Research On 3D Point Cloud Model For Laser Measurement Of Complex Ring Forgings

Complex ring forgings are the key components of modern mechanical equipment,which are widely used in many fields.As a precision component,most equipment has high precision requirements for them.In order to ensure the production quality of ring forgings,3D laser scanner is usually used to reconstruct the model and measure the dimensional accuracy of ring forgings.However,the 3D point cloud model of ring forging obtained by 3D laser scanner has a huge amount of data,which is not conducive to model reconstruction.Therefore,its point cloud model needs to be simplified.In addition,the ring forging need to be scanned and registered for many times to ensure that the point cloud model will not be missing,incomplete,dislocation and other problems.In order to solve the above problems,the simplification and registration of 3D point cloud model of complex ring forging are studied in this paper.The main research contents are as follows:First of all,there is a huge amount of data in the three-dimensional point cloud model of ring forging.In order to simplify it,this paper proposes a triangle folding algorithm based on the combination of single median error metric and firefly algorithm.The algorithm derives the single median error function according to the spatial distance change of the median after the triangle grid is folded.Then,the error function is regarded as the brightness of the firefly,and the firefly algorithm is used to quickly calculate the minimum value of the function,and the triangle grid with the minimum value of the function is folded.The final simplified model is obtained after several triangle mesh folds.Secondly,in order to obtain a complete point cloud model of ring forging,this paper proposes a registration algorithm based on subspace similarity.The registration process of the point cloud model is divided into two parts.The first part is the coarse registration of the model when the spatial position relationship between the point cloud to be registered and the target point cloud is unknown.In this part,an improved point-surface feature histogram is used to describe the features,and the gray correlation analysis is used to find the matching point pairs.The second part is to use the closest distance sorting to pair after obtaining a certain initial position,and measure the registration error through the subspace similarity and Hilbert Schmidt independence criterion.After continuous iteration,the registration error is reduced as much as possible to ensure the registration accuracy.Finally,the 3D point cloud model of the existing ring forging in the laboratory is obtained through the three-dimensional laser scanner,and it is processed according to the model simplification algorithm and registration algorithm proposed in this paper.The size keeping ability of the algorithm is verified by measuring the size of the experimental results.In addition,several groups of contrast experiments are carried out to verify the effectiveness of the algorithm.