海量多媒体图像压缩中的冗余零系数消除算法设计  

Design of Redundant Zero Coefficient Elimination Algorithm in Massive Multimedia Image Compression

作　　者：马蕾[1] 贺鹏飞[2] MA Lei;HE Pengfei(Telecommunication Engineering Institute,Beijing Polytechnic,Beijing 100176,China;Department of Information Network,China Population and Development Research Center,Beijing 100081,China)

机构地区：[1]北京电子科技职业学院电信工程学院,北京100176 [2]中国人口与发展研究中心信息网络部,北京100081

出　　处：《自动化仪表》2023年第7期65-69,共5页Process Automation Instrumentation

基　　金：北京电子科技职业学院重大科研基金资助项目(2021Z007-KXZ)。

摘　　要：为了提高海量多媒体图像压缩的效果,提出一种针对海量多媒体图像压缩中的冗余零系数消除算法。计算图像信息熵,并构建金字塔模型,以提取多媒体图像的信息特征。将图像压缩分为一维离散余弦变换(DCT)压缩与二维DCT压缩,确保图像压缩后结构不被破坏。结合零系数数量变换曲线,确定最佳聚焦位置,建立冗余零系数消除模型,即可完成压缩图像冗余零系数消除。试验结果表明,所提方法可提高海量多媒体图像压缩的峰值信噪比,并且缩小图像所占用的空间。所提方法能够有效提高图像压缩的质量,使图像结构更加完整,故实际应用价值更高。To improve the effect of massive multimedia image compression,a redundant zero coefficient elimination algorithm in massive multimedia image compression is proposed.The image information entropy is calculated,and a pyramid model is constructed to extract the information features of multimedia images.The image compression is divided into one-dimensional discrete cosine transform(DCT)compression and two-dimensional DCT compression to ensure the structure of the image is not destroyed after compression.The redundant zero coefficients of compressed images can be eliminated by combining the zero coefficient number transformation curve,determining the best focus position,and establishing the redundant zero coefficient elimination model.The experimental results show that the proposed method can improve the peak signal-to-noise ratio of massive multimedia image compression and can reduce the space occupied by the image.The proposed method can effectively improve the quality of image compression and make the image structure more complete,so it is of higher practical application value.

关 键 词：多媒体图像 冗余零系数 金字塔模型 信息特征 峰值信噪比 离散余弦变换压缩 图像结构 

分 类 号：TH126.1[机械工程—机械设计及理论]