一种多目标光斑自适应快速检测算法  

A fast adaptive detection algorithm for multi-objective spot

作　　者：吴鹏飞[1] 贾璐瑶 WU Pengfei;JIA Luyao(College of Automation and Information Engineering,Xi’an University of Technology,Xi’an710048,China)

机构地区：[1]西安理工大学自动化与信息工程学院

出　　处：《激光杂志》2019年第6期27-31,共5页Laser Journal

基　　金：陕西省重点产业创新链计划(No.2017ZDCXLGGYG01);西安市高校院所人才服务企业工程计划(No.2017080CG/RC043(XALG001));西安市碑林区科技计划项目(No.GX1816)

摘　　要：在激光光斑探测系统中,需要对多个目标光斑信息进行提取,针对这种情况,提出了一种多目标光斑自适应快速检测算法。该算法首先对整个多目标光斑图像与背景图像进行差分处理、中值滤波、边缘检测等预处理,使用迭代法实时获取不同的阈值对光斑图像进行二值化处理,利用图像的对称特点快速精准的提取光斑区域像素面积、质心位置坐标、直径等参数。最后通过实验验证该算法可以在不同背景光干扰的情况下自适应的选取阈值,且能有效、稳定的对光斑区域进行分割,实现多目标光斑的快速检测。In the laser spot detection system,it is necessary to extract the information of multiple target spots.For this situation,a multi-target spot adaptive fast detection algorithm is proposed.The algorithm firstly pre-performs the whole multi-target spot image and background image,such as differential processing,median filtering and edge detection.It uses the iterative method to obtain different thresholds in real time to binarize the spot image,and uses the symmetrical feature of the image to quickly and accurately extract the pixel areas,centroid position coordinates,diameter and other parameters.Finally,the experimental results show that the algorithm can adaptively select thresholds under the condition of different background light interference,and can effectively and stably segment the spot area to achieve rapid detection of multi-target spots.

关 键 词：多目标光斑检测 差分运动图像 背景噪声 迭代算法 自适应阈值分割 

分 类 号：TN247[电子电信—物理电子学]