基于NCC改进算法的SMT电路器件定位

发布时间：2018-04-22 06:39

本文选题：SMT + NCC　；参考：《计算机应用》2017年S1期

【摘要】：随着电子产品的微型化发展,对表面贴装技术(SMT)中电路器件的定位精度和速度提出了越来越高的要求。针对SMT贴合工艺中标记点匹配速度慢、边缘检测精度差的问题,为了提高模板匹配效率,在经典NCC算法的基础上提出了一种缩小搜索区域的NCC算法(RSR-NCC),通过提取模板图像中关键信息与搜索图像进行匹配,实现电路器件的定位。在保证匹配准确性的情况下,该算法可减少计算量,同时增加关键信息的权重。分别采用传统NCC算法、SSDA算法、NCC+粗搜索算法与RSR-NCC算法对SMT贴合过程中随机采集的图像进行匹配定位,结果显示RSR-NCC算法在保证匹配结果准确的基础上,匹配时间分别为NCC算法、SSDA算法、NCC+粗搜索算法的1%、2.8%、8.5%,明显提高了匹配速度,证明了该算法的优越性。
[Abstract]:With the development of the microminiaturization of electronic products, the positioning precision and speed of the circuit devices in the surface mount technology (SMT) are getting higher and higher. In order to improve the template matching efficiency, a narrow search is proposed to improve the template matching efficiency in order to improve the template matching efficiency in order to solve the problem of slow matching speed and poor edge detection precision in the SMT bonding process. The NCC algorithm (RSR-NCC) of the cable region is used to match the key information in the template image and the search image to realize the location of the circuit device. In the case of ensuring the accuracy of the matching, the algorithm can reduce the computation and increase the weight of the key information. The traditional NCC algorithm, the SSDA algorithm, the NCC+ rough search algorithm and the RSR-NCC algorithm are used respectively. According to the matching and positioning of the random collected images in the SMT closing process, the results show that on the basis of ensuring the accuracy of the matching results, the matching time of the RSR-NCC algorithm is NCC, SSDA, and NCC+ coarse search algorithm 1%, 2.8%, 8.5%, and the matching speed is obviously improved, and the superiority of the algorithm is proved.

【作者单位】：中国矿业大学信息与电气工程学院;中国人民解放军第五七一五工厂;
【基金】：国家重点研发计划项目(2016YFC0801808)
【分类号】：TN405;TP391.41

【参考文献】