1. 硅电池片自动串焊表面缺陷在线视觉检测研究.
- Author
-
陈智强, 楼佩煌, and 钱晓明
- Abstract
Aiming at the online quality inspection of welded silicon battery chipsꎬ an online quality inspection method of welded silicon bat ̄ tery chips based on visual identification technology was proposed in this paperꎬ including feature recognition and visual measurement. In the feature recognition stageꎬ an improved region growing algorithm was proposed to finish the high ̄speed and stable detection of gridline ̄fracture featuresꎻ For the scratch and dirty features caused by weldingꎬ the regions of interest was adaptive thresholdedꎬ then the defects were filtered by invariant moments feature. In the field of visual measurementꎬ using the symmetry of target imageꎬ pixel based measurements of solder strip offsets and chip spacings were carried out through a number of sampling rectangles. The acquisition of inter chip images at each trans ̄ mission beat was enabledꎬ the original image was segmented into a number of regions of interest by the edge location methodꎬso that high ̄ speed online detection was completed. The results of actual online test indicate thatꎬ the visual inspection method of welded silicon battery chips can identify target defects quickly and accuratelyꎬ and it has the characteristics of good recognition stability. [ABSTRACT FROM AUTHOR]
- Published
- 2018
- Full Text
- View/download PDF