基于视频图像处理的快速对焦方法研究

对焦速度与精度是自动对焦系统中最重要的两个参数，它们决定系统能否迅速地定位到焦点处，以获取清晰的图像。通过对比连续帧图像局部区域清晰度评价函数值来判断对焦趋势，同时控制对焦驱动电机的转动方向，最后进行反向搜索并与最大判据值比较以寻找到最佳成像位置，成功对焦所需最长时间小于2s。经过大量实验证明该方法具有较好的实时性和精确性。

Fast Auto-Focusing Algorithm Based on Video Image Processing

The velocity and precision are two important aspects of the auto-locus system. They can decide the whether systern can quickly locate the focusing point to obtain clear image. The auto-focusing algorithm based on the image sharpness m CCD camera systems is studied in this paper. Firstly, the method discerns the focus trend through comparing evaluation data of serial sub images, then it changes the motor running direction according to the focus trend. Secondly, it searchs the best focusing position in reverse direction depending on judgement condition, and the maximum time of successful focusing is less than two seconds. The experimental results show that the algorithm has good real time property and accuracy degree.