摘要
本文在彩色图像的HSI颜色空间中,根据其分量的不相关特性,针对H、S分量建立了彩色图像的复变函数模型,并给出其边缘检测方法,得到H、S分量的边缘图像。抽取彩色图像的I分量构造灰度图像,利用当前成熟的灰度边缘检测方法(基于小波的多尺度边缘检测方法)得到I分量的边缘图像。最后,利用信息融合技术,将上述获得的两幅边缘图像进行信息组合得到最终的边缘检测结果。从试验的结果来看,本文所采用的方案充分利用了彩色图像所包含的信息,对图像的彩色边缘比较敏感,获得了满意的边缘检测结果。
Firstly, based on HSI color space, complex variable function model correlative to H and S components is proposed and the correlative edge detection method is presented, and then the edge image of H and S components can be achieved by the method. Secondly, the grey image is acquired by extracting I component from the color image, and processed with current mature grey edge detection method （accurate image edge detection based on wavelet multi-scale analysis） to obtain the edge image of I component. Finally, taking advantage of data fusion technique, the two edge images above mentioned can be combined into an accurate edge image of the color image. The experimental results show that the information of color image is efficiently utilized and the edge images combined arc satisfying by applying this framework.
出处
《光电工程》
EI
CAS
CSCD
北大核心
2008年第2期90-96,共7页
Opto-Electronic Engineering
基金
国家863计划资助项目(2002AA411010)
关键词
彩色图像
复变函数
HSI颜色空间
边缘检测
color image
complex variables function
HSI color space
edge detection
