基于BEMD图像特征点的图像压缩方法

摘要

BEMD as a relatively new multi-scale analysis technology, adaptively extracting composition in line with visual perception from images has its own unique advantages. This article explores the BEMD applications and promotion in the field of image compression. It proposes image compression concept based on feature points from BEMD components of image. Extrema (both maxima and minima) and zero-crossings of IMF components and residual as feature points coefficient are quantized and coded, receiver decodes. The feature points reconstruct each IMF and residual component by interpolation, and then reconstruct the original image. Experiments prove its feasibility.%BEMD作为一种较新的多尺度分析方法,在自适应的提取图像符合视觉感知的成分上有其独特的优势.本文探索研究了BEMD 在图像压缩领域的应用推广.提出了基于BEMD提取图像分量的特征点的图像压缩理念.对各IMF分量和残余提取他们的极值点和过零点,将极值点和过零点视为特征点,对特征点系数做量化编码,接收端解码,再对特征 点插值分别 重建各IMF和残余分量,进而重建原图像.通过实验证明了其可行性.