Transmission method for video or moving pictures by compressing block differences
AssigneeTelecom Finland Oy
InventorDan Björklund
Life of this patent
Estimating the term…
Abstract
A method is disclosed for transmitting video images over a data communication network, in which method the video image is encoded at the sending end, transmitted, and decoded at the receiving end. The first frame is compressed and transmitted to the receiving end where it is stored in image memory. A copy of the first frame is retained at the transmitting end and divided into blocks. The next frame is also divided into blocks. The blocks of the first frame and the blocks of the next frame are compared to determine a block difference for each block. For each block, the difference between successive blocks is compressed and transmitted if the block difference for the block exceeds a predetermined threshold.
Claims7
A method for transmitting video stream comprising temporally successive frames from a transmitting end to a receiving end in a data communication network, comprising:
- a) storing a current frame in a reference memory provided at the transmitting end;
- b) compressing said current frame;
- c) transmitting the compressed current frame to the receiving end;
- d) decompressing the transmitted current frame at the receiving end;
- e) storing the decompressed current frame in an image memory provided at the receiving end;
- f) dividing a first comparison frame into blocks, said first comparison frame being the frame currently stored in the reference memory;
- g) dividing a second comparison frame into blocks, said second comparison frame being the frame temporally subsequent to the frame currently stored in the reference memory;
- h) comparing the luminance components of a predetermined number of spatially corresponding pixels of a first comparison block and a second comparison block, said first comparison block being a block of the first comparison frame and said second comparison block being a block of the second comparison frame spatially corresponding to the first comparison block;
- i) determining a block comparison parameter equal to the number of said compared luminance components of spatially corresponding pixels the difference between which exceeds a predetermined pixel component threshold;
- j) determining a block subtraction value by subtracting each pixel of the first comparison block from a spatially corresponding pixel of the second comparison block, if the determined block comparison parameter exceeds a predetermined parameter threshold;
- k) compressing the determined block subtraction value, if the determined block comparison parameter exceeds a predetermined parameter threshold;
- l) transmitting the compressed block subtraction value to the receiving end, if the determined block comparison parameter exceeds a predetermined parameter threshold;
- m) transmitting block position data to the receiving end indicating the spatial position of the second comparison block, if the determined block comparison parameter exceeds a predetermined parameter threshold;
- n) decompressing the transmitted block subtraction value at the receiving end, if the determined block comparison parameter exceeds a predetermined parameter threshold;
- o) integrating the decompressed block subtraction value into the frame stored in the image memory using the transmitted block position data, if the determined block comparison parameter exceeds a predetermined parameter threshold;
- p) storing the second comparison block in the reference memory;
- q) repeating steps h-p with another pair of comparison blocks spatially subsequent to the previously processed pair of comparison blocks until each block of the first comparison frame and the second comparison frame has been processed;
- r) transmitting a predetermined part of the second comparison frame to the receiving end and integrating said part into the frame stored in the image memory;
- and
- s) repeating steps f-r with another frame of the video stream temporally subsequent to the previously processed frame.
Dependencies are inferred from the claim text.