基于深度学习的电子元件焊点缺陷检测方法

Solder Joint Defect Detection for Electronic Components Based on Deep Learning

提出一种用于训练掩模区域卷积神经网络(Mask R-CNN)的半自动生成焊点图像掩模的方法。由于传统的通过人工标注获取掩模的方法费时费力,提出了一种简便快捷的基于GrabCut获取图像掩模的方法。该方法由两个阶段组成:第一阶段为基于GrabCut的焊点图像分割,输出像素级分割结果,从而获得所输入图像掩模;第二阶段实现基于Mask R-CNN的焊点表面缺陷检测方法,可以实现对缺陷的定位、分类和分割。试验结果证实了该方法的有效性,在保证Mask R-CNN方法检测精度的前提下,能快速、简单地获取训练Mask R-CNN所需的焊点掩模。

A method of semi-automatic generation of solder joint image mask for training mask region convolutional neural network(Mask R-CNN)is proposed.Because the traditional manual annotation method of obtaining mask is time-consuming and laborious,a simple and fast method of obtaining image mask based on GrabCut is proposed.The method consists of two stages.The first stage is the solder joint image segmentation based on GrabCut.The pixel level segmentation results are output to obtain the input image mask.In the second stage,the detection method of solder joint surface defects based on Mask R-CNN is implemented,which can realize the location,classification and segmentation of defects.Experimental results confirm the effectiveness of the proposed method.Under the premise of ensuring the detection accuracy of Mask R-CNN method,the solder joint mask required for training Mask R-CNN can be quickly and simply obtained.