基于总变分的MPEG解码算法  

Total variation based MPEG decompression algorithm

作　　者：肖孝军 陈智斌[1] 文有为[1] Xiao Xiaojun Chen Zhibin Wen Youwei(Faculty of Science, Kunming University of Science ＆ Technology, Kunming 650500, Chin)

机构地区：[1]昆明理工大学理学院,昆明650500

出　　处：《计算机应用研究》2017年第9期2825-2829,共5页Application Research of Computers

基　　金：国家自然科学基金资助项目(11101193)

摘　　要：针对低比特MPEG图像序列出现的压缩痕迹,提出一种后处理正则化方法进行MPEG解码。首先由自适应量化得到一个原始视频DCT系数的量化区间,重建的视频序列投影到此区间内,视频是连续的静止图像,得到耦合时间维度和分离时间维度的两种总变分模型;最后利用经典的原—对偶算法求解提出来的凸优化模型,得到后处理MPEG解码视频序列。实验结果表明,总变分正则化函数能够一定程度上减轻压缩痕迹,提高解码视频的质量。Focused on the defect that compression artifacts in low-bit MPEG images sequence, this paper proposed a post-pro- cessing technique of regularization to decompress MPEG objection. Firstly, the quantization interval of the DCT coefficients of video original could be gotten because of adaptive quantization, and the reconstructed video sequences were projected onto the interval. Secondly, because of videos were continuous still images sequences, it presented two total variation models that all di- mensions were coupled and temporal dimension was decoupled from two spatial dimensions. Finally, it exploited a classical pri- mal-dual algorithm to solve the convex model so as to obtain a post-processing MPEG video sequences. The experimental resuits demonstrate that the total variation functions can reduce compression artifacts to some extent and improve the quality of the decoded videos.

关 键 词：MPEG压缩 压缩痕迹 总变分 原-对偶算法 

分 类 号：TN919.81[电子电信—通信与信息系统]