\r杨敬钰¹，徐 吉¹，李 坤²，胡浩丰³，郭 鑫\r¹\r
\r
AuthorsHTML:\r杨敬钰¹，徐 吉¹，李 坤²，胡浩丰³，郭 鑫\r¹\r
\r
AuthorsListE:\rYang Jingyu ¹，Xu Ji ¹，Li Kun ²，Hu Haofeng ³，Guo Xin\r¹\r
\r
AuthorsHTMLE:\rYang Jingyu ¹，Xu Ji ¹，Li Kun ²，Hu Haofeng ³，Guo Xin\r¹\r
\r
Unit:\r\r1. 天津大学电气自动化与信息工程学院，天津 300072；\r
\r\r2. 天津大学智能与计算学部，天津 300350；
\r
\r3. 天津大学精密仪器与光电子工程学院，天津 300072\r
\r
Unit_EngLish:\r1. School of Electrical and Information Engineering，Tianjin University，Tianjin 300072，China；
2. College of Intelligence and Computing，Tianjin University，Tianjin 300350，China；
3. School of Precision Instruments and Opto-Electronics Engineering，Tianjin University，Tianjin 300072，China\r
\r
Abstract_Chinese:\r\r由偏振图像获取的偏振信息可用于图像增强与场景检测，近年来受到科研界与工业界的广泛关注．传统的偏振图像捕捉方法主要是通过旋转彩色相机前摆放的偏振片来采集多个偏振角度的图像或者使用偏振相机捕捉图像．前者只适合采集静态场景且手动换偏振片时容易造成相机镜头的抖动而无法对齐各个偏振角的图像，后者，即特制的偏振相机，由于制作工艺复杂，因此价格高昂．偏振图像采集设备制约了该领域的研究与发展．基于上述原因，搭建了由\r4\r个普通彩色相机组成的廉价偏振相机系统，该系统在\r4\r个相机前分别前置\r4\r个偏振角度的偏振片，即\r0\r°\r、\r45\r°\r、\r90\r°\r、\r135\r°\r，并固定在特制的相机架上，用以保证系统中相机之间的相对位置不变．四相机组成同步电路连接到电脑，利用软件控制同步采集场景偏振图像，减少人工触碰相机引起的误差．为了对齐不同偏振角度的图像获取场景偏振信息，采用多分辨率变分光流方法，利用梯度恒常模型对光照不敏感的特性，计算视图间的密集流场，利用斯托克斯矢量求解场景图像的偏振信息，最后，为了测量提出系统的性能，给出了该偏振系统的一种应用——低曝光图像增强，与传统彩色图像增强相比，在场景极暗处，偏振图像可以区分物体与背景，恢复图像细节，取得了更好的恢复效果．\r\r
\r
Abstract_English:\r\rPolarization information obtained from polarization images\r，\rwhich has received extensive attention from scientific research and industrial circles in recent years\r，\rcan be used for image enhancement and scene detection\r．\rThe traditional methods to capture a polarization image are mainly to collect images with multiple polarization angles by rotating the polarizer placed in front of the color camera or to use a use polarization camera\r．\rThe former is only suitable for static scene acquisition and easily causes camera lens movements when the polarizer is changed manually\r．\rAlternatively\r，\ra special polarization camera is expensive because of its complex manufacturing process\r，\rwhich restricts the research and development in this field\r．\rBased on the above reasons\r，\ran inexpensive polarization camera system composed of four ordinary color cameras was built\r．\rThe system has four polarization angles in front of the four cameras\r，\r0°\r，\r45°\r，\r90° and 135°\r．\rThe system is fixed on a special camera rack to ensure that the relative positions of the cameras in the system remain unchanged\r．\rThe four cameras are connected to the computer by a synchronous circuit\r．\rThe software is used to control the synchronous acquisition of the polarization images in the scene to reduce the error caused by the manual touch of the camera\r．\rIn order to obtain the polarization information of the scene by aligning images with different polarization angles\r，\ra multi-resolution variational optical flow method was adopted\r．\rThe gradient constant model was used to calculate the dense flow field between views\r．\rThe Stokes vector was used to solve the polarization information of the scene images\r．\rFinally\r，\rin order to measure the performance of the proposed system\r，\ra low exposure application of the polarization system is provided\r．\rCompared with traditional color image enhancement\r，\rthe polarization image can distinguish object from the background\r，\rrestore image details and achieve better restoration results in extremely dark scenes\r．\r\r
\r
Keyword_Chinese:偏振相机；低照度图像；图像恢复；图像增强\r

Keywords_English:polarization camera；low illumination image；image restoration；image enhancement\r


PDF全文下载地址:http://xbzrb.tju.edu.cn/#/digest?ArticleID=6417