Screen content coding using code repository for compound image compression

Compound Images are stacked with text, graphics and continuous tone contents. Screen Content Coding (SCC) improves the compression quality of images with multilayer of graphics and text that are generated by digital devices rather than camera-captured images, since these images possess repetitive patterns. We proposed an Intra-Frame Dictionary Coding based on hashing technique without modifying the HEVC syntaxes. In our proposed system the current blocks are mapped with related blocks holding the same hash value. These blocks are considered as prediction blocks and the best block is selected from the prediction blocks and encoded in a repository. Further the proposed coding efficiency can be enhanced by employing the Rate Distortion Optimization. The proposed system results when compared with the existing system displays a better performance of 95% in terms of compression ratio. The proposed algorithm is coded in MATLAB and the results were analyzed.