摘要
本文对方向分解图像编码进行了系统研究。与Kunt方案相比,提出了更有效的边缘信息选择和编码算法,改进了内插方法和模型,并采用了新的方向滤波器设计方法,基于上述改进,建立了一套完整的匹配视觉生理机制的方向分解图像编码算法。系统模拟实验表明,当数据压缩比近60:1时,采用本方案仍能获得较好的图像质量。
In this paper, the directional decomposition-based image coding is investigeted systematically Compared with kunt's work, we have presented more efficient algorithm of edge information selection and edge coding, improved the method and model of interpolation and adopted new design methods for direction filter. Based on the above-mentioned improvements, a set of algorithms of directional decomposition-based image coding has been established completely, which is matched with physiological mechanism of vision The system simulations show that good quality of the reconstruction image can be still obtained by our coding scheme when the data compression ratio is near 60:1.
出处
《通信学报》
EI
CSCD
北大核心
1991年第3期23-30,共8页
Journal on Communications
基金
国家自然科学基金
关键词
图象编码
匹配
视觉
生理机制
picture coding, image data compression, source coding.
