Development And Application Of Cylindrical Pellet Comprehensive Detection System Based On Machine Vision

Based on the actual situation of the company at the present stage,the defect and size detection of boron carbide cylindrical pellets mainly rely on manual completion.This method can only meet the needs of small batch s ampling inspection,but in the actual mass production process,the traditional manual inspection method is rather tedious and inefficient,which increases the cost burden of enterprises and affects the production quality and efficiency at the same time.In recent years,Machine Vision technology has been rapidly developed and widely used due to its good performance.Machine vision has many advantages such as non-contact,high precision,real-time,online,etc.It has a good application status and development prospects in many fields such as industrial automation,medicine,intelligent transportation and so on.This system is an online monitoring system for boron carbide pellets.The main function is to detect the size and defect of the boron carbide pellets.Based on the analysis of the research background and the development status at home and abroad,and the extensive research on the reference literature,the author has carried out in-depth research on the machine vision inspection system and given the specific solutions.(1)The system is mainly divided into two parts: optical system and material conveying system.The optical imaging system performs a systematic analysis of each inspection station,and selects an appropriate optical lens,camera,and imaging environment design.The material handling system is designed in detail for feeding,waiting,four-station visual feeding and material sorting devices.(2)Introduces the defect recognition algorithm part of the boron carbide pellet.A detailed algorithm for the cylinder block defects and the core end face defects is introduced.The ROI region of interest extraction algorithm,image enhancement technology,and defect recognition algorithm are introduced in detail.(3)Develops the size measurement algorithm for boron carbide pellets.The fit to the end circle is fitted by the least squares method and compared with the Hough transform method.For the side dimension measurement of the boron carbide pellet,in order to accurately fit the edge straight line and avoid other interference,the Radon variation method is selected for fitting,and a good application effect is obtained.(4)According to the defect identification and size detection algorithms proposed in the previous chapters,the corresponding experiment s are designed to verify the robustness and adaptability of the algorithm.The experimental results show that the algorithm performs well in defect identification and dimensional measurement,which satisfies the design requirements.