Wavelet Transformed based Fast Fractal Image Compression

The main requirement of compression system in multimedia applications with large image size (more than or equal to 512x512 pixels) is high compression ratio with acceptable visual quality. Fractal image coding is one of the alternatives for high compression rate but due to exhaustive encoding time, it has limited practical applications. In this paper, wavelet transform based fast fractal image coding is proposed. FFT based fractal image coding with variable quadtree partition is applied to the approximation subband and three detail subbands of the wavelet transformed image. Quadtree partitioned wavelet subtree is constructed after wavelet decomposition of fractal decoded approximation subband image. The self similarities existing in wavelet subtree are exploited by predicting the coefficients at finer scale from those at coarser scale using affine transformation. The proposed method improves the visual quality of the decoded image at very low bit rate and achieves average 92% reduction of coding time compared to the pure fast fractal image coding. Results also show that the JPEG based compressed image PSNR value reduced by 4-5 dB when the compression ratio around 60:1-70:1.