Defects Inspection And MATLAB Simulation Of Chips BGA Packages Based On Machine Vision

Semiconductor field is one of the most important application fields of machine vision technology. As the mainstream technology of current semiconductor packaging, Ball Grid Array (BGA) certainly needs the support of machine vision technology. This paper adopts machine vision technology to solve the problem of defects inspection during the study on BGA ball attachment workcell. The main contents are as follows:1. Based on the study of BGA packaging technology and BGA ball attachment technics,the characteristics of BGA packaging and the technical difficulties during the development of BGA ball attachment workcell were acquainted. Then the requirements of machine vision system were put forward.2. The article started with the packaging technics of BGA ball attachment workcell, analyzed the indispensability and essentiality to integrate machine vision technology in BGA ball mounting. Introduced the overview of the development and applications of machine vision,and researched the key technology on illumination,camera,image processing,etc.3. The theories and methods of digital image processing were studied, which included image pre-processing, feature extraction, recognition algorithm,etc,especially focused on Blob Analysis algorithm. Image capture control, image pre-processing, and defects inspection through programs have been achieved.4. Machine vision experiment platform has been constructed. According to the defect sorts of BGA chips,defects inspection process was devised. By the use of the software MATLAB,simulation system has been established and experiments have been done,finally ball defects inspection is actualized.