Research On Image Alignment Method Of Machine Vision

With the increasing integration of electronic components and size reduction, and productivity has increased annually, the motion accuracy and speed of electronic packaging equipment ware required more and more, traditional electronic packaging devices has been unable to meet the realities of needs, visual alignment technology as the key technology of electronic packaging equipment to ensure high speed, high-precision running, having made major equipment manufacturers and academic put attention on it.The structure of alignment system based on the image alignment algorithm in machine vision was systematically investigated in this dissertation. Main research works are as follows:(1)Based on the commonly used line detection algorithm, a new line detection algorithm was presented which based on angle and can improve the speed and accuracy. Based on the conclusions of previous results, the angle between lines was used to determine the line. Articles described the steps of algorithm in detail, and did a lot of experiments which have proved the feasibility of the algorithm and the superiority compared with the correlation algorithm. (2)Thesis involved in the visual system of image alignment for research work. Through carefully conclude the study of the current situation at home and abroad, present a new flag based on the geometric characteristics of the image alignment algorithm. After several experiments, firstly, doing two times filter before and after binary; secondly, morphological edge detection algorithm is used to extract the image edges; finally, align the geometric characteristics of selected signs simplified Hough transform algorithm and the minimum 2 by the method of combining line extraction. The alignment algorithm effectively improved the image of the treatment effect and alignment speed.(3)Articles on the image alignment algorithm designed in this paper was practical applications in the flip-chip devices, and the positioning accuracy was studied, the errors of visual systems ware analyzed by many experiments in different situations. The results show that the algorithm can satisfy the requirements of electronic packaging equipment.