摘要
基于暗通道先验去雾的图像增质方法在目标探测中表现良好,但其以光强信息为载体,光学维度单一的不足导致其目标表征效能下降.本文借助偏振对物理属性的敏感特性,提出在传统暗通道先验去雾方法中引入偏振信息来增强不同物体之间的辨识程度.研究了暗通道先验去雾方法中退散射与偏振探测的理论,并搭建机械式偏振滤波成像设备在雾天环境对所提方法的目标表征功能进行了实验验证.研究表明,基于偏振的暗通道先验去雾方法能够同时获取物体的光强与偏振信息,与传统暗通道先验去雾方法相比,利用目标与背景的偏振差异能够明显地提高二者对比度.此研究结果可应用于现有的偏振成像仪器系统,实现退散射与偏振信息的实时提取,进一步提高雾天目标探测与表征的效率.
The image enhancement method based on dark channel priori defogging performs well in target detection,but it takes the light intensity information as the carrier and the single optical dimension leads the target characterization efficiency to decline. Based on the sensitivity of polarization to physical properties, in this paper a proposal is made that polarization information is introduced into the traditional dark channel priori defogging method to enhance the recognition degree between different objects. The theory of backscattering and polarization detection in dark channel priori defogging method is studied, and the mechanical polarization filtering imaging equipment is built to verify the target characterization function of the proposed method in foggy environment. The research shows that the dark channel priori defogging method based on polarization can obtain the light intensity and polarization information of the object at the same time. Compared with the traditional dark channel priori defogging method, using the polarization difference between the target and the background can significantly improve their contrast. This research result can be applied to the existing polarization imaging instrument system to realize real-time backscattering and polarization information extraction, and further improve the efficiency of target detection and characterization in fog.
作者
霍永胜
Huo Yong-Sheng(Key Laboratory of Instrument Science and Dynamic Testing,Ministry of Education,North University of China,Taiyuan 030051,China)
出处
《物理学报》
SCIE
EI
CAS
CSCD
北大核心
2022年第14期112-120,共9页
Acta Physica Sinica
基金
国家自然科学基金(批准号:11847069,62005251)资助的课题.
关键词
偏振
暗通道先验去雾
目标表征
退散射
polarization
haze removal using dark channel prior
target representation
elimination of scattering
