基于图像分割的改进型预测编码

Improved Predictive Coding Based on Image Segmentation

传统的预测编码是通过将图像转化为预测差来消除像素间的冗余度,从而实现图像压缩。为了进一步减小预测差的熵,提出了一种基于图像分割的改进型预测编码算法。实验结果表明,该算法具有比传统预测编码更好的压缩效果。同时,还比较了各种不同的分割方法对图像压缩性能的影响。

In this paper, an algorithm of improved predictive coding based on image segmentation is proposed. To reduce the value of the predictive error＇s entropy, the algorithm divides the image into two parts first, the background region where the gray value changes gently and the edge region Where the gray value changes sharply, then the traditional predictive coding is carded on these two parts. The experiment shows that the algorithm has better compression result than traditional predictive coding. And different influences on compression performance by different methods of image segmentation are also compared.