Technology Research Of Crown Cap Defect Detection System Based On Vision

With the rapid development of modern industry,efficiency and accuracy of manual operation in the industry have been far from being able to meet the social needs currently."Chinese Artificial Intelligent in manufacturing industry "2025" indicates that the revolution has come.The intellective operation has a higher initiative,selectivity and accuracy than manual operation.It can efficiently complete a variety of industrial work.With the passage of time,industrial intelligence will replace manual operation.In the manufacture of caps,in order to effectively improve the efficiency of the production,the vision system by the machine to replace manual has become the mainstream.In the paper,caps of the crown cover is used as the object.Based on detections from the dislocations,the stains of the printing surfaces,the missings or the offsets of the claws,we designed and completed hardware and software system.The paper main completed the following works:(1)Based on the digital iterative method,the PDE digital filter was designed.In the process of the date selective adaptive filtering,the continuous PDE was discretized.We used corresponding digital filters engendered with digital iteration method and preprocessed the date of the original noise image.While utilizing amplitude information we got the density function of acquisition data noise and target measurement.Based on the date of the detection threshold,the amplitude likelihood ratio could be calculated.The simulation results show that this method guaranteed a high signal-to-noise ratio,and accomplished the process of the data selective adaptive filtering.(2)A weight matrix from PDE filter was established.Based on the amplitude likelihood ratio from the PDA filter,the frequency response based on the variational PDE filter was got.Through calculating the cost function of the data of selective adaptive filter,the algorithm of the filtering could be optimized.The simulation results show that this method improved the efficiency of the utilization on the filtering.(3)An algorithm of selective adaptive filtering based on variational PDE is proposed.In this paper,a selective adaptive filtering method based on variational PDE is designed for the defects of the algorithm on RLS filtering and LMS filtering.The simulation results show that the method not only has high signal-to-noise ratio,but also improve the robustness of the algorithm of the data selective adaptive filtering,and also protects the edge information of the image effectually.(4)We designed detection system for printing surface and claw of the crown cover cap.The algorithms of the detection system for hardware and software were designed.We solved error in processes and completed the defection testing for the bottle seal printing surface dirt,printing surface offset,missing and misplaced and other defects.Results show that the system is stable and the detection efficiency is high.