一种基于高维形象几何理论的彩色图像增强算法  

Color Image Enhancement Approach Based on High-dimensional Space Geometry

作　　者：谭乐怡[1] 王守觉[2] 

机构地区：[1]同济大学电子信息与工程学院计算机系,上海201804 [2]中国科学院半导体研究所,北京100083

出　　处：《计算机科学》2008年第12期214-215,223,共3页Computer Science

摘　　要：基于近年来发展起来的全新理论——高维形象几何与仿生信息学,提出一种彩色图像增强算法。本算法将彩色图像看作是高维空间的点,通过对多幅图片的分析,推导出点与点之间的内在联系,提出"模糊-再模糊-清晰"的增强算法,并由实验证明该算法对模糊图像的清晰化有很好的效果。A novel algorithm on color image enhancement was proposed. The approach is based on point location in high-dimensional space geometry method, and it is quite different from traditional ways. A color image is looked as a point in high-dimensional space, we tried to find out the intrinsic relationship between the points （images） via experi- ments or analysis. Then, a ＂blurring-blurring-debluring＂ algorithm was proposed. Experiments proved the availability of this algorithm.

关 键 词：高维形象几何 仿生信息学 彩色图像增强 模糊图像清晰化 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]