurality of image parts, for each one of which a video data transmission mode is selected from among at least two modes which have different sampling structures, the system comprising means for determining a motion vector for each one of the image parts, and means for transmitting, on an associated digital channel, data defining the mode selected for each one of the image parts and the motion vectors determined for image parts for which the first mode is selected, characterized in that said system further comprises: a) means for associating a first image part of said first image for which said first mode has been selected, with a group of a plurality of image parts of said preceding image, said group containing a second image part occupying, in said preceding image, the same position as said first image part occupies in said first image; b) means for making comparisons between the motion vector determined for said first image part and each of the motion vectors determined for respective image parts of said group to determine if one of said comparisons is positive, wherein, if no comparison is positive, said first image part is processed according to a second mode, and wherein if one of the comparisons is positive, the selection of said first mode is confirmed and a data item is transmitted which identifies the one of said image parts of said group for which the comparison was positive.

Claim 6

A high definition television transmission system, in which a first image and a preceding image are divided into a plurality of image parts, for each one of which a video data transmission mode is selected from among at least two modes which have different sampling structures, the system comprising means for determining a motion vector for each one of the image parts, and means for transmitting, on an associated digital channel, data defining the mode selected for each one of the image parts and the motion vectors determined for image parts for which the first mode is selected, characterized in that said system further comprises: a) means for associating a first image part of said first image for which said first mode has been selected, with a group of a plurality of image parts of said preceding image, said group containing a second image part occupying, in said preceding image, the same position as said first image part occupies in said first image; b) means for making comparisons between the motion vector determined for said first image part and each of the motion vectors determined for respective image parts of said group to determine if one of said comparisons is positive, wherein, if no comparison is positive, said first image part is processed according to a second mode, and wherein if one of the comparisons is positive, the selection of said first mode is confirmed and a data item is transmitted which identifies the one of said image parts of said group for which the comparison was positive.

Claim 7

The high definition television system of claim 4, wherein a comparison is determined to be positive if there is an identity between the coordinates of the motion vectors being compared.

Claim 8

The high definition television system of claim 4, further comprising: means for decoding motion vectors determined for said first image and said preceding image; means for comparing the decoded motion vectors and for determining said selected mode.

Claim 9

The high definition television system of claim 6, wherein a comparison is determined to be positive if the difference between the motion vector of said first image part and the motion vector of a compared image part is the smallest of those motion vectors compared.

Claim 10

The high definition television system of claim 6, wherein a comparison is determined to be positive if the difference between the motion vector of said first image part and the motion vector of a compared image part is lower than a predetermined threshold.

Claim 11

A high definition television transmission system according to claim 4, further comprising: a) means for determining within said motion vectors determined for the image parts of said preceding image, a sub-set of vectors which are encountered most frequently in said preceding image; and b) means for transmitting digitally the characteristics of said sub-set of vectors whereby, if the motion vector determined for said first image part is one of said sub-set of vectors, the first mode is selected for transmitting video data related to said first image part and its motion vector is defined by reference to said sub-set, and, if a motion vector of said first image part is not one of said sub-set of vectors, said second mode is selected to transmit data related to said first image part.

Claim 12

A high definition television receiver for representing transmitted images, in which each image is divided into a plurality of image parts, for each one of which a video data reproduction mode making use of motion vector data is used, said motion vector data being received for each one of the image parts to be processed in this mode, the receiver being equipped with means for creating, on the basis of the motion vector data, an intermediate image between two images received consecutively, said creating means further comprising at least one memory for recording motion vector data, characterized in that said receiver is further equipped with means for creating a part of a second image and on the basis of a received digital data item identifying a motion vector of a part of a first image applying it to the part of the second image.

Claim 13

A high definition television receiver for representing transmitted images, in which each image is divided into a plurality of image parts, for each one of which a video data reproduction mode making use of motion vector data is used, said motion vector data being received for each one of the image parts to be processed in this mode, the receiver being equipped with means for creating, on the basis of the motion vector data, an intermediate image between two images received consecutively, said creating means further comprising at least one memory for recording motion vector data, characterized in that said receiver is further equipped with means for creating a part of a second image and on the basis of a received digital data item identifying a motion vector of a part of a first image applying it to the part of the second image.