Dimensional Measurement And Packaging Defect Detection Station Design Based On Machine Vision

Intelligent Manufacturing(IM)is the key to enterprise transformation in the new century.For the food industry,in order to enhance the competitiveness of enterprises and form a good brand effect,many enterprises have begun to integrate modern intelligent technology to form a good processing ecology.One of the key nodes in the food processing is the detection and immediate elimination of inferior products.The inspection indicators are often concentrated on visual attributes such as size,color,texture,and packaging.Therefore,research on a good intelligent visual detection mechanism become the key to the transformation of food companies.This article takes the production project of Chongqing Health Food Limited company as the background and designs an intelligent workstation for the measurement of size and the detection of packaging defect based on machine vision.Firstly,it analyzed the design requirements of intelligent workstations,and completed the design of the overall plan based on the existing production line.Secondly,the structure of station was designed,the conveying displacement control was realized through PID algorithm,and the multi-station system was optimized by using edge computing.Then,based on the system standardization research of camera calibration,light stability and image preprocessing,the size measurement algorithm was developed and tested based on edge detection and minimum outline fitting technology;packaging defects detection algorithm were developed and tested based on Siam neural network.And optimized the accuracy of the algorithm through multi-source data fusion.On this basis,for the two design forms of single station and multi-station,the two system frameworks of C/S and B/S were used to complete the design of the software system in combination with pyqt,django,html,MYSQL and other development tools.And carried out an instantiation test on specific functions.