High-Speed Optoelectronic System for Surface Inspection of Fuel Pellets

Two high-speed optoelectronic methods and computer vision system for surface inspection of nuclear reactor fuel pellets are considered. In the proposed methods the reflected images of pellet surfaces are taken by digital cameras, which provide high contrast of defective areas against the frame background. Image processing algorithms for identification of defective products are given. A system with productivity of up to 10 pellets per second was designed. Results of real time processing of the obtained images show that the probability of defective pellets detection is not less than 95%. In case of visual inspection by operators the probability of false solution by operator may be unacceptable, namely defective fuel pellets are considered as fit production. The above- mentioned drawbacks determine the urgency of the problem of automating the detection of defective pellets. Moreover, it is important that the productivity of visual inspection system must be very high: from 4 to 10 pellets/sec. To solve this problem, we have proposed to use optoelectronic methods and computer vision system combined with a mathematical apparatus for image processing. In this report two high-speed methods and automatic system for side and edge surfaces inspection are presented. The experimental results of the system testing are given.