Target object escape route prediction method and deploying and controlling platform

Abstract

The invention provides a target object escape route prediction method and a deploying and controlling platform. The target object escape route prediction method comprises the steps of obtaining feature information of the target object; acquiring a plurality of target monitoring videos according to the feature information, and shooting suspected objects in the target monitoring videos, wherein thesimilarity between the suspected objects and the target objects is larger than a set threshold value; acquiring historical tracks of the suspected object according to the plurality of target monitoring videos; acquiring real-time movement information of the suspected object according to the first monitoring video; determining a fork where the suspected object is about to arrive according to the real-time movement information, wherein the fork corresponds to at least three roads, and the at least three roads comprise a road where the suspected object is located currently and at least two to-be-selected roads; and predicting respective implementation probabilities of the at least two to-be-selected roads according to the historical tracks and the traffic light states and congestion states ofthe at least two to-be-selected roads. According to the invention, the target object can be quickly and effectively captured.

Description

technical field [0001] The present application relates to video surveillance technology, and in particular to a method for predicting escape routes of target objects and a deployment and control platform. Background technique [0002] In order to maintain the safety of people and property and bring convenience to traffic, many intersections, public places, residential areas, elevators, etc. in the city have installed cameras to collect images around the clock, providing a solid foundation for the detection of cases and the execution of traffic violations. Favorable evidence. Especially when the suspects in the case (including suspects or suspect vehicles) show up or sudden illegal incidents (such as traffic violations), the investigators need to hunt down the suspects, and the surveillance video captured by the camera becomes a powerful auxiliary evidence. . [0003] In the traditional solution, investigators need to manually view a large number of surveillance videos, loo...

AI Technical Summary

Problems solved by technology

But obviously, this kind of efficiency is very low. Just watching the surveillance video will consume a lot of manpower and time. What's more, the subjective experience is too dependent on the qualifications of the investigators. It is very likely that the predicted escape route is wrong. The hunt plan is also ineffective

Images