Application Of 3D Point Cloud Technology In Forensic Appraisal

Forensic appraisal is an important part of the judicial process and provides a key scientific basis for sentencing judgments.At present,the mainstream measurement method of body surface injury for forensic staff is the traditional measurement method of manual contact,while the digital scar measurement method mainly relies on two-dimensional images for design,which cannot meet the accuracy requirements well.Based on this,this article proposes an intelligent forensic identification system,which provides a more accurate and convenient choice for forensic identification.Based on the application background of scar identification,this article designs the algorithms for the scar area measurement,the auricle defect ratio calculation and the scar length measurement for the body surface injury identification in the development of this system.The implement of scar area measurement is mainly based on the reconstruction of the scar.The target area is then obtained by calculating the total area of the triangle patch in the grid file.The problems in the calculation of the auricle defect ratio is solved by the point cloud registration algorithm.First,through filtering and Principal Component Analysis(PCA)are used to preprocess the point cloud of the human body surface,and the auricle data is automatically extracted by step edge amplitude on the head data.For large changes in the structure of the auricle,the ear hole area is added as the feature point of the Iterative Closest Point(ICP)registration to optimize the registration effect.In response to the professional needs of forensic appraisal,a method for calculating scar length with points as the core is proposed.In the projection of the scar data,the convex hull points and the smallest circumscribed rectangle are used to extract the endpoints.Afterwards,in order to solve the problem of deviation when the endpoints are directly used for connecting on the surface,the idea of segmented connection is used.On the projection plane,the curve thinning algorithm that integrates three-dimensional information is used to obtain key points between the endpoints.Finally,points used for curved surface connection are found on the 3D point cloud through the nearest neighbor search.Then the trend line is connected by the key points in turn,and the length of the scar is calculated by it.Finally,the above functions are integrated to design and implement an intelligent forensic identification system,based on the basic process of forensic injury identification.And the "Human Body Injury Degree Appraisal Standards" is organized into a rule database to realize automatic injury identification.The results of testing the system show that the system can effectively complete the main process of forensic identification and improve the efficiency.Also,taking the current calculation method as reference,the error of the measurement about the scar length and the auricle defect ratio is within ±3%,which meets the accuracy requirements of injury identification.And compared with traditional calculation methods,the use of three-dimensional measurement technology can more intuitively reflect the specific conditions of the injury site,which is convenient for subsequent injury analysis.