An Improved Method of Remote Sensing Image Compression Based on Fractal and Wavelet Domain

One of the efficient solutions for data transmission and storage of huge remote sensing images is the high-fidelity and rapid compression. The hybrid algorithms based on fractal coding and wavelet transformation have already been developed and are becoming very popular at present. In this paper, an improved method of remote sensing image compression based on fractal coding and wavelet domain is proposed. Firstly, the method uses the wavelet-transformation to create the similarity between sub-bands of an image, and the fractal algorithm is exploited to code for adjacent high-frequency bands with similarity character, it is must be mentioned that some improvement has been used to speed up searching for similarity blocks during the step. Consequently, fractal iterative decoding and wavelet inverse transform are applied to reconstruct the image. From the above, it can be concluded the method proposed in this paper not only makes full use of the characteristic of wavelet multi-resolution, but also considers the self-similarity among objects. Finally, some experiments have been given, and the results demonstrates that the method can do well in the remote sensing image compression, less losing of the image information and more recovering of image quality. Moreover, ensuring the signal-noise ratio of the image, it can better increase the image coding efficiency * Email: [email protected].