This research project aims to develop a waferWafer inspection tool for the wafer fabrication process. In the wafer fabrication process, physical defectsDefects present a significant challenge for engineers, requiring them to monitor, detect, minimize, and promptly address any faults in production. This approach significantly reduces production costs while ensuring that product quality is maintained. The newly developed tool provides a new approach to strategy, which is capable of locating the defect and immediately analyzeAnalyze the parameters of the defect. DefectsDefects like foreign inclusions, porosityPorosity, and cracks are prevalent and can be readily identified using the new system. In the system, first the imageImage is captured by a high-resolution camera, and immediately the captured image is translated into a graphical pattern format. The MatLab engine is used to support the translation procedure. Finally, the graphical pattern is compared with the imagesImage pattern, which is saved in the system database to decide all the parameters of defectsDefects such as type of area, perimeter, defect, size, etc.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Wafer Defects Inspection Using Image Processing Techniques

  • Mahzan Johar,
  • Muhammad Khusairi Osman,
  • Zakiah Ahmad,
  • Mohamed Yusof Radzak,
  • Mohd Fauzi Abu Hassan

摘要

This research project aims to develop a waferWafer inspection tool for the wafer fabrication process. In the wafer fabrication process, physical defectsDefects present a significant challenge for engineers, requiring them to monitor, detect, minimize, and promptly address any faults in production. This approach significantly reduces production costs while ensuring that product quality is maintained. The newly developed tool provides a new approach to strategy, which is capable of locating the defect and immediately analyzeAnalyze the parameters of the defect. DefectsDefects like foreign inclusions, porosityPorosity, and cracks are prevalent and can be readily identified using the new system. In the system, first the imageImage is captured by a high-resolution camera, and immediately the captured image is translated into a graphical pattern format. The MatLab engine is used to support the translation procedure. Finally, the graphical pattern is compared with the imagesImage pattern, which is saved in the system database to decide all the parameters of defectsDefects such as type of area, perimeter, defect, size, etc.