在大型光学跟踪探测系统中, 频繁的自动对焦会导致光轴晃动, 引起拍摄的图像产生抖动, 对光学设备测量精度产生影响。针对实时光电探测设备的自动对焦技术, 提出一种对当前图像的清晰程度判断的快速算法。根据图像的清晰程度, 决定光学设备的自动对焦触发时间。首先对目标区域进行变换, 得到典型目标区域的梯度, 检测目标的边缘线及其方向, 对梯度数值拟合, 计算边缘的锐度分布。根据光学测量设备的分割精度要求事先确定阈值, 通过锐度分布函数值, 决定光学测量系统是否需要自动对焦。通过对上万帧不同序列图像的测试, 该算法判断的有效性达到93%。判断一帧图像的时间在2 ms以内。

In optics tracking and detection system, frequent autofocus always result in shake of axes, which will cause image to wobble. A new algorithm that can estimate current frame image of optic equipment be in focus or not is proposed in this paper. According to current image be in focus or not and corresponding threshold, optic control system can choose the time of auto focus automatism. The algorithm measures object edge and edge direction, then get several grads points along the normal line and calculate edge sharpness value. Based on the measure precision of optic-equipment, a threshold value will be set beforehand. If edge sharpness value is more than threshold value, it can conclude current frame digital image is in focus. If image is out of focus, optics system then takes autofocus program. This algorithm test several thousands of digital images be get from optics tracking and detection system, the result shows that correctness of this algorithm is more than 93%. It can finish processing one image during the time of 2 ms.