Title :
A transmission-type pinhole detection system for high-speed aluminum foil
Author :
Xin-Bin Luo ; Shan Fu ; Xiu-Qin Huang ; Qing-Qing Xing
Author_Institution :
Sch. of Aeronaut. & Astronaut., Shanghai Jiao Tong Univ., Shanghai, China
Abstract :
A transmission-type pinhole detection system is proposed for high-speed aluminum foil, which could automatically detect and identify pinhole defects on the aluminum foil strip. The transmitting light passes through the pinholes in the foil strip and forms a bright spot area in the camera image to detect the pinholes. However, the pseudo-defects generated by the external factors will affect the identification of pinhole defects; thus, a unique mechanical device is used to shield the external interference. On this basis, a simple, fast, and effective detection and classification algorithm is adopted to meet the requirements of real-time detection and identification. Subsequently, based on the demand of production application, a novel algorithm to merge pinhole defects is designed. This system has been evaluated and applied in an aluminum plant and the field application results show that it is a stable and reliable system for real-time pinhole detection.
Keywords :
aluminium; cameras; foils; image processing; production engineering computing; quality control; aluminum foil strip; aluminum plant; bright spot area; camera image; classification algorithm; detection algorithm; high-speed aluminum foil; merge pinhole defects; production application; pseudo-defects; real-time detection; real-time identification; real-time pinhole detection; transmission-type pinhole detection system; transmitting light passes; unique mechanical device; Abstracts; Classification algorithms; Light emitting diodes; Image processing; Non-destructive detection; Pinhole detection; Quality control; Surface defect;
Conference_Titel :
Machine Learning and Cybernetics (ICMLC), 2014 International Conference on
Conference_Location :
Lanzhou
Print_ISBN :
978-1-4799-4216-9
DOI :
10.1109/ICMLC.2014.7009119