An intelligent illumination control system for image measurement
Image measurement based on machine vision is a promising method for the precise measurements of workpiece. The measuring precision depends on high quality image data. The lamp-house and illumination scheme affect the image quality and edge extracting effect directly. It will bring out the characteristics of the object as the system required. In order to increase the contrast, they should make the difference of the object and the background evident. All of the above assure that the result of measurement is the best. This paper introduces an illumination control method which is used in image measurement, including system configuration, lamp-house structure, control principle, illuminating effect evaluation, control optimization strategy, etc. This article adopts the maximum between-class variance to evaluate the contrast between the object and the background, and uses the average gradient method to assess the image quality of surface features. In order to increase the efficiency of searching the optimal control vector point, it uses fuzzy expert system to determine the lightening control points and limit the search space for cetain objects to measure. Experiments show that, the effects of illumination control for image measurement are prominent with this method. Meanwhile, the real-time performance of the system could meet the engineering demands.
illumination control image quality evaluation fuzzy expert system machine vision image measurement
Zhou hu Yang Jianguo
College of mechanics Engineering Donghua university, Shanghai, China
国际会议
2010 International Conference on Measurement and Control Engineering(2010年IEEE测量与控制工程国际会议 ICMCE2010)
成都
英文
270-275
2010-11-16（万方平台首次上网日期，不代表论文的发表时间）