Method and apparatus for predicting range of motion of target

Abstract

The invention provides a method and apparatus for predicting the range of motion of a target. The method comprises the steps of obtaining environment attribute information and first path length information of respective paths set in an actual area for a target to select; iteratively solving a multiple linear regression function constructed based on the environment attribute information and the first path length information by using a particle swarm method to obtain second path length information corresponding to each path, independent and dependent variables of the multiple linear regression function being the environment attribute information and the first path length information respectively; according to the respective second path length information and the starting positions and total time of motion of the target, determining the respective end positions of motion corresponding to the target, the total time of motion being the time taken by the target to move from the starting positions of motion to the end positions of motion; and obtaining a predicted range of motion of the target according to the end positions of motion. The machine learning technology is used to predict the range of motion, the prediction accuracy is high, the time and effort are saved, and the practicality is high.

Description

technical field [0001] The present invention relates to the technical field of machine learning, in particular, to a method and device for predicting a range of motion of a target. Background technique [0002] Nowadays, the anti-investigation awareness of criminal targets is increasing year by year. Faced with the strict police control, suspects will choose to flee at the first opportunity in any troubled situation. Facing the fleeing of the criminal target, the police need to accurately delineate the escape range of the suspect in the first time (for example, within 20 minutes), deploy a large number of people within the range, and arrest the criminal target. [0003] The relevant technology provides a method for predicting the target escape range based on social science. This method mainly uses the relevant experience of experts in sociology and psychology, and comprehensively considers multiple on-site environmental factors such as road network, weather, and time, and T...

AI Technical Summary

Problems solved by technology

[0004] The inventor found in the research that the target escape range prediction method based on social science is mainly based on expert experience, and the accuracy of range prediction is low due to the inability to comprehensively evaluate various factors that affect the escape range. Accurate range prediction will lead to subsequent waste of more manpower and material resources to round up criminal targets

Images