基于人眼对比度敏感视觉特性的彩色图像压缩技术

Color Image Compression Technology Based on Contrast Sensitivity Characteristics of Human Vision System

结合人眼对比度敏感视觉特性的测量过程和图像离散余弦变换特征,提出了一种基于人眼感知特性的彩色图像压缩技术方案。方案首先将彩色图像转换到CLELAB颜色空间描述,并分别对3个分量进行离散余弦变换处理,再结合人眼对比度敏感函数模型和DCT变换域系数特征对变换域系数进行量化处理,并采用Huffman算法进行编码压缩,最后采用逆过程进行解压并匹配出解压后的彩色图像。通过仿真实验,结果表明:在压缩比为98.355 5∶1时,PSNR能达到37dB以上,人眼无法分辨出解压缩图像与源图像之间的差异;且编码效率能达到0.74以上,冗余度在0.26以下。表明提出的彩色图像的压缩技术方案在保证编码质量和图像质量的前提下能够实现较高的压缩比,是一种可行的、有效的彩色图像压缩技术方案。

According to the measurement process of the human visual contrast sensitivity characteristics and characteristics of Discrete Cosine Transform of the image,a color image compression technology program is proposed based on human visual characteristics.In the program,the color image is converted and described in CLELAB color space at first,and each of the three components of the color image is processed by Discrete Cosine Transform.Secondly,the transform domain coefficients are quantified according to the contrast sensitivity function model of human vision and coefficient＇s characteristics of Discrete Cosine Transform.The color image is encoded using Huffman algorithm.Lastly compressed images are decompressed using the inverse process,and the color images are matched after decompression.The simulation results show that PSNR can reach 37 dB or more when the compression ratio is 98.355 5∶1,and the human eye can not distinguish the differences between the source image and decompressed image,and the coding efficiency can reach more than 0.74,redundancy in 0.26 below,which shows that the color image compression technology solutions proposed can achieve a higher compression ratio under the premise of ensuring the encoding quality and image quality,and is a feasible and effective solution for color image compression technology.