基于动目标检测的夜间视频实时增强技术

Real-Time Night Image Enhancing Based on Moving Target Detection

为了在普通PC环境中实现夜间视频的实时增强,对Ardely提出的DADPEQU夜间图像增强方法进行了改进,取得了更适合夜间图像增强的算法。设计了基于动目标检测的夜间视频实时增强策略,通过背景模板的使用,大大减少了增强处理的计算量,大幅提高了处理速度。将改进后的算法与增强策略结合,对夜间视频图像实时增强进行了实验,结果证明该算法既能取得较好的增强效果,又能满足实时处理的速度要求。

To achieve real time night image enhancement in ordinary PC environment, the Ardely＇s enhancemeht algorithm which is called DADPEQU is improved first, and a more suitable algorithm is obtained. Then, a newly designed strategy which is based on moving target detection is used in real-time night image enhancement. The key of the new strategy are the using of background templet. The improved algorithm and new strategy can observably improve the speed of computing. The improved algorithm and new strategy are proved to be efficient in real time night image enhancement.