Application of Feature Detection in XXY Alignment Platform

• Main Authors: LIN, HONG-YU, 林泓諭
• Other Authors: Mao, Wei-Lung
• Format: Others
• Language: zh-TW
• Published: 2019
• Online Access: http://ndltd.ncl.edu.tw/handle/5cpd37

碩士 === 國立雲林科技大學 === 電機工程系 === 107 === Due to the development of Industry 4.0, many traditional industries and technology industries have begun to be automated, and the precision platform has also begun to be widely used, and has gradually replaced manpower. Due to the limited precision measured by human eyes, if artificial alignment is used, In addition to the poor precision, it is also very time consuming, so we use the Charge-Coupled Device (CCD) for aligning with the XXY precision alignment platform and using image recognition instead of the manual alignment system. This thesis is divided into three parts: (1) introduction to the architecture of the XXY alignment platform, (2) introduction to image processing, (3) introduction of feature detection, and design of a set of alignment programs using SURF feature detection. In this study, the NI motion control card is combined with a stepper servo motor to move the XXY platform, and OpenCV is used for image processing, such as binarization, edge detection, mathematical morphology, and template comparison. Because the main purpose of this paper is to design a set of alignment programs, but because of the use of the pattern comparison method, it will cause poor comparison or platform rotation, so we use the feature detection SURF [ 4] method, to compare the graphics in our wafer, and successfully solve the problem that the comparison is not accurate with the rotation, and the feedback of the identified results to the controller, so that the XXY platform moves to achieve the image alignment.