Research of Detecting Copper Foreign Material in Nickel Foam Based on Computer Vision
Nickel foam is the main material of nickel-hydrogen batteries.Nickel foam with a small amount of copper impurities can cause a battery short circuit and result in the battery explosion.In order to prevent nickel foam with copper foreign material out of the factory, this paper proposed an approach of detecting copper foreign material in nickel foam using computer vision.First, the system got the nickel foam images through CDD color cameras.Then finding binary copper areas of the image in RGB color space according to the principle that the reflectance spectrum of nickel foam and copper material are different;after that, using open operation of morphology on binary images to de-noise and repair;finally, to make the system more robust and reduce false alarms,SVM classifier was introduced to further filter out false alarms.The result has proved that the approach has 0.02% false alarm rate, but it has 0.00% missed rate and meets requirements of measurement accuracy.
nickel foam computer vision reflectance spectrum color space Support Vector Machine (SVM) classifier false alarm
Long Hu Songhua He Shengping Xia
College of Information Science and Engineering, Hunan University, Changsha Hunan 410082,China ATR National Key Laboratory, National University of Defense Technology, Changsha Hunan 410073, China
国际会议
三亚
英文
185-191
2015-12-26（万方平台首次上网日期，不代表论文的发表时间）