| As the rapid development in Internet and multimedia technology, exchanging of image information is becoming more widely. People have higher quality requirements in browsing and transmission of images. With limitations of storage space, image information is needed to compress before storage and transmission. In recent years, wavelet-based embedded coding algorithms are constantly emerging. Achieving a compression, it can reconstruct multiple images in different quality requirements. So the research of high quality embedded compression algorithm based on existed embedded image compression methods has great practical significance.This paper is organized as follows:1,This paper presents a research of block-based embedded coding method, and key technologies of JPEG2000 standard are in-depth studied.2,Taking human visual perception into account, the article presents an object-based embedded coding with optimized truncation method based on traditional EBCOT algorithm. In this algorithm, the original image is segmented into different visual interesting objects, and these objects can be encoded independently using different embedded encoding strategies and generated separate streams. Simulation results show that the proposed improved algorithm is effective.3,Based on the request of object-based embedded coding, at first, this paper put PCRD algorithm to control object-based embedded code streams. And then combining the characteristics of human visual perception, a rate control algorithm for object-based embedded coding is proposed. Combining the characteristics of human visual perception, the algorithm estimates the importance of each object and determines the coding priority on the basis of average entropy theory. The algorithm carries the optimized interception on the visual objects'coding stream and reloading. Simulation results indicated that the new algorithm can improve the overall visual effect of the reconstructed image comparaed with PCRD algorithm. |
PDF Full Text Request