基于区域特征加权的JPEG改进算法

JPEG标准在对彩色图像编码时,为了提高图像的压缩率,需要将图像从RGB颜色空间转换到YCbCr颜色空间并对颜色分量Cb与Cr进行重采样.颜色空间转换与重采样会引起还原图像光谱畸变现象.在研究图像融合技术的基础上,提出一种基于区域特征的动态加权色彩空间变化方法.根据主观目视和客观评价指标,对还原图像进行了比较和分析.仿真试验表明,提出的方法可以提高还原图像的光谱分辨率,同时很好保持了原图像的空间分辨率.

AN IMPROVED ALGORITHM OF STANDARD JPEG BASED ON WEIGHT REGIONAL FEATURES

In order to enhance the compression ratio of picture,JPEG standard needs to transform the picture from the RGB color space to the YCbCr color space and carry on downsampled resample to the chrominance components(CbCr)when encoding color image.This kind of downsampled resample can cause severe spectrum distortion phenomenon on reconstructed image.According to thorough studies on the principle of IHS fusion method,a new color transformation method based on weighted regional features is proposed.Based on objective and subjective image performance measure,reconstructed pictures are compared and analyzed.The simulated experiment indicates that the proposed method can enhance the spectrum resolution of reconstructed image,and at the same time maintain the spatial resolution of original image.