摘要
提出了一种基于边缘信息测度的四叉树自适应分块的DCT压缩算法。该方法首先对原始图像建立的边缘信息测度图进行四叉树分块,继而根据分割好的各个块的大小和块内包含边缘点的多少来进行不同压缩比的DCT压缩编码。尺度大的块内压缩比较高;尺度小的块内压缩比相对较低。计算机模拟实验证明,单位bit恢复信噪比(SNBR)比非自适应法提高了30%。
A new DCT coding method based on edge information measure and adaptive quadtree segmentation was proposed to enhance compression ratio of DCT coding and transmission efficiency.The quadtree algorithm was used to segment the image adaptively according to the pixels type and thresholds.Then DCT was applied to each block with different compression ratio.The larger the block size is,the higher the compression ratio we take.The simulation experiment shows that the compression ratio and SNR carried by transmitting 1 bit(SNBR) is bigger than 30 % of nonadaptive method.
出处
《光电子．激光》
EI
CAS
CSCD
北大核心
2003年第10期1081-1084,1093,共5页
Journal of Optoelectronics·Laser
基金
国家交通部攻关资助项目(200136422504)
