Optical EngineeringImage coding schemes based on subsampling vector quantization
|Format||Member Price||Non-Member Price|
|GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free.||Check Access|
Three variable-rate image-coding schemes are proposed in this paper. The goal of these schemes is to exploit variable-size segmentation of images using subsampling vector quantization. In the schemes, three simple block type classifiers are designed to decide the block activities of the image blocks. If two adjacent blocks are sufficiently similar in block activities, then they are encoded by subsampling vector quantization. Otherwise, the traditional vector quantization scheme is employed to encode each image block independently. According to the experimental results, the new schemes cannot only achieve significant reduction of bit rates, but also provide good subjective image quality. Therefore, the proposed schemes are suitable for the compression of digital images.