Research On 3D Reconstruction System Based On Line Laser Scanning

Currently,machine vision technology is widely used in all aspects of society.Among them,the measurement technology based on three-dimensional reconstruction has become an important research direction of machine vision,which is widely used in intelligent automated production,robot navigation,medical assistance and other fields.The traditional contact measurement is not suitable for the current development direction due to its slow working efficiency,high maintenance cost and strict accuracy requirements of the measuring device.According to the number of cameras,the 3d reconstruction technology can be divided into monocular camera,binocular camera and multiocular camera,and can also be divided into active(such as structural light)and passive(such as binocular stereo vision)according to the nature of the light source.The3 d reconstruction system studied in this paper is based on the line laser scanning of monocular camera,which has a wide application prospect in scientific research and industrial production.Compared with the traditional measurement method,it has the advantages of simple structure,convenient maintenance,non-contact,high efficiency and more application scenes.The main tasks of this paper are as follows:(1)Set up a 3d reconstruction system based on line laser scanning,analyzed technical requirements,integrated practicability and economy,selected the relevant hardware: sensor,lens,stepping motor,line laser transmitter,driver,controller,power supply,etc.,and ran and debugged the whole system.(2)Analyzed and established the mathematical model based on the linear laser 3d reconstruction system,completed the camera calibration,laser calibration and motion axis calibration,and proposed a new laser calibration algorithm.(3)Preprocessing and center extraction are carried out for the collected light bar images.In view of the problems of too long computation time or poor extraction accuracy in some traditional center extraction algorithms,a new improved algorithm based on gray-scale barycenter method and gaussian fitting method is proposed.(4)The whole system process was completed,and three-dimensional object reconstruction and measurement experiments were carried out.The experimental results proved the feasibility of the system scheme,and analyzed the factors affecting the accuracy of the reconstruction process,as well as some scenarios where the system is not applicable.