Research On Algorithm Of Volume Measurement Based On Binocular Stereo Vision

Computer binocular stereo vision belongs to machine vision and it is also the frontier research direction in computer vision research.Through the cameras in the binocular system,the computer can capture external information.Image science is the premise and foundation of stereo vision theory.In recent years,as the research and application of image science has gained much more importance,the technology tends to mature in the future,and the computer binocular stereo vision has become a new hot spot in the field of computer vision.In this paper,the binocular stereoscopic principle and Hough transform are combined to measure the regular cuboid volume.This algorithm is mainly used in the logistics field assembly line system.The vertices of the regular bubble cargo in the logistics are taken as the calculated feature points,and the coordinates of the target point in the space are determined according to the binocular stereo vision principle and the Hough transformation line detection algorithm.Based on the binocular stereo vision transformation,the spatial coordinates of the vertex of the object are detected.The Matlab software is used to study and simulate the binocular stereo vision and Hough transform linear detection algorithm,and the parameters and image coordinates are optimized to the space.The transformation relationship between the coordinates,the detected image coordinates of the total of eight target vertices captured by the left and right cameras are used as the feature point coordinates into the optimized coordinate transformation equation,the matching parallax is calculated,and finally the object is completed by the triangle ranging formula.The three-dimensional coordinates of the vertices are calculated.This paper builds an experimental platform,completes the software design,verification and error analysis of the ranging system,and compares a variety of different algorithms to derive the optimal design.The volumetric measurement of regular bubble cargoes on the assembly line in the logistics can be completed in real time.The experimental results show that the algorithm based on binocular stereo vision and Hough transform can effectively extract the vertex information of regular objects from the background.After a lot of experiments,this paper proves that the regular object can be detected by the distance measurement algorithm designed in this paper.The length,height,and height data are consistent with the measured distance.