High efficiency coding apparatus

Claim 2

The apparatus of claim 1, wherein sample values of video signals of consecutive two fields entered by the sample value decimating means are decimated of sample values in the different horizontal and vertical positions between fields and in every other sample value.

Claim 3

The apparatus of claim 1, wherein the decimating means determines by using the sample points existing between lines in which the transmission sample points belong to a same field, between lines adjacent on the screen and belonging to consecutive fields, or between lines on the same position on the screen and belonging to consecutive frames, and the positions of the sample points to be decimated differ between lines.

Claim 4

The apparatus of claim 1, wherein the transmission field composing means prepares two fields composed of transmission sample values of horizontal H and vertical V lines obtained by the decimating means, separates the two fields into V groups containing two lines between adjacent fields on the screen, sequentially arranges the transmission sample values of different fields in every transmission sample value in every one of V groups to compose the information of one line composed of 2H transmission sample values, thereby composing one transmission field of horizontal direction 2H and vertical direction V lines.

Claim 5

A high efficiency coding apparatus comprising sample value decimating means for decimating sample values of input video signals in every other sample value to obtain transmission sample values, block forming means for collecting a plurality of transmission sample points obtained by the decimating means over plural fields to form blocks, motion detecting means for judging whether a moving picture block with a large motion or a still picture block with a small motion among fields in the block in every block obtained by the block forming means, orthogonal transformation means for transforming orthogonally on the basis of the orthogonal transform in each field when detected as moving picture block by the motion detecting means, or transforming orthogonally by combining plural fields as one still picture when detected as still picture block, coding means for coding the orthogonally transformed transmission sample value, and information transmission means for coding and transmitting the information showing whether detected as moving picture block or detected as still picture block by the motion detecting means.

Claim 6

The apparatus of claim 5, wherein sample values of video signals of consecutive two fields entered by the sample value decimating means are decimated of sample values in the different horizontal and vertical positions between fields and in every other sample value.

Claim 7

The apparatus of claim 5, wherein the decimating means determines by using the sample points existing between lines in which the transmission sample points belong to a same field, between lines adjacent on the screen and belonging to consecutive fields, or between lines on the same position on the screen and belonging to consecutive frames, and the positions of the sample points to be decimated differ between lines.

Claim 8

The apparatus of claim 5, wherein the orthogonal transformation means processes a moving picture by three-dimensional orthogonal transformation, and a still picture block by two-dimensional orthogonal transformation.

Claim 9

The apparatus of claim 5, wherein the orthogonal transformation means separates each block into blocks of every field and orthogonally transforms the operation result of the sum and difference in the horizontal and vertical directions between blocks when the input block is a moving picture block, and separates into blocks of frame unit and orthogonally transforms the operation result of sum and difference of vertical components between blocks in the case of a still picture block.

Claim 10

The apparatus of claim 5, wherein the transmission field composing means prepares two fields composed of transmission sample values of horizontal H and vertical V lines obtained by the decimating means, separates the two fields into V groups containing two lines between adjacent fields on the screen, sequentially arranges the transmission sample values of different fields in every transmission sample value in every one of V groups to compose the information of one line composed of 2H transmission sample values, thereby composing one transmission field of horizontal direction 2H and vertical direction V lines.

Claim 11

A high efficiency coding apparatus comprising decimating means for decimating sample values to obtain transmission sample points by using a filter utilizing only the sample points in the same field in the portion with a large motion of the input video signal, and using a filter utilizing sample values between plural fields or between plural frames in the portion with a small motion, block forming means for collecting a plurality of transmission sample points obtained by the decimating means over plural fields to form blocks, orthogonal transformation means for transforming orthogonally transmission sample points within the block obtained by the block forming means, and coding means for coding the orthogonally transformed transmission sample values. In the moving picture portion, sample values are decimated by using the filter utilizing the sample values in the same field, and in the still picture portion, sample values are decimated by using the filter utilizing the sample values between plural frames.

Claim 12

The apparatus of claim 11, wherein the decimating means determines by using the sample points existing between lines in which the transmission sample points belong to a same field, between lines adjacent on the screen and belonging to consecutive fields, or between lines on the same position on the screen and belonging to consecutive frames, and the positions of the sample points to be decimated differ between lines.

Claim 13

A high efficiency coding apparatus comprising block forming means for collecting a plurality of transmission sample points of the input video signal over plural fields to form blocks, motion detecting means for judging whether a moving picture block with a large motion or a still picture block with a small motion among fields in the block in every block obtained by the block forming means, decimating means for decimating sample values to obtain transmission sample points by using a filter utilizing only the sample points in the same field for the sample values in the block detected as a moving picture block by the motion detecting means, and using a filter utilizing sample values between plural fields or between plural frames for the sample values in the block detected as a still picture block, orthogonal transformation means for transforming orthogonally on the basis of the orthogonal transform in every plural fields when detected as a moving picture block by the motion detecting means, and transforming orthogonally by combining the plural fields as one still picture when detected as a still picture block, coding means for coding the orthogonally transformed transmission sample values, and information transmission means for coding and transmitting the information showing whether detected as a moving picture block or detected as a still picture block by the motion detecting means.

Claim 14

The apparatus of claim 13, wherein wherein the decimating means determines by using the sample points existing between lines in which the transmission sample points belong to a same field, between lines adjacent on the screen and belonging to consecutive fields, or between lines on the same position on the screen and belonging to consecutive frames, and the positions of the sample points to be decimated differ between lines.

Claim 15

The apparatus of claim 13, wherein the orthogonal transformation means processes a moving picture by three-dimensional orthogonal transformation, and a still picture block by two-dimensional orthogonal transformation.

Claim 16

The apparatus of claim 13, wherein the orthogonal transformation means separates each block into blocks of every field and orthogonally transforms the operation result of the sum and difference in the horizontal and vertical directions between blocks when the input block is a moving picture block, and separates into blocks of frame unit and orthogonally transforms the operation result of sum and difference of vertical components between blocks in the case of a still picture block.

Claim 17

A high efficiency coding apparatus comprising sample value decimating means for decimating sample values of input video signals in every other sample value to obtain transmission sample values, transmission field composing means for composing one field of transmission field from transmission sample values of consecutive two input fields obtained by the sample value decimating means, first block forming means for forming blocks from the transmission sample values contained in one field or plural fields of the transmission field obtained by the transmission field composing means, second block forming means for forming blocks containing the same sample value number as the first block forming means from the sample values contained in one field or plural fields of the input field, and orthogonal transformation means for orthogonally transforming in every block obtained by the first block forming means when high compression is required, and orthogonally transforming in every block obtained by the second block forming means when low compression is required.

Claim 18

A decoding apparatus for reproducing a compressed signal from a medium, by composing one field or consecutive two fields of transmission field by using a sub-sample filter, and forming an orthogonal transform block by using the one or plural transmission fields, which comprises a transformer for inversely and orthogonally transforming the reproduction signal from the medium, a memory for storing the output signal of the transformer and sending out one identical transmission field consecutively in two-field period, and a filter for interpolating a non-transmission sample value on the basis of the transmission sample value sent out from the memory.

Claim 19

The apparatus of claim 18, wherein the filter is realized by a filter in line for interpolating by using the transmitted sample values on the same line comprising the sample points to be interpolated.

Claim 20

A high efficiency coding apparatus comprising filter means for processing input video signals by band limiting and frequency converting, and decimating means for decimating the video signals limited in band by the filter means.

Claim 21

The apparatus of claim 20, wherein the filter means is a filter in which the input video signal is separated into four regions of low band, medium band, high band, and ultrahigh band in the horizontal direction on the two-dimensional frequency in the horizontal direction and vertical direction, and the medium band is further separated into two regions of medium band of vertical low band and and medium band of vertical high band, and the high band is separated into two regions of high band of vertical low band and high band of vertical high band, and at this time the components of low band, middle band of vertical low band and high band of vertical low band are taken out, and the components of the high band of vertical low band are converted in frequency to high band of vertical high band.

Claim 22

The apparatus of claim 20, wherein the filter means is composed of a first filter in which the input video signal is separated into four regions of low band, medium band, high band, and ultrahigh band in the horizontal direction on the two-dimensional frequency in the horizontal direction and vertical direction, and the medium band is further separated into two regions of medium band of vertical low band and and medium band of vertical high band, and the high band is separated into two regions of high band of vertical low band and high band of vertical high band, and at this time the components of low band and middle band of vertical low band are taken out as first components, a second filter for converting the components of the high band of vertical low band in frequency to high band of vertical high band to be taken out as second components, and an adder for adding the first components of the present field obtained in the first filter and the second components of the adjacent fields or adjacent frames obtained in the second filter to obtain the sum as the sample value of a new present field, and adding the second components of the present field obtained in the second filter and the first components of the adjacent fields or adjacent frames obtained by the first filter to obtain the sum as a new sample value of the adjacent fields or adjacent frames.