用于目标跟踪的大视场CCD相机自动调光方法及实现

Method and Realization of Auto Light Control of Large Visual Field CCD Camera Used in Target Tacking

针对用于目标跟踪的大视场CCD相机的特殊结构,为了获得高质量图像,提出了一种基于电子快门无级调节的自动调光方法,即通过实时调整CCD相机的曝光时间来实现快速、高精度调光。文中提出了多区域灰度加权的图像灰度均值提取方法,消除了因大视场CCD相机背景图像数据过多而给自动调光带来的影响。系统采用FPGA内置RAM,通过乒乓缓存策略保证调光算法的实时性。通过对CCD相机的辐射定标和外景成像实验表明:该调光方法在辐亮度为25.697～260.548W/m2.sr区间内具有良好的图像质量,不仅满足调光速度快、实时性好、适应性强的要求,而且与无调光模式工作状态比较,CCD相机的动态范围增大了20倍,满足调光范围宽的要求。

According to the special structure of the large visual field CCD camera used in moving target tacking, a new method based on adjusting the electronic shutter steplessly is proposed to get the high quality image. Through adjusting the exposal time in real time, we could make the light adjusting fast and accurately. To draw the gray scale of the image, this paper put forward a new method named the gray scale weighting of different regional. Since the large visual field CCD eamera has a lot of back- ground image data, it will bring much influence to the auto light control. The new gray Scale drawing method which is advanced in writing could eliminate the influence. Using the RAM of FPGA, the system ensured that the light-adjusting arithmetic could work in real time by the ping-pang buffer usage strategy. The radiation scale and the result of location experiment indicated that the method could offer best image quality when the radiant luminance is from 25 W/m2 · sr to 260 W/m2 p· sr. The dynamic range of the new system had increased 20 times compared with that of the system without the auto light control. The new method satisfied not only the requirment of the wide dynamic range, but also the fast speed, the real-time, and the good adaptability.