| With the development of digital devices and the computer industry, digital image has been an important carrier for information exchange in people’s life. However, there is a large amount of data volume in an original image, and the image must be efficiently compressed to be used in practical. Fractal image compression encodes an image by the iterated function system based on the image self-similarity. However, traditional fractal image encoding algorithms are high computational and time consuming and cannot be directly used in practical applications. Therefore, how to effectively reduce the encoding time of fractal algorithm and at the same time ensure the excellent visual quality of decoding image is a very important research direction in the field of fractal image compression.In this paper, a large number of domestic and international digital image compression algorithms are researched, and aiming at the compression of medical abdominal CT image and static digital image, a Fractal-SPIHT-WT and Huffman based medical image ROI compression algorithm and a Gabor transform and fractal theory based image compression algorithm are respectively proposed. On the one hand, it meets the requirements of medical image for without distortion in some local interesting regions, and it reduces the amount of image data and the encoding time at the same time; On the other hand, it accelerates the fractal encoding algorithm as much as possible and reduces the complexity of the algorithm within the acceptable range of the human visual system for general digital image. During the process of ROI extraction, aiming at the structure characteristics of abdominal CT image and the drawbacks of traditional MRF algorithm, this paper proposes a ribs fitting based image pre-processing algorithm and devotes it to the MRF model in wavelet domain.In order to verify the effectiveness of the proposed algorithm, some comparisons with traditional image encoding algorithms are made, and experimental results show that the proposed algorithm owns good robustness, which can both accelerate the fractal encoding speed and keep excellent visual quality of decoding image. |
