| 基于细节特征融合的低照度全景图像增强 |
| |
| 引用本文： | 王殿伟,韩鹏飞,李大湘,刘颖,许志杰,王晶.基于细节特征融合的低照度全景图像增强[J].控制与决策,2019,34(12):2673-2678. |
| |
| 作者姓名： | 王殿伟 韩鹏飞 李大湘 刘颖 许志杰 王晶 |
| |
| 作者单位： | 西安邮电大学 通信与信息工程学院,西安,710121;西安邮电大学 通信与信息工程学院,西安 710121;电子信息现场勘验应用技术公安部重点实验室,西安 710121;哈德斯菲尔德大学 计算机与工程学院,Huddersfield,UK HD13DH;谢菲尔德哈雷姆大学 计算机学院,Sheffield,UK S11WB |
| |
| 基金项目： | 西安邮电大学研究生创新基金项目(CXJJ2017057). |
| |
| 摘 要： | 为了提高低照度条件下采集的全景图像的视觉效果,提出一种基于细节特征加权融合的低照度全景图像增强算法.首先,利用双边滤波算法提取出图像的光照分量,并分别采用自适应伽马校正和对比度受限的自适应直方图均衡化算法对光照分量进行处理;然后,与原始光照信息进行加权融合得到校正后的光照分量,并在反射分量调整时,提出一种自适应调整函数来校正反射信息;最后,将光照分量与反射分量合并,以实现对低照度全景图像的增强.实验结果表明,所提出的算法在提高图像亮度的同时,可以增强图像细节信息,去除噪声,使增强后图像色彩信息更加丰富自然.
|
| 关 键 词： | 细节特征 双边滤波 低照度全景图 图像增强 |
Low-light panoramic image enhancement based on detail-feature fusion |
| |
| WANG Dian-wei,HAN Peng-fei,LI Da-xiang,LIU Ying,XU Zhi-jie and WANG Jing.Low-light panoramic image enhancement based on detail-feature fusion[J].Control and Decision,2019,34(12):2673-2678. |
| |
| Authors: | WANG Dian-wei HAN Peng-fei LI Da-xiang LIU Ying XU Zhi-jie and WANG Jing |
| |
| Affiliation: | School of Telecommunication and Information Engineering,Xián University of Posts and Telecommunications,Xián710121,China,School of Telecommunication and Information Engineering,Xián University of Posts and Telecommunications,Xián710121,China,School of Telecommunication and Information Engineering,Xián University of Posts and Telecommunications,Xián710121,China;Key Laboratory of Electronic Information Application Technology for Scene Investigation,Ministry of Public Security,Xián710121,China,School of Telecommunication and Information Engineering,Xián University of Posts and Telecommunications,Xián710121,China;Key Laboratory of Electronic Information Application Technology for Scene Investigation,Ministry of Public Security,Xián710121,China,School of Computing and Engineering,University of Huddersfield,Huddersfield,HD1 3DH,UK and Department of Computing, Sheffield Hallam University,Sheffield,S1 1WB,UK |
| |
| Abstract: | A low-illumination panoramic image enhancement algorithm based on a weighted fusion of detail-features is proposed, which can improve the quality of a low-illuminance panoramic image. Firstly, the illuminance component is extracted by bilateral filtering, and enhanced by adaptive gamma correction and contrast-limited adaptive histogram equalization. Then, the three illumination information is fused to get the final illumination component. In the reflection component estimation, an adaptive adjustment function is proposed to correct the reflection information. Finally, the corrected light component and the reflection component are multiplied to get the enhanced image. The experimental results show that the proposed algorithm can not only improve the image brightness and gain more detailed information, but also remove the noise, and it makes the color of an image more natural and abundant. |
| |
| Keywords: | |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《控制与决策》浏览原始摘要信息 |
|
点击此处可从《控制与决策》下载全文 |
|
