Image enhancement of shadow region based on polarization imaging

Mohamed Reda, Linghao Shen, Yongqiang Zhao

科研成果: 书/报告/会议事项章节会议稿件同行评审

4 引用 (Scopus)

摘要

Object of shadow area suffers from low contrast and noise, causing difficulties for object detection and recognition. However, conventional image enhancement methods perform inefficiently in severer shadow conditions, where the contrast enhancement is not obvious, or the noise amplification is huge. In fact, Polarization information can be preserved even in very heavy shadows. Combining the polarization information with the intensity and chromaticity information of the original image can overcome the above problems. This paper proposes a method of object enhancement in shadow based on polarization imaging. Firstly, this paper puts forward degree of polarization (DOP) Invariance Prior and DOP-Chromatic Consistency Prior. Then, based on these two priors, a feedback image enhancement method based on polarization imaging is proposed. Experiments show that the proposed method is superior to the other conventional image enhancement methods.

源语言英语
主期刊名Pattern Recognition and Computer Vision 2nd Chinese Conference, PRCV 2019, Proceedings, Part II
编辑Zhouchen Lin, Liang Wang, Tieniu Tan, Jian Yang, Guangming Shi, Nanning Zheng, Xilin Chen, Yanning Zhang
出版商Springer
736-748
页数13
ISBN(印刷版)9783030317225
DOI
出版状态已出版 - 2019
活动2nd Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2019 - Xi'an, 中国
期限: 8 11月 201911 11月 2019

出版系列

姓名Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
11858 LNCS
ISSN(印刷版)0302-9743
ISSN(电子版)1611-3349

会议

会议2nd Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2019
国家/地区中国
Xi'an
时期8/11/1911/11/19

指纹

探究 'Image enhancement of shadow region based on polarization imaging' 的科研主题。它们共同构成独一无二的指纹。

引用此