基于SURF算法的电路板元件定位及检测

为解决人工电子元件检测日益困难的问题,提出了一种基于加速鲁棒特性(speed-up robust features,SURF)算法的电路板元器件的定位及检测方法.针对SURF算法在实际图像匹配应用中会发生较大概率误匹配的问题,提出了一种提高匹配概率的算法.首先用最近邻方法进行粗匹配,然后利用曲线拟合剔除错误匹配点对,最后利用匹配点对的坐标位置关系找出元件的位置,用改进的7个不变矩方法进行产品检测.实验证明了此方法的有效性和可行性,提高了特征点匹配的精度,实现了电路板元件的精确定位和检测.

Circuit Board Components Positioning and Testing Based on SURF Algorithm

In order to solve the problem of the increasing difficulty in detecting artificial electronic components, a method for circuit board components positioning and testing based on speed-up robust features（SURF）was pro-posed. SURF algorithm has a high probability of error matching points in practical application. Therefore, a method for improving matching probability was proposed. First, a coarse points matching was performed by the nearest neighbor method. Then error matching points were eliminated using the method of curve fitting. Finally, the position of the components was found in the circuit board through the coordinated relation of the matching points, using the seven optimized invariant moment features method for product testing. The experimental results have proved the fea-sibility and effectiveness of the proposed method. This method improves the accuracy of the feature points matching and realizes the accurate positioning and testing of circuit board components.