Research On Furniture Plate Location Detection System Based On Line-laser Scanning

In recent years,the loading and stacking process of furniture plate in the processing and manufacturing process is mostly completed by manual operation.Under the background of this production mode,the application of machine vision technology in the process of plate loading and unloading can realize plate stacking process automation,which can greatly improve the flexibility level of enterprise automation.For the stacking and dismantling of plates,the first thing to do is to obtain the position of plates in space,that is,three-dimensional coordinate information.According to the dimension of processing information,the technologies based on machine vision can be divided into those based on traditional two-dimensional vision and those based on three-dimensional vision.In the plate production and processing workshop,the environment is more complex,with many dust particles,light factors such as instability.However,the plane camera based on traditional visual technology cannot cope with these environments well,and its anti-interference ability is insufficient,which cannot reflect the stability of the visual system.Therefore,in this paper,the laser linear sensor with strong anti-interference ability is used to scan and detect the plate,and a 3d measuring system based on line laser scanning is developed to obtain the spatial position of the plate,which mainly includes the following aspects.Firstly,the source,research background and three-dimensional measurement technology are introduced and studied.The working principle of sensor triangulation is studied.Based on the overall demand analysis of the system,each component module of the board position detection and the working process of the system were designed,and the experimental platform of the detection system experiment was built according to the design scheme of the detection system.Second,the study on the depth of plate image preprocessing algorithm design,mainly analyses the image enhancement,filtering denoising and plank object extraction algorithm,and then the corresponding image algorithms respectively experimental treatment,the treatment effect,through the comparative analysis between them in order to select the piecewise linear transformation,median filtering and threshold segmentation algorithm based on highly on the depth of plate image preprocessing,gained good pretreatment effect.Then,on this basis,in order to obtain the 3d coordinate information of the plate,the algorithm of 3d reconstruction and pose estimation based on the depth data of the plate is studied.In order to reconstruct the actual model of plate,an algorithm of surface fitting reconstruction was designed to obtain its stereo model.By analyzing the characteristics of the curvature change of the surface contour of the plate,the X and Y surfaces of the plate model are fitted by the method based on the least square principle,and the validity of the reconstruction algorithm is verified in the experiment.Further,in order to determine the spatial coordinates and posture of the plate,the target posture estimation method of the plate was studied and analyzed,and the 3d matching algorithm based on the CAD 3d data model of the plate was designed to obtain its 3d state information.Finally,the software of the plate position detection system is designed.The error distribution of the coordinate and attitude data of the plate in the experiment is analyzed,and the experimental results show that the system can meet the requirements of the plate position detection.