一种粗精结合的快速图像自动调焦算法  

Algorithm of rapid autofocus combining rough and fine adjustment

作　　者：林兆华[1] 米阳[1] 葛兵[1] 

机构地区：[1]中国科学院长春光学精密机械与物理研究所,长春130033

出　　处：《国外电子测量技术》2015年第10期28-31,共4页Foreign Electronic Measurement Technology

摘　　要：为保证光测设备获得清晰图像,提出一种粗调与精调相结合的自动快速调焦算法,该算法运用图像清晰度评价函数自动搜索到正焦面,使用改进的Kirsch边缘检测算子评价函数以大步长进行粗略调焦,使用小波变换为基础的评价函数实现小范围、小步距的精确调焦。该算法聚焦区域的选取采用非均匀采样函数,大大减少了数据处理的工作量和所需时间。最佳聚焦位置的搜索采用一种改进的爬山搜索算法与搜索策略,有效地提高了搜索效率,通过实验表明:该算法能够满足对图像的调焦要求,取得较好的使用效果。Auto-focusing technique is one of the key technologies to ensure the high quality image for the optical measurement system.An algorithm of rapid autofocus combining rough and fine adjustment is proposed.This method can automatically search the focal plane by evaluating the image definition.First,the image edge function based on the improved Kirsch operator with high performance is used to rough focusing.The evaluation function based on wavelet transform is used to fine focusing in narrow range later.Un-uniform sampling function is used as the method calculating the focusing window,which can reduce the work of the data processing and the calculating time.An improved mountain climbing searching algorithm and tactics of the optimum focus position,which can significantly accelerate the search speed.The experimental results indicate that this algorithm can meet the focusing requirements of the optical measurement system,which can obtain good practical value.

关 键 词：光测设备 图像处理 评价函数 自动调焦 

分 类 号：TP202[自动化与计算机技术—检测技术与自动化装置] TN16[自动化与计算机技术—控制科学与工程]