Image encoding apparatus and method with both lossy and lossless means
This invention is to generate encoded data within a target amount. A first encoding unit generates lossy encoded data of each pixel block by using a quantization matrix Q i specified by a parameter i. A second encoding unit generates lossless encoded data of each pixel block. Let Lx be the lossless encoded data length, and Ly be the lossy encoded data length. A control unit determines, using a nonlinear boundary function f i,j ( ) specified by the parameters i and j, whether condition: Ly<f i,j (Lx) is satisfied and stores the determination result as history information in a history memory unit. One of the two encoded data is stored in a memory based on the determination result. If the encoded data amount stored in the memory has exceeded the target amount, the control unit updates at least one of the encoding parameters i and j on the basis of the history information.
1. An image encoding apparatus for encoding each tile of an image, comprising:
at least one processor configured to function as:
a lossless encoding unit for losslessly encoding a tile of interest and outputting lossless encoded data;
a lossy encoding unit for lossy-encoding a tile of interest and outputting lossy encoded data;
a selection unit for selectively outputting one of the encoded data by comparing a code length of the lossless encoded data of the tile of interest with a code length of the lossy encoded data, and holding a total lossless encoded data amount MLS of the selectively outputted lossless encoded data and a total lossy encoded data amount MJP of the selectively outputted lossy encoded data; and
a decision unit for, when a total encoded data amount of the image has exceeded a target amount, deciding a re-encoding method for re-starting encoding of the image by comparing MLS with MJP.
2. The apparatus according to claim 1 , wherein said selection unit executes a weighted comparison of MLS and MJP to make it easy to select the lossless encoded data.
3. The apparatus according to claim 1 , wherein in re-encoding when the total encoded data amount has exceeded the target amount, and MLS is not less than MJP, said selection unit executes a weighted comparison of MLS and MJP to make it hard to select the lossless encoded data.
4. The apparatus according to claim 1 , wherein in re-encoding when the total encoded data amount has exceeded the target amount, and MLS is not less than MJP, said lossy encoding unit executes lossy encoding at a compression ratio higher than that of a previous time.
5. The apparatus according to claim 4 , wherein said lossy encoding unit changes the compression ratio by changing a quantization parameter.
6. The apparatus according to claim 1 , wherein decision unit decides a re-encoding method by executing a weighted comparison of MLS and MJP.
7. An image encoding method of encoding each tile of an image, comprising steps of:
losslessly encoding a tile of interest and outputting lossless encoded data;
lossy-encoding a tile of interest and outputting lossy encoded data;
selectively outputting one of the encoded data by comparing a code length of the lossless encoded data of the tile of interest with a code length of the lossy encoded data, and holding a total lossless encoded data amount MLS of the selectively outputted lossless encoded data and a total lossy encoded data amount MJP of the selectively outputted lossy encoded data; and
deciding a re-encoding method, when a total encoded data amount of the image has exceeded a target amount, for re-starting encoding of the image by comparing MLS with MJP.
8. The method according to claim 7 , wherein in the selecting step, a weighted comparison of MLS and MJP is executed to make it easy to select the lossless encoded data.
9. The method according to claim 7 , wherein in re-encoding when the total encoded data amount has exceeded the target amount, and MLS is not less than MJP, in the selecting step, a weighted comparison of MLS and MJP is executed to make it hard to select the lossless encoded data.
10. The method according to claim 7 , wherein in re-encoding when the total encoded data amount has exceeded the target amount, and MLS is not less than MJP, in the lossy encoding step, lossy encoding is executed at a compression ratio higher than that of a previous time.
11. The method according to claim 10 , wherein in the lossy encoding step, the compression ratio is changed by changing a quantization parameter.
12. A non-transitory computer-readable medium storing a computer program for controlling an information processing apparatus, which when executed by a processor causes said processor to perform the steps of an image encoding method according to claim 7 .