基于图像匹配的电路板故障检测与定位算法

自动光学检查是利用图像处理算法进行电路板故障检测与定位的技术,它可以解决传统电路板测试无法克服的缺陷.图像相减是自动光学检查中最简单和最直接的办法,但具有一定的局限性,这是由于二值化预处理造成的.本文提出一种基于图像匹配的加载电路板故障检测与定位算法,无需进行二值化预处理操作,利用图像差的绝对值、平方以及相关操作,检测与定位元件的倒装与漏装故障,并具有与图像相减类似的简易性,实验证实了该算法的性能.

Image matching based on fault detection and location algorithm of circuit board

Automatic Optical Inspection(AOI) is a technique for fault detection and location of electrical circuit board using digital image processing algorithms.It can provide an advantageous alternative for the unsolvable problems of traditional circuit board tester.Image subtraction is one of the easiest and simplest AOI techniques,but there are some limitations due to binary pre-processing in image subtraction.An image matching based fault detection and location algorithm of circuit board is proposed in this paper.It doesn't need the operation of binary pre-processing.Some image matching techniques including absolute value of image difference,square value of image difference,and correlation operation in this algorithm.The fault such as component reversely-mounted and non-mounted is detected and located.The proposed method in this paper is similar to image subtraction for the sake of its simple nature.The performance of the proposed algorithm is validated in experiments.