Research On Weighting Technologe Of Concrete Aggregate Based On Machine Vision

For the natural stacking of aggregates,the weight measurement of aggregates on the belt is a common problem.At present,the weight measurement is mainly carried out by manual measurement and instrument measurement.Manual measurement has high precision,high cost and poor real-time performance.In individual cases,manual measurement will be more difficult.In order to quickly measure the aggregate pile weight,reduce equipment costs.This paper presents a method for measuring the weight of aggregate pile based on machine vision.This method uses non-contact,image acquisition,image segmentation,edge detection,shape detection,stereo matching,three-dimensional reconstruction and other technologies to quickly measure the weight of aggregate pile.The measurement principle is to use the camera to obtain the volume and density data of the target aggregate pile,and finally obtain the weight of the target aggregate pile.Experiments show that the weight measurement method based on binocular stereo vision can solve the problem of rapid measurement of aggregate heap.The main contents of this paper are as follows :(1)In the calibration of binocular vision,the experimental platform is built,and the advantages and disadvantages of traditional calibration method,traditional calibration method and active vision calibration method are compared.Zhang Zhengyou calibration method is used for calibration,and the reprojection error is reduced by repeated operation in matlab to meet the measurement requirements.(2)An improved GA genetic algorithm is proposed,which adds the process of splitting and merging.Compared with the previous GA algorithm,A * algorithm and PSO-GA,the convergence property,path optimality and convergence speed of the algorithm are better.Finally,it is proved that the method can segment the edge of aggregate pile well.(3)A volume measurement method based on semi-global matching algorithm(SGM)is proposed.The matching cost Census transform is improved,and the speed in the high-resolution image operation process becomes faster.The semi-global matching algorithm(SGM)and the global matching algorithm have the same matching strategy,which is divided into four parts : cost calculation,cost aggregation,disparity calculation and disparity optimization.In this paper,Open CV is used to realize the semi-global matching algorithm,and the dense disparity map of the image to be measured is obtained.The three-dimensional coordinate information of the object to be measured is reconstructed by triangulation principle,and the volume of the object to be measured is solved by integral.The experiment proves that the scheme is feasible.(4)An improved watershed algorithm is proposed to measure the edge segmentation ability of aggregate,which verifies the practicability of calculating the gap rate of aggregate on a single color belt.Establish an image library,optimize the watershed algorithm,and use the watershed algorithm to solve the surface gap rate of the aggregate pile.A large number of experiments were made to find the correlation between aggregate bulk density and gap rate,and the linear relationship between them was given.Finally,the scheme of measuring the weight of aggregate pile by machine vision is realized.Figure 39 Table 9 Reference 103...