一种基于小波变换和多级树集合分裂算法的图像多描述编码改进算法

Algorithm for Multiple Description Image Coding Based on Wavelet Transform and Set Partitioning In Hierarchical Trees

在研究基于嵌入式零树小波的多描述编码算法基础上,提出了一种结合小波变换和多级树集合分裂算法的图像多描述编码改进算法,阐述了该算法的原理,并用Matlab来对2种算法进行程序仿真。仿真结果证明,后者比前者在相同的条件(除所有描述都接收的情况)下能得到质量更高的重构图像,提高了数据传输的鲁棒性。

On the basis of EZW-based(embedded-zero wavelet tramsform)multiple description coding algorithm,an improved algorithm for multiple description image coding is proposed in combination of wavelet transform with the algorithm of multi-grade set separation,its principle is deseribed.Matlab is used to simulate the program of the 2 algorithms.The results shows that under the same conditions the later can obtain a reconstructed image with its quality higher than that of the former(except for receiving all the descriptions),by which the robustness of data transmission is improved.