High-efficiency coding method, high-efficiency coding apparatus, recording and reproducing apparatus, and information transmission system
First Claim
1. A high-efficiency coding method comprising the steps of:
- a block-dividing step for dividing image information extracted from a predetermined area of a preceding frame to provide a plurality of reference block of a predetermined unit at every frame;
a difference absolute value sum calculation step for calculating a difference absolute value sum of pixel data located within said plurality of reference blocks and pixel data located within a remarkable block extracted from a present frame;
a first motion detection step for obtaining first motion vector data based on first difference absolute value sum data of minimum value from a plurality of difference absolute value sum data provided at every plurality of reference blocks;
a second motion detection step for obtaining second motion vector data based on second difference absolute value sum data of minimum value from a plurality of difference absolute value sum data provided at every plurality of reference blocks corresponding to one field located within said plurality of reference blocks;
a comparison step for comparing said first difference absolute value sum data and said second difference absolute value sum data;
a selection step for selecting said first motion vector data or said second motion vector data based on a compared result of said comparison step;
a motion compensation step for motion-compensating image information of said preceding frame based on said first motion vector data or said second motion vector data; and
an encoding step for effecting encoding by calculating a difference between said image information of present frame and said motion-compensated image information of preceding frame.
1 Assignment
0 Petitions
Accused Products
Abstract
A high-efficiency coding apparatus is provided which can increase accuracy in motion compensation and also increase processing speed without increasing the circuit scale as compared with the case that motion detection is carried out with respect to both of even and odd fields. Thus, this high-efficiency coding apparatus can perform accurate encoding, and can realize satisfactory transmission or recording of information. This high-efficiency coding apparatus is formed with frame memories for storing image information of a present frame therein, a frame memory for storing image information of a preceding frame therein, a frame type/2-field type motion detecting circuit for obtaining frame motion vector data of half-pel accuracy and 2-field motion vector data of half-pel accuracy by effecting motion detection of pixel data located within reference blocks located within the search area of a preceding frame and pixel data located within a remarkable block of the present frame, and selectively outputting these motion vector data. A motion compensating circuit effects motion compensation based on the selected output and an adding circuit effects encoding of the image information of the present frame and motion-compensated image information.
24 Citations
15 Claims
-
1. A high-efficiency coding method comprising the steps of:
-
a block-dividing step for dividing image information extracted from a predetermined area of a preceding frame to provide a plurality of reference block of a predetermined unit at every frame; a difference absolute value sum calculation step for calculating a difference absolute value sum of pixel data located within said plurality of reference blocks and pixel data located within a remarkable block extracted from a present frame; a first motion detection step for obtaining first motion vector data based on first difference absolute value sum data of minimum value from a plurality of difference absolute value sum data provided at every plurality of reference blocks; a second motion detection step for obtaining second motion vector data based on second difference absolute value sum data of minimum value from a plurality of difference absolute value sum data provided at every plurality of reference blocks corresponding to one field located within said plurality of reference blocks; a comparison step for comparing said first difference absolute value sum data and said second difference absolute value sum data; a selection step for selecting said first motion vector data or said second motion vector data based on a compared result of said comparison step; a motion compensation step for motion-compensating image information of said preceding frame based on said first motion vector data or said second motion vector data; and an encoding step for effecting encoding by calculating a difference between said image information of present frame and said motion-compensated image information of preceding frame.
-
-
2. A high-efficiency coding method comprising the steps of:
-
a block-dividing step for dividing image information extracted from a predetermined area of a preceding frame to provide a plurality of reference blocks of a predetermined unit at every frame; a first motion detection step for obtaining first motion vector data by effecting a first motion detection by use of pixel data located within said plurality of reference blocks and pixel data located within a remarkable block extracted from a present frame; a second motion detection step for obtaining second motion vector data by effecting a second motion detection by use of pixel data located within said plurality of reference blocks corresponding to one field provided within said plurality of reference blocks and pixel data located within said remarkable block extracted from said present frame; a first interpolation pixel data generation step for obtaining a plurality of interpolation reference blocks by generating interpolation pixel data between pixel data and pixel data within a predetermined area including pixel data located at positions in said preceding frame indicated by said first motion vector data; a first difference absolute value sum calculation step for calculating a first difference absolute value sum of pixel data located within said interpolation reference blocks and pixel data located within said remarkable block extracted from said present frame; a third motion detection step for obtaining third motion vector data based on first difference absolute value sum data of minimum value from a plurality of difference absolute value sum data located at every plurality of interpolation reference blocks; a second interpolation pixel data generation step for obtaining a plurality of interpolation reference blocks by generating interpolation pixel data between pixel data and pixel data within a predetermined area of one field including pixel data located at positions in said preceding frame indicated by said second motion vector data; a second difference absolute value sum calculation step for calculating a second difference absolute value sum of pixel data located within said interpolation reference blocks and pixel data located within said remarkable block extracted from said present frame; a fourth motion detection step for obtaining fourth motion vector data based on second absolute value sum data of minimum value from a plurality of difference absolute value sum data provided at every plurality of interpolation reference blocks; a comparison step for comparing said first difference absolute value sum data and said second difference absolute value sum data; a selection step for selecting said third motion vector data or said fourth motion vector data based on a compared result of said comparison step; a motion-compensation step for motion-compensating said image information of preceding frame based on said third motion vector data or said fourth motion vector data; and an encoding step for effecting encoding by calculating a difference between said image information of present frame and said motion-compensated image information of preceding frame.
-
-
3. A high-efficiency coding method comprising the steps of:
-
a block-dividing step for dividing image information extracted from a predetermined area of preceding frame to provide a plurality of reference blocks of a predetermined unit at every frame; a first motion detection step for obtaining first motion vector data by effecting a first motion detection by use of pixel data located within said plurality of reference blocks and pixel data located within a remarkable block extracted from a present frame; a second motion detection step for obtaining second motion vector data by effecting a second motion detection by use of pixel data located within a plurality of reference blocks corresponding to one field of said plurality of reference blocks and pixel data located within said remarkable block extracted from said present frame; a first interpolation pixel data generation step for obtaining a plurality of interpolation reference blocks by generating interpolation pixel data between pixel data and pixel data within a predetermined area including pixel data located at positions in said preceding frame indicated by said first motion vector data; a first difference absolute value sum calculation step for calculating a first difference absolute value sum calculation by use of pixel data located within said interpolation reference blocks and pixel data located within said remarkable block extracted from said present frame; a third motion detection step for obtaining third motion vector data based on first difference absolute value sum data of minimum value from a plurality of difference absolute value sum data located at every plurality of interpolation reference blocks; a second interpolation pixel data generation step for obtaining a plurality of interpolation reference blocks by generating interpolation pixel data between pixel data and pixel data within a predetermined area of one field including pixel data located positions in said preceding frame indicated by said second motion vector data; a second difference absolute value sum calculation step for effecting a second difference absolute value sum calculation by use of a part of pixel data located within said interpolation reference blocks and a part of pixel data located within said remarkable block extracted from said present frame; a fourth motion detection step for obtaining fourth motion vector data based on second difference absolute value sum data of minimum value from a plurality of difference absolute value sum data located at every plurality of interpolation reference blocks; a comparison step for comparing said first difference absolute value sum data and said second difference absolute value sum data; a selection step for selecting said third motion vector data or said fourth motion vector data based on a compared result of said comparison step; a motion compensation step for motion-compensating said image information of preceding frame based on said third motion vector data or said fourth motion vector data; and an encoding step for effecting encoding by calculating a difference between said image information of present frame and said image information of preceding frame.
-
-
4. A high-efficiency coding apparatus comprising:
-
first memory means for storing image information of present frame to be encoded therein; second memory means for storing image information of preceding frame therein; first motion detecting means for obtaining first motion vector data by effecting a motion detection by use of pixel data located within reference blocks sequentially set within a predetermined range of said image information of preceding frame stored in said second memory means and pixel data located within a remarkable block in said image information of present frame stored in said first memory means; second motion detecting means for obtaining second motion vector data by effecting a motion detection by use of pixel data located within reference blocks corresponding to one field of reference blocks sequentially set within a predetermined range of said image information of preceding frame stored in said second memory means and pixel data located within said remarkable block of said image information of present frame stored in said first memory means; third motion detecting means for generating interpolation pixel data between pixel data and pixel data within a predetermined area including pixel data in said image information of preceding frame indicated by said first motion vector data supplied thereto from said first motion detecting means, calculating difference absolute value sum of interpolation pixel data located within a plurality of interpolation reference blocks composed of said interpolation pixel data and pixel data located within said remarkable block, and obtaining third motion vector data based on first difference absolute value sum data of minimum value from a plurality of difference absolute value sum data obtained by said difference absolute value sum calculation; fourth motion detecting means for generating interpolation pixel data between pixel data and pixel data within a predetermined area of said one field including pixel data located in said image information of preceding frame indicated by said second motion vector data from said second motion detecting means, calculating difference absolute value sum of interpolation pixel data located within a plurality of interpolation reference blocks composed of said interpolation pixel data and pixel data located within said remarkable block, and obtaining fourth motion vector data based on second difference absolute value sum data of minimum value from a plurality of difference absolute value sum data obtained by said difference absolute value sum calculation; selecting means for selecting said third motion vector data supplied thereto from said third motion detecting means or said fourth motion vector data supplied thereto from said fourth motion detecting means based on a compared result of said first difference absolute value sum data supplied thereto from said second motion detecting means and said second difference absolute value sum data supplied thereto from said third motion detecting means; motion compensating means for motion-compensating said image information of preceding frame based on said third motion vector data or said fourth motion vector data supplied thereto from said selecting means; and encoding means for effecting encoding based on said motion-compensated image information of preceding frame supplied thereto from said motion compensating means and said image information of present frame stored in said memory means.
-
-
5. A high-efficiency coding apparatus comprising:
-
first memory means for storing image information of present frame to be encoded therein; second memory means for storing image information of preceding frame therein; first difference absolute value sum calculating means for calculating difference absolute value sum of pixel data located within reference blocks sequentially set within a predetermined range of said image information of preceding frame stored in said second memory means and pixel data located within a remarkable block in said image information of present frame stored in said first memory means; first minimum value selecting means for obtaining first motion vector data based on first difference absolute value sum data of minimum value from a plurality of difference absolute value sum data supplied thereto from said first difference absolute value sum calculating means; second minimum value selecting means for obtaining second motion vector data based on second difference absolute value sum data of minimum value of difference absolute value sum data which result from calculating difference absolute value sum of pixel data located within a reference block corresponding to one field corresponding to reference blocks sequentially set within a predetermined range of said image information of preceding frame stored in said second memory means and pixel data located within said remarkable block in said image information of present frame of a plurality of difference absolute value sum data supplied thereto from said first difference absolute value sum calculating means; first interpolation pixel generating means for generating interpolation pixel data between pixel data and pixel data within a predetermined area including pixel data in said image information of preceding frame indicated by said first motion vector data supplied thereto from said first minimum value selecting means; second difference absolute value sum calculating means for calculating difference absolute value sum of interpolation pixel data located within a plurality of interpolation reference blocks composed of interpolation pixel data supplied thereto from said first interpolation pixel generating means and pixel data located within said remarkable block; third minimum value selecting means for obtaining third motion vector data based on third difference absolute value sum data of minimum value from a plurality of difference absolute value sum data supplied thereto from said second difference absolute value sum calculating means; second interpolation pixel generating means for generating interpolation pixel data between pixel data and pixel data within a predetermined area including pixel data located in said image information of preceding frame indicated by said second vector data supplied thereto from said second minimum value selecting means; third difference absolute value sum calculating means for calculating difference absolute value sum of interpolation pixel data located within a plurality of interpolation reference blocks composed of interpolation pixel data supplied thereto from said second interpolation pixel generating means; fourth minimum value selecting means for obtaining fourth motion vector data based on fourth difference absolute value sum data of minimum value from a plurality of difference absolute value sum data obtained by said third difference absolute value calculating means; comparing means for comparing said third difference absolute value sum data supplied thereto from said third minimum value selecting means and said fourth difference absolute value sum data supplied thereto from said fourth minimum value selecting means; selecting means for selecting said third motion vector data supplied thereto from said third minimum value selecting means or said fourth motion vector data supplied thereto from said minimum value selecting means based on a compared result of said comparing means; motion compensating means for motion-compensating said image information of preceding frame based on said third motion vector data or said fourth motion vector data supplied thereto from said selecting means; and encoding means for effecting encoding based on said motion-compensated image information of preceding frame supplied thereto from said motion compensating means and image information of present frame stored in said first memory means.
-
-
6. A high-efficiency coding apparatus comprising:
-
first memory means for storing image information of present frame to be encoded therein; second memory means for storing image information of preceding frame; first difference absolute value sum calculating means for calculating difference absolute value sum of pixel data located within reference blocks sequentially set within a predetermined range of said image information of preceding frame stored in said second memory means, and pixel data located within a remarkable block in said image information of preceding frame stored in said memory means; first minimum value selecting means for obtaining first integer pixel accuracy motion vector data based on first difference absolute value sum data of minimum value from a plurality of difference absolute value sum data supplied thereto from said difference absolute value sum calculating means; second minimum value selecting means for obtaining second integer pixel accuracy motion vector data based on second difference absolute value sum data of minimum value in difference absolute value sum data which result from calculating difference absolute value sum of pixel data located within reference block corresponding to one field located within reference blocks sequentially set within a predetermined range in said image information of preceding frame stored in said second memory means and pixel data located within the remarkable block of said image information of present frame stored in said first memory means from a plurality of difference absolute value sum data supplied thereto from said first difference absolute value sum calculating means; first interpolation pixel generating means for generating interpolation pixel data between pixel data and pixel data within a predetermined area including pixel data in said image information of preceding frame indicated by said first integer pixel accuracy motion vector data supplied thereto from said first minimum value selecting means; first motion detecting means for calculating difference absolute value sum of interpolation pixel data located within sub-interpolation reference block smaller than a plurality of interpolation reference block composed of interpolation pixel data from said first interpolation pixel generating means and pixel data located within sub-remarkable block smaller than said remarkable block, and obtaining first half-pel accuracy motion vector data baaed on third difference absolute value sum data of minimum value of a plurality of difference absolute value sum data obtained by said difference absolute value sum calculation; second interpolation pixel generating means for generating interpolation pixel data between pixel data and pixel data within a predetermined area including pixel data located in said image information of preceding frame indicated by said second integer pixel accuracy motion vector data supplied thereto from said second minimum value selecting means; second motion detecting means for calculating difference absolute value sum of interpolation pixel data located within a sub-interpolation reference block smaller than a plurality said interpolation reference blocks supplied thereto from said second interpolation pixel generating means and pixel data located within a sub-remarkable block smaller than said remarkable block, and obtaining second half-pel accuracy motion vector data based on fourth difference absolute value sum data of minimum value from a plurality of difference absolute value sum data obtained by said difference absolute value sum calculation; comparing means for comparing said third difference absolute value sum data supplied thereto from said first motion detecting means and said fourth difference absolute value sum data supplied thereto from said second motion detecting means; selecting means for selecting said first half-pel accuracy motion vector data from said first motion detecting circuit or said second half-pel accuracy motion vector data supplied thereto from said second motion detecting means based on a compared result of said comparing means; motion compensating means for motion-compensating said image information of preceding frame based on said half-pel accuracy motion vector or said second half-pel accuracy motion vector data supplied thereto from said selecting means; and encoding means for effecting encoding based on said motion-compensated image information of preceding frame supplied thereto from said motion compensating means and said image information of present frame stored in said first memory means. - View Dependent Claims (7)
-
-
8. A recording and reproducing apparatus comprising:
-
first memory means for storing image information of present frame to be encoded therein; second memory means for storing image information of preceding frame therein; first motion detecting means for obtaining first motion vector data by effecting a motion detection by use of pixel data located within reference blocks sequentially set within a predetermined range in said image information of preceding frame stored in said second memory means, and pixel data located within a remarkable block in said image information of preceding frame stored in said first memory means; second motion detecting means for obtaining second motion vector data by effecting a motion detection by use of pixel data located within reference block corresponding to one field of reference blocks sequentially set within a predetermined range in said image information of preceding frame stored in said second memory means, and pixel data located within a remarkable block in said image information of present frame stored in said first memory means; third motion detecting means for generating interpolation pixel data between pixel data and pixel data within a predetermined area including pixel data located in said image information of preceding frame indicated by said first motion vector data supplied thereto from said first motion detecting means, calculating difference absolute value sum of interpolation pixel data located within a plurality of interpolation reference blocks composed of said interpolation pixel data and pixel data located within said remarkable block, and obtaining third motion vector data based on first difference absolute value sum data of minimum value from a plurality of difference absolute value sum data obtained by said difference absolute value sum calculation; fourth motion detecting means for generating interpolation pixel data between pixel data and pixel data within a predetermined area of one field including pixel data located in said image information of preceding frame indicated by said second motion vector data supplied thereto from said first motion detecting means, calculating difference absolute value sum of interpolation pixel data located within a plurality of interpolation reference blocks composed of said interpolation pixel data, and pixel data located within said remarkable block, and obtaining fourth motion vector data based on second difference absolute value sum data of minimum value from a plurality of difference absolute sum value data obtained by said difference absolute value sum calculation; selecting means for selecting said third motion vector data supplied thereto from said third motion detecting means or said fourth motion vector data supplied thereto from said fourth motion detecting means based on a compared result of said second difference absolute value sum data supplied thereto from said second motion detecting means, and said third difference absolute value sum data supplied thereto from said third motion detecting means; motion compensating means for motion-compensating said image information of preceding frame based on said third motion vector data or said fourth motion vector data supplied thereto from said selecting means; encoding means for effecting encoding based on said motion-compensated image information of preceding frame supplied thereto from said motion compensating means, and said image information of present frame stored in said first memory means; recording means for recording an output from said encoding means on a recording medium; reproducing means for reproducing recording information recorded on said recording medium; decoding means for decoding a reproduced output from said reproducing means; and reproduced signal processing means for reproduced-signal-processing an output supplied thereto from said decoding means.
-
-
9. A recording and reproducing apparatus comprising:
-
recording signal processing means for processing image information to be recorded for recording; first memory means for storing image information of present frame to be encoded therein; second memory means for storing image information of preceding frame therein; first difference absolute value sum calculating means for calculating difference absolute value sum of pixel data located within reference blocks sequentially set within a predetermined range in said image information of preceding frame stored in said second memory means, and pixel data located within a remarkable block in said image information of present frame stored in said first memory means; first minimum value selecting means for obtaining first motion vector data based on first difference absolute value sum data of minimum value from a plurality of difference absolute value sum data supplied thereto from said difference absolute value sum calculating means; second minimum value selecting means for obtaining second motion vector data based on second difference absolute value sum data of minimum value of difference absolute value sum data which result from calculating difference absolute value sum of pixel data located within reference block corresponding to one field of reference blocks sequentially set within a predetermined range of said image information of preceding frame stored in said second memory means and pixel data located within a remarkable block in said image information of present frame stored in said memory means of a plurality of difference absolute value sum data supplied thereto from said first difference absolute value sum calculating means; first interpolation pixel generating means for generating interpolation pixel data between pixel data and pixel data within a predetermined area including pixel data in said image information of preceding frame indicated by said first motion vector data supplied thereto from said first minimum value selecting means; second difference absolute value sum calculating means for calculating difference absolute value sum of interpolation pixel data located within a plurality of interpolation reference blocks composed of interpolation pixel data supplied thereto from said first interpolation pixel generating means, and pixel data located within said remarkable block; third minimum value selecting means for obtaining third motion vector data based on third difference absolute value sum data of minimum value of a plurality of difference absolute value sum data supplied thereto from said second difference absolute value sum calculating means; second interpolation pixel generating means for generating interpolation pixel data between pixel data and pixel data within a predetermined area of one field including pixel data in said image information of preceding frame indicated by said second motion vector data supplied from said second minimum value selecting means; third difference absolute value sum calculating means for calculating difference absolute value sum of interpolation pixel data located within a plurality of interpolation pixel data supplied thereto from said second interpolation pixel generating means and pixel data located within said remarkable block; fourth minimum value selecting means for obtaining fourth motion vector data based on fourth difference absolute value sum data of minimum value from a plurality of difference absolute value sum data obtained by said third difference absolute value calculating means; comparing means for comparing said third difference absolute value sum data supplied thereto from said third minimum value selecting means, and said fourth difference absolute value sum data supplied thereto from said fourth minimum value selecting means; selecting means for selecting said third motion vector data supplied thereto from said third minimum value selecting means and said fourth motion vector data supplied thereto from said fourth minimum value selecting means based on a compared result from said comparing means; motion compensating means for motion-compensating said preceding frame image information based on said third motion vector data or said fourth motion vector data supplied thereto from said selecting means; encoding means for effecting encoding based on said motion-compensated preceding frame image information supplied thereto from said motion compensating means and said present frame image information stored in said first memory means; recording means for recording an output from said encoding means on a recording medium; reproducing means for reproducing recorded information from said recording medium; decoding means for decoding a reproduced output supplied thereto from said reproducing means; and reproduced signal processing means for reproduced-signal-processing an output supplied thereto from said decoding means.
-
-
10. A recording and reproducing apparatus comprising:
-
recording-signal-processing means for processing recorded image information for recording; first memory means for storing encoded image information of present frame; second memory means for storing image information of preceding frame; first difference value sum calculating means for calculating difference absolute value sum of pixel data located within reference blocks sequentially set within a predetermined range of image information of preceding frame stored in said second memory means, and pixel data located within a remarkable block of said image information of preceding frame stored in said first memory means; first minimum value selecting means for obtaining motion vector data of first integer pixel accuracy based on first difference absolute value sum data of smallest value of a plurality of difference absolute value sum data supplied thereto from said difference absolute value sum calculating means; second minimum value selecting means for obtaining motion vector data of second integer pixel accuracy based on second difference absolute value sum data of smallest value of difference absolute value sum data which result from calculating difference absolute value sum of pixel data located within a reference block corresponding to one field of reference blocks sequentially set within a predetermined range in image information of preceding frame stored in said second memory means and pixel data located within a remarkable block in said image information of present frame stored in said first memory means of a plurality of difference absolute value sum data supplied thereto from said first difference absolute value sum calculating means; first interpolation pixel generating means for generating interpolation pixel data between pixel data and pixel data within a predetermined area including pixel data in said image information of preceding frame indicated by said motion vector data of first integer pixel accuracy supplied thereto from said first minimum value selecting means; first motion detecting means for calculating difference absolute value sum of interpolation pixel data located within a sub-interpolation reference block smaller than said plurality of interpolation reference blocks composed of said interpolation pixel data supplied thereto from said first interpolation pixel generating means, and pixel data located within a sub-remarkable block smaller than said remarkable block, and obtaining motion vector data of first half-pel accuracy based on third difference absolute value sum data of smallest value of a plurality of difference absolute value sum data obtained by said difference absolute value sum calculation; second interpolation pixel generating means for generating interpolation pixel data between pixel data and pixel data within a predetermined area of one field including pixel data located in said image information of preceding frame indicated by said motion vector data of second integer pixel accuracy supplied thereto from said second minimum value selecting means; second motion detecting means for calculating difference absolute value sum of interpolation pixel data located within a sub-interpolation reference block smaller than said plurality of interpolation reference blocks composed of interpolation pixel data supplied thereto from said second interpolation pixel generating means and pixel data located within a sub-remarkable block smaller than said remarkable block, and obtaining motion vector data of second half-pel accuracy based on fourth difference absolute value sum data of smallest value of said plurality of difference absolute value sum data obtained by said difference absolute value sum calculation; comparing means for comparing said third difference absolute value sum data supplied thereto from said first motion detecting means and said fourth difference absolute value sum data supplied thereto from said second motion detecting means; selecting means for selecting said motion vector data of first half-pel accuracy supplied thereto from said first motion detecting means or said motion vector data of second half-pel accuracy supplied thereto from said second motion detecting means based on a compared result from said comparing means; motion compensating means for motion-compensating said image information of preceding frame based on said third motion vector data or said fourth motion vector data supplied thereto from said selecting means; encoding means for effecting encoding based on said motion-compensated image information of preceding frame supplied thereto from said motion compensating means and said image information of present frame stored in said first memory means; recording means for recording an output from said encoding means on a recording medium; reproducing means for reproducing recorded information from said recording medium; decoding means for decoding a reproduced output supplied thereto from said reproducing means; and reproduced-signal-processing means for reproduced-signal-processing an output supplied thereto from said decoding means. - View Dependent Claims (11)
-
-
12. An information transmission system comprising:
-
first memory means for storing encoded image information of present frame therein; second memory means for storing image information of preceding frame therein; first motion detecting means for obtaining first motion vector data by effecting a motion detection of pixel data located within reference blocks sequentially set within a predetermined range of said image information of preceding frame stored in said second memory means, and pixel data located within a remarkable block in said image information of present frame stored in said first memory means; second motion detecting means for obtaining second motion vector data by effecting a motion detection of pixel data located within a reference block corresponding to one field located within reference blocks sequentially set within a predetermined range of said image information of preceding frame stored in said second memory means, and pixel data located within a remarkable block in said image information of present frame stored in said first memory means; third motion detecting means for generating interpolation pixel data between pixel data and pixel data within a predetermined area including pixel data in said image information of preceding frame indicated by said first motion vector data supplied thereto from said first motion detecting means, calculating difference absolute value sum of interpolation pixel data located within a plurality of interpolation reference blocks, and obtaining third motion vector data based on first difference absolute value sum data of smallest value of a plurality of difference absolute value sum data obtained by said difference absolute value sum calculation; fourth motion detecting means for generating interpolation pixel data between pixel data and pixel data located within a predetermined area of one field including pixel data in said image information of preceding frame indicated by said second motion vector data supplied thereto from said first motion detecting means, calculating difference absolute value sum of interpolation pixel data located within a plurality of reference blocks composed of said interpolation pixel data and pixel data located within said remarkable block, and obtaining fourth motion vector data based on second difference absolute value sum data of smallest value of a plurality of difference absolute value sum data obtained by said difference absolute value sum calculation; selecting means for selecting said third motion vector data supplied thereto from said third motion detecting means or said fourth motion vector data supplied thereto from said fourth motion detecting means based on a compared result of said second difference absolute value sum data supplied thereto from said second motion detecting means and said third difference absolute value sum data supplied thereto from said third motion detecting means; motion compensating means for motion-compensating said image information of preceding frame based on aid third motion vector data or said fourth motion vector data supplied thereto from said selecting means; encoding means for effecting encoding based on said motion-compensated image information of preceding frame supplied thereto from said motion compensating means, and said image information of present frame stored in said first memory means; transmitting means for transmitting an output supplied thereto from said encoding means; and decoding means for decoding said information transmitted.
-
-
13. An information transmission system comprising:
-
first memory means for storing encoded image information of present frame therein; second memory means for storing image information of preceding frame therein; first difference absolute value calculating means for calculating difference absolute value sum of pixel data located within reference blocks sequentially set within a predetermined range of said image information of preceding frame stored in said second memory means, and pixel data located within a remarkable block in said image information of present frame; first minimum value selecting means for obtaining first motion vector data based on first difference absolute value sum data of smallest value of a plurality of difference absolute value sum data supplied thereto from said difference absolute value sum calculating means; second minimum value selecting means for obtaining second motion vector data based on second difference absolute value sum data of smallest value located within difference absolute value sum data which result from calculating difference absolute value sum of pixel data located within a reference block corresponding to one field of reference blocks sequentially set within a predetermined range of said image information of preceding frame stored in said second memory means, and pixel data located within said remarkable block in said image information of present frame stored in said first memory means of a plurality of difference absolute value sum data supplied thereto from said first difference absolute value sum calculating means; first interpolation pixel generating means for generating interpolation pixel data between pixel data and pixel data within a predetermined area including pixel data located in said image information of preceding frame indicated by said first motion vector data supplied thereto from said first minimum value selecting means; second difference absolute value sum calculating means for calculating difference absolute value sum calculation of interpolation pixel data located with a plurality of interpolation reference blocks composed of said interpolation pixel data supplied thereto from said first interpolation pixel generating means, and pixel data located within said remarkable block; third minimum value selecting means for obtaining third motion vector data based on third difference absolute value sum data of smallest value of a plurality of difference absolute value sum data supplied thereto from said second difference absolute value sum calculating means; second interpolation pixel generating means for generating interpolation pixel data between pixel data and pixel data within a predetermined area of one field including pixel data located in said image information of preceding frame indicated by said second motion vector data supplied thereto from said second minimum value selecting means; third difference absolute value sum calculating means for calculating difference absolute value sum of interpolation pixel data located within a plurality of interpolation reference blocks composed of interpolation pixel data supplied thereto from said second interpolation pixel generating means, and pixel data located within said remarkable block; fourth minimum value selecting means for obtaining fourth motion vector data based on fourth difference absolute value sum data of smallest value of a plurality of difference absolute value sum data obtained by said third difference absolute value sum calculating means; comparing means for comparing said third difference absolute value sum data supplied thereto from said third minimum value selecting means and said fourth difference absolute value sum data supplied thereto from said fourth minimum value selecting means; selecting means for selecting said third motion vector data supplied thereto from said third minimum value selecting means or said fourth motion vector data supplied thereto from said fourth minimum value selecting means; motion compensating means for motion-compensating said image information of preceding frame based on said third motion vector data or said fourth motion vector data supplied thereto from said selecting means; encoding means for effecting encoding based on said motion-compensated image information of preceding frame supplied thereto from said motion compensating means and said image information of present frame stored in said first memory means; transmitting means for transmitting an output supplied thereto from said encoding means; and decoding means for decoding transmitted information.
-
-
14. An information transmission system comprising:
-
first memory means for storing image information of present frame to be encoded therein; second memory means for storing image information of preceding frame therein; first difference absolute value sum calculating means for calculating difference absolute value sum of pixel data located within reference blocks sequentially set within a predetermined range of said image information of preceding frame stored in said second memory means; first minimum value selecting means for obtaining motion vector data of first integer pixel accuracy based on first difference absolute value sum data of smallest value of a plurality of difference absolute value sum data from said difference absolute value sum calculating means; second minimum value selecting means for obtaining motion vector data of second integer pixel accuracy based on second difference absolute value sum of smallest value of difference absolute value sum data which results from calculating pixel data located within a reference block corresponding to one field located within reference blocks sequentially set in a predetermined range of said image information of preceding frame stored in said second memory means and pixel data located within a remarkable block in said image information of present frame stored in said first memory means; first interpolation pixel generating means for generating interpolation pixel data between pixel data and pixel data within a predetermined area including pixel data located in said image information of preceding frame indicated by said motion vector data of first integer pixel accuracy supplied thereto from said first minimum value selecting means; third motion detecting means for calculating difference absolute value sum of interpolation pixel data located within a sub-interpolation reference block smaller than a plurality of interpolation reference blocks composed of interpolation pixel data supplied thereto from said first interpolation pixel generating means and pixel data located within a sub-remarkable block smaller than the remarkable block, and obtaining motion vector data of first half-pel accuracy based on third difference absolute value sum data of smallest value of a plurality of difference absolute value sum data obtained by difference absolute value sum calculation; second interpolation pixel generating means for generating interpolation pixel data between pixel data and pixel data within a predetermined area of one field including pixel data located in said image information of preceding frame indicated by said motion vector data of second integer pixel accuracy supplied thereto from said second minimum value selecting means; second motion detecting means for calculating difference absolute value sum of interpolation pixel data located within a sub-interpolation reference block smaller than a plurality of interpolation reference blocks composed of interpolation pixel data supplied from said second interpolation pixel generating means and pixel data located within a sub-remarkable block smaller than the remarkable block, and obtaining motion vector data of second half-pel accuracy based on fourth difference absolute value sum data of smallest value in a plurality of difference absolute value sum data obtained by said difference absolute value sum data; comparing means for comparing said third difference absolute value sum data supplied thereto from said first motion detecting means and said fourth difference absolute value sum data supplied thereto from said second motion detecting means; selecting means for selecting said motion vector data of first half-pel accuracy supplied thereto from said first motion detecting means and said motion vector data of second half-pel accuracy supplied thereto from said second motion detecting means based on a compared result from said comparing means; motion compensating means for motion-compensating said image information of preceding frame based on said third motion vector data or said fourth motion vector data supplied thereto from said selecting means; encoding means for effecting encoding based on said motion-compensated image information of preceding frame supplied thereto from said motion compensating means and said image information of present frame stored in said first memory means; transmitting means for transmitting an output supplied thereto from said encoding means; and decoding means for decoding transmitted information. - View Dependent Claims (15)
-
Specification