DT-CWT和色调映射相结合的低照度图像增强算法  

Low Illumination Image Enhancement Algorithm Based on DT-CWT and Tone Mapping

作　　者：宋瑞霞[1] 李达[1] 余建德[2] Song Ruixia;Li Da;Yu Jiande(College of Sciences,North China University of Technology;Beijing 100144,2Faculty of Information Technology,Macao University of Sciences and Technology,Macao 999078)

机构地区：[1]北方工业大学理学院,北京100144 [2]澳门科技大学资讯科技学院,澳门999078

出　　处：《计算机辅助设计与图形学学报》2018年第7期1305-1312,共8页Journal of Computer-Aided Design & Computer Graphics

基　　金：国家自然科学基金(61272026;61571046);国家重点研发计划(2017YFF0209806);澳门科技发展基金(097/2013/A3)

摘　　要：为了提高低照度图像的质量,提出基于双树复小波变换(DT-CWT)与色调映射相结合的图像增强算法.首先将RGB图像转换到HSV色彩空间,然后对亮度分量V进行双树复小波分解,分离出高、低频子带;然后对高频子带进行Butterworth滤波,增强图像细节、抑制噪声;对低频子带,则使用文中改进的色调映射算法来调整图像亮度;最后进行DT-CWT逆变换得到处理后的V分量,并将之与H分量、S分量合成,转换为清晰的RGB图像.文中充分利用DT-CWT能准确表达图像细节的优良特性,并在Retinex基础上改进了色调映射算法,实现对低照度图像的增强.实验结果表明,该算法可以明显地改善低照度彩色图像的视觉效果.To improve the quality of low illumination image, based on DT-CWT and tone mapping, an image enhancement algorithm is proposed. First, the RGB image is converted into the HSV color space whose luminance component V will be divided into high and low frequency sub-bands by DT-CWT. The high-frequency sub-band is processed by Butterworth filter to enhance the image details and suppress the noise while the low-frequency sub-band is managed by the improved tone mapping to adjust the image illumination. Finally, DT-CWT inverse transformation is applied to obtain the reconstructed V component which is synthesized with the H and S components to obtain a clear RGB image. In this paper, we make full use of excellent characteristics of accurate image-detail expression of DT-CWT which is combined with the improved tone mapping algorithm based on Retinex theory to achieve the enhancement of low illumination image. The experiment results show that the proposed algorithm can obviously improve the visual effect of the low illumination color image.

关 键 词：低照度图像 双数复小波变换 双边滤波 色调映射 RETINEX 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]