Wavelet-based Image Coding Optimization Algorithm Research

Image Coding is one of the most important research topics of image processing. With the high-performance image coding algorithms, we can make large-scale image data stored in the limited storage space possible. Although the image coding algorithm has been concerned and studied in the field of image processing, there are still many defects in the existing algorithms. Based on the wavelet domain image coding algorithm theories, this paper discusses the current image coding algorithm, studies and improves the performance of it.First, we import the WBCT transform to image coding algorithm, this makes it more precise to describe the geometric contour of the image. The experiment results prove that, compared with wavelet transform, this method can handle a higher image energy accumulation effect and get a better description of human visual effects. Secondly, based on the high-performances advantages of multi-core computing platforms, a cache-aware algorithm of the Haar wavelet transform is proposed in this paper. Through reconstructing the raw data and changing the calculation order, we highly improve the performance and potential of the cache. At the same time, we use the advantages of multi-threaded parallel techniques. Experiments show that, the algorithm can access a higher performance on different computing platforms.At last, we combine the rapid Haar wavelet transform and image coding algorithm. The experimental data shows that the algorithm can get a larger increase when compressing the high-resolution images with low bpp. In the original image coding algorithm, the memory management is a great difficult for programmer. In this paper, we also realize a image coding algorithm based on a state array not the original list link. The experiment shows that the method based on state array needs less and stable storage and easy to manage memory.