面向高空飞艇平台的视频增强与稳像  

Video enhancement and stabilization for high altitude airship platform

作　　者：王书民[1] 张爱武[1] 王飞[1] 胡少兴[2] 孙卫东[3] 仰继连[3] 

机构地区：[1]首都师范大学三维信息获取与应用教育部重点实验室,北京100048 [2]北京航空航天大学机械及自动化学院,北京100083 [3]清华大学电子工程系,北京100084

出　　处：《吉林大学学报（工学版）》2014年第1期203-210,共8页Journal of Jilin University:Engineering and Technology Edition

基　　金：国家科技支撑计划项目(2012BAH31B01);北京市自然科学基金重点项目(B类)(KZ201310028035);国家自然科学基金项目(41071255)

摘　　要：针对高空飞艇平台获取的视频监测数据存在的问题,提出了帧内增强去雾和帧间去除高频振动分量的稳像方法。采用基于Retinex图像增强技术进行帧内增强处理,去除薄雾,保留图像细节,还原真实色彩;建立四自由度全局运动估计模型,并采用中值滤波方法进行运动补偿,去除高频振动分量,实现帧间稳像;并给出未定义区域重建算法。通过实际应用,证明了本文算法能够很好地实现飞艇视频稳像,在帧内增强和帧间稳像方面均有良好效果,可满足后期监测分析需求。An image stabilization method is proposed for the problems existing in the video acquired from high altitude airship platform. It can remove the fog in the intra-frame images and filter the high frequency vibration components in the inter-frame images. The image enhancement technique Retinex is adopted to remove the fog in the intra frame images. It retains the details of the image, and restores the genuine color of image. Then, the four freedom global motion mode is established to estimate the global motion, and the median filter is adopted to smooth the global motion and remove the high frequency vibration components. It achieves the image stabilization in the inter-frame images. The algorithm of constructing the undefined area is given at the same time. Through the practical application, the algorithm proposed in this paper achieves the image stabilization for the video acquired from airship platform perfectly. It gets good image stabilization results in the intra-frame images and inter-frame images, which meet the requirements of monitoring and analyzing.

关 键 词：计算机应用 RETINEX算法 运动估计 运动补偿 视频稳像 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]