基于信息融合的PCB红外热像检测关键技术研究

发布时间：2018-01-22 03:35

本文关键词： 印制电路板检测红外热像技术图像融合故障诊断　出处：《红外技术》2017年09期 　论文类型：期刊论文

【摘要】：利用红外热像技术对PCB(Printed Circuit Board,印制电路板)的板载元件工作状态进行探测,是一种新型的非接触式的检测方法。针对传统检测方法未将PCB的可见光图像作为检测信息的有益补充,本文把两种图像进行融合,将有用的信息有机地结合了起来。首先检测可见光与红外图像的最大稳定极值区域特征,利用简化的SURF描述子实现多模态图像的配准。接下来采用结合非下采样轮廓波变换和压缩感知技术进行红外与可见光图像融合,获得兼具元器件位置和工作属性等信息的图像。最后使用ID3决策树获得被测故障的最大的类别信息,通过实验表明了基于信息融合的红外热像技术进行PCB检测的可行性,同时也对局限性进行了分析和阐述。
[Abstract]:Infrared thermal imaging technology is used to detect the working state of PCB(Printed Circuit board (printed circuit board). It is a new non-contact detection method. Because the traditional detection method does not take the visible light image of PCB as a useful supplement to the detection information, the two images are fused in this paper. The useful information is combined organically. Firstly, the maximum stable extremum region features of visible and infrared images are detected. The simplified SURF descriptor is used to realize the registration of multi-modal images. Then the infrared and visible images are fused by combining the non-downsampling contour wave transform and compressed sensing technology. Finally, the ID3 decision tree is used to obtain the largest category information of the fault under test. Experiments show the feasibility of infrared thermal image technology based on information fusion for PCB detection. At the same time, the limitation is analyzed and explained.
【作者单位】：洛阳理工学院工程训练中心;
【基金】：河南省科技攻关计划项目(162102210279,112102210522)
【分类号】：TN41;TP391.41
【正文快照】： PCB(printed circuit board,印制电路板)是电子电路的载体,各种电子元件以PCB为载体组合在一起,从而实现电路的整体功能。PCB的制造属于板卡级的封装,是整机总成的基础,它的质量直接影响电子产品整机的质量,可靠的电路板是保障电子产品质量的决定要素。产品的质量是一个企业旺

【相似文献】