| Summary: | 碩士 === 國立勤益科技大學 === 機械工程系 === 97 === Machine vision inspection technologies have been widely used in the areas of manufacturing industrial, processing monitoring, quality control, etc. A machine vision system comprises a CCD camera, an image-to-digital card, the illuminant source, the vision software, etc. The main purpose of this research is to set up a machine vision system for pattern matching inspection and pattern size measurement. This research simulates a manufacturing company. The setup system is flexible and used to replace the human workers to reduce the inspection error due to physical and/or mental fatigue. The processes of the inspection include (1) the image captured by a CCD camera, (2) define the region of interest (ROI), (3) the image signal processed by the LabVIEW vision software, (4) determine if the test sample pass the quality requirement. There are two scenarios in this research. One scenario is for pattern matching inspection, which is used to inspect the shape characteristics of the test sample to match the predefined specimen. The second scenario is for pattern size measurement. In this research I inspected a calibration card with four different sample locations to examine the pattern matching. I also used nine gauge blocks to measure their pattern sizes. The measurement accuracy and inspection reliability were discussed. Some conclusions and future work were addressed at the end of this dissertation.
|
|---|
