摘要
遥感图像增强是对遥感图像进行后继处理的必要步骤且在遥感图像的处理中占有独特的地位。当前,有很多图像增强方法被应用到遥感图像的增强当中,文中引入Retinex增强算法,该算法可以实现对图像进行自适应的增强处理。为了使Retinex算法更具有普遍性和自适应性,采用多尺度Retinex算法对图像进行增强。多尺度Retinex算法即结合多个尺度上的单Retinex算法,因而可以进一步提高Retinex算法的自适应能力和对图像的增强效果。应用小波变换方法和多尺度Retinex算法对遥感图像进行增强对比实验,实验结果表明多尺度Retinex算法在遥感图像的增强方面具有优越性,并且可以取得令人满意的增强效果。
Remote sensing image enhancement occupies a peculiar position in remote sensing image processing and is an important preprocessing step for subsequent analysis. Numerous image enhancement techniques are available for remote sensing image enhancement. The Retinex algorithm, capable of self-adjusting for enhancing images, is introduced. To improve the capability of self-adjusting for enhancing images, the multi-scale Retinex （MSR） is adopted in this study. The MSR is superior to the SSR in combination of scales, which can improve the capability of self-adjusting and the effectiveness of the enhanced image at the same time. A remote sensing image enhancement experiment is conducted using wavelet transform and MSR respectively. The experiment results demonstrate that MSR is superior over wavelet transform, and can achieve satisfying results for enhancing remote sensing images.
出处
《红外技术》
CSCD
北大核心
2007年第9期536-540,共5页
Infrared Technology
基金
安徽省自然科学基金(编号:070412059)
安徽大学研究生创新项目(编号:20073050)
关键词
遥感图像
图像增强
小波变换
多尺度RETINEX
remote sensing image
image enhancement
wavelet transform
multi-scale Retinex （MSR）
