综合诊断系统多维度重构小孔光斑中心测量方法  

Method for measuring laser spot center based on multi-dimensional reconstruction in integrated diagnostic system

作　　者：王拯洲[1,2,3] 胡炳樑[3] 殷勤业[1] 曹世康[3] 李东坚[3] 李红光[3] 

机构地区：[1]西安交通大学电子与信息学院,陕西西安710049 [2]中国科学院大学,北京100049 [3]中国科学院西安光学精密机械研究所,陕西西安710119

出　　处：《红外与激光工程》2015年第B12期73-79,共7页Infrared and Laser Engineering

基　　金：国家863计划(051Z331BOA);国家自然科学基金(11327303);国家自然科学基金(61405244)

摘　　要：为了对准直过程中能量弱、分布不均匀的小孔光斑信息进行增强并抑制噪声,消除光斑不稳定对计算结果的影响,文中提出了一种新方法。首先,将多幅分时采集的图像构建一个多维的图像立方体,利用MNF变换将数据信息主要集中在第一维,去除各维图像之间的相关性;其次,使用Kmeans方法将第一维图像分为光斑和背景,对分类图像进行数学形态学处理,寻找范围最大光斑的边缘为小孔光斑的边缘;最后,使用最小二乘法进行圆拟合来计算小孔图像圆心。实验结果表明,该方法能提高弱对比度的小孔光斑区域检测的准确率,达到理想光斑区域的97.15%,获得的小孔中心和半径误差小于2个像素,实现了综合诊断系统对小孔光斑的精确测量。In order to enhance the weak and asymmetrical image of small laser spot and suppress the noise information, eliminate unstable laser spot effect on the calculation results, a new method was proposed in this paper. Firstly, a multi-dimensional image cube was constructed by many images captured in different time, the primary information was concentrated in the first dimensional image, the correlation of each dimensional image was removed. Secondly, the first dimensional image was divided into laser spot and background by Kmeans method, the classified image was processed by mathematical morphology, the edge of largest spot was searched as edge of small laser spot. Finally, the least square method of circle fitting was used to calculate small laser spot center. The experimental results show that the method can improve the accurate rate on checking the weak and asymmetrical laser spot, the binary laser spot area obtained by new method is 97.15% of ideal spot area, the error of small laser spot center and radius is less than 2 pixels, and realize the accuracy measure of small laser spot in the integrated diagnostic system.

关 键 词：综合诊断系统 多维度重构 小孔激光光斑 MNF Kmeans 最小二乘法 圆拟合 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]