| Summary: | 碩士 === 國立高雄應用科技大學 === 電機工程系博碩士班 === 102 === Quality inspection of components plays a critical role in industrial inspection. During the past decade, machine vision system has been widely used in manufacturing industry to replace the manual inspection such as process monitoring. A vision-based inspection system for carbon film resistors is proposed in this thesis. A series of image processing techniques are used to remove the background of an inspection image captured from CCD camera. Then, a method for finding the location of the edges of color codes and resistors is applied to inspection image for the convenience of feature extraction. In order to solve the problem of non-uniform illumination distribution of color codes, the interquartile range is introduced. Finally, the Euclidean distance is used to inspect the quality of resistors on the conveyor, namely the shape of the resistor and surface defects on the resistor. Experimental results show that the proposed system can detect the defects on the resistor effectively.
|
|---|
