eduzhai > Applied Sciences > Engineering >

Deep Learning Based Defect Detection for Solder Joints on Industrial X-Ray Circuit Board Images

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 6 pages

Abstract: Quality control is of vital importance during electronics production. As themethods of producing electronic circuits improve, there is an increasing chanceof solder defects during assembling the printed circuit board (PCB). Manytechnologies have been incorporated for inspecting failed soldering, such asX-ray imaging, optical imaging, and thermal imaging. With some advancedalgorithms, the new technologies are expected to control the production qualitybased on the digital images. However, current algorithms sometimes are notaccurate enough to meet the quality control. Specialists are needed to do afollow-up checking. For automated X-ray inspection, joint of interest on theX-ray image is located by region of interest (ROI) and inspected by somealgorithms. Some incorrect ROIs deteriorate the inspection algorithm. The highdimension of X-ray images and the varying sizes of image dimensions alsochallenge the inspection algorithms. On the other hand, recent advances on deeplearning shed light on image-based tasks and are competitive to human levels.In this paper, deep learning is incorporated in X-ray imaging based qualitycontrol during PCB quality inspection. Two artificial intelligence (AI) basedmodels are proposed and compared for joint defect detection. The noised ROIproblem and the varying sizes of imaging dimension problem are addressed. Theefficacy of the proposed methods are verified through experimenting on areal-world 3D X-ray dataset. By incorporating the proposed methods, specialistinspection workload is largely saved.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×