Enhanced Iteration-Free Fractal Image Coding Algorithm With Efficient Search and Storage Space

An enhanced iteration free fractal algorithm is proposed in this research paper to design an efficient domain pool for image compression. The proposed methodology reduces the coding process time, intensive computation tasks and also the memory requirements. The redundancies in the domain pool are reduced by the Linde Buzo Gray (LBG) Algorithm. For each range block, vector features such as mean value, edge strength, and texture strength are used to delete the irrelevant domain block. A pruning condition for terminating the searching process to find the best domain block from the domain pool is used. The codes are stored efficiently by comparing the values of the previous coded range blocks. The performance of the proposed method is compared with the existing iteration free fractal code for the benchmark images on the parameters like coding time, memory capacity and image quality. From the results of the computer simulation, the proposed method achieves excellent performance in coding time. The enhancement scheme for iteration free fractal image coding using vector quantization resulted in a reduction of 5.7 times and 11.5 times than the existing iteration free fractal code method for the single block partition of size 8x8 and 4x4 respectively on the Lena image for the codebook of size 16. The reduction in time is still higher in using code books of higher levels.