Image processing system
AssigneeSeiko Epson Corporation
InventorsTsuyoshi Tamura, Toshiki Miyane, Teruhisa Ishikawa
Life of this patent
Estimating the term…
Loading patent…
AssigneeSeiko Epson Corporation
InventorsTsuyoshi Tamura, Toshiki Miyane, Teruhisa Ishikawa
Estimating the term…
The present invention provides an image decoding and reconstruction system for successively reading out and decoding compression video data constituting one screen page and reconstructing an image with a memory of a small capacity. The system of the invention divides an image of one screen page into 8x8-pixel blocks, compresses video data by discrete cosine transform, successively reads data in the unit of a 2×2 matrix of the 8×8-pixel blocks, and stores the video data along horizontal scanning lines in one of two memories while reading other video data previously stored along the scanning lines out of the other memory. The invention also provides an interblock distortion removal filter for eliminating interblock distortions due to orthogonal transform. The filter eliminates interblock distortions generated on the boundary between vertically adjacent blocks in each 16-line data area before writing video data in the memory, and eliminates distortions between horizontally adjacent blocks in the data area when reading video data out of the memory. The filter also eliminates interblock distortions on the boundary between adjacent 16-line data areas by reading last line data of a first block out of one memory while reading first line data of a second block out of the other memory.
An interblock distortion removal filter used in a process of decoding compression data to video data, said compression data being generated by dividing original video data constituting one screen page into a plurality of M×N-pixel blocks and coding said video data by each block according to an irreversible coding method, M and N representing integers greater than one but less than the number of pixels along the least pixel dimension of said screen page, said filter comprising:
An interblock distortion removal filter used in a process of decoding compression data to video data, said compression data being generated by dividing original video data constituting one screen page into a plurality of M×N-pixel blocks and coding said video data by each block according to an irreversible coding method, M and N representing integers greater than one but less than the number of pixels along the least pixel dimension of said screen page;
An interblock distortion removal filter used in a process of decoding compression data to video data, said compression data being generated by dividing original video data constituting one screen page into a plurality of M×N-pixel blocks and coding said video data by each block according to an irreversible coding method, M and N representing integers greater than one but less than the number of pixels along the least pixel dimension of said screen page;
A method of decoding compression data to video data, said compression data being generated by dividing original video data constituting one screen page into a plurality of M×N-pixel blocks and coding said video data by each block according to an irreversible coding method, said method comprising the steps of:
An interblock distortion removal filter used in a process of decoding compression data to video data, said compression data being generated by dividing original video data constituting one screen page into a plurality of M×N-pixel blocks and coding said video data by each block according to an irreversible coding method, M and N representing integers greater than one but less than the number of pixels along the least pixel dimension of said screen page, said filter comprising:
A method for removing interblock distortion in a process of decoding compression data to video data, said compression data being generated by dividing original video data constituting one screen page into a plurality of M×N-pixel blocks and coding said video data by each block according to an irreversible coding method, M and N representing integers greater that one but less than the number of pixels along the least pixel dimension of said screen page, said method comprising the steps of:
Dependencies are inferred from the claim text.