100mL Soft Bag High Speed Inspection Robot

In the production of medical drugs,some foreign bodies may enter the liquid in the production process,such as plastic shavings,fibers and hair.Once these foreign bodies enter the human body,they may cause psychological discomfort at least,and even may be life-threatening.The country stipulates that infusion products must be tested before leaving the factory.There are mature automated testing solutions for glass infusion bottles and plastic infusion bottles,but soft bags still need manual testing in pharmaceutical companies.The result of manual testing is subjective.Therefore,it is valuable to develop an automatic detection device for soft bags foreign bodies inspection.In this paper,a 100 mL high-speed lamp inspection robot for testing soft bags based on machine vision technology will be developed according to the inspection method of the soft infusion bag in the Chinese Pharmacopoeia and the operation of the workers on the production line.According to the detection content required by the factory,the regional infusion soft bag shooting scheme,lighting method and image acquisition method will be designed to improve the quality of the photographs.Based on the pictures taken,a foreign body detection algorithm will be developed,and the task of foreign body detection in a 100 mL infusion soft bag will be completed.The main work is as follows:(1)It will be introduced that how to manually detect foreign objects in soft bags and popular foreign body detection methods for infusion products.Referring to the manual inspection method,the first generation infusion soft bag lamp inspection robot will be designed and realized,and the second version will be designed to overcome its shortcomings.(2)The material and impurity characteristics of the infusion bag will be analyzed.According to these characteristics,a lighting scheme combining bottom lighting and back lighting will be designed.In order to meet the requirement of detection precision,the parameters of the camera will be calculated,and the manipulator will be selected to realize the motion control of the camera,which can improve the contrast between the foreign object and the background and the clarity of the foreign object image.According to the specification of infusion bag,a method of image acquisition will be designed.(3)A suitable detection algorithm will be designed using image processing technology and deep learning technology.The basic feature,HOG feature and convolutional neural network feature of the image will be extracted by the corresponding algorithm,and the neural network will be used for feature fusion,image classification,and finally the foreign body will be detected.