Scene abnormal state identification method and system and intelligent terminal

Abstract

The invention discloses a scene abnormal state identification method and system and an intelligent terminal, and the method comprises the steps of collecting the human body biological posture information in a preset scene image so as to obtain the image information of all collected human body targets in the preset scene image; in a preset time period, tracking and recording the position information and the posture information of each collected human body target under a plurality of time nodes, and respectively storing each time node and the position information and the posture information of each collected human body target; based on the posture information of each collected human body target at each time node, estimating and outputting an action change track of the collected human body target within the preset time period; and according to the output action change track, obtaining a behavior state which is about to be made by the collected human body target, and outputting a responsestrategy according to whether the estimated behavior state is abnormal or not. The method, the system and the intelligent terminal are suitable for the scene monitoring when an area is small or many object targets exist, the identification precision is higher, and the identification speed is faster.

Description

technical field [0001] The invention relates to the technical field of intelligent identification, in particular to a method and system for identifying abnormal states of scenes for multi-target monitoring. Background technique [0002] In recent years, with the rapid development of artificial intelligence, intelligent identification and monitoring technology based on artificial intelligence has brought great changes to people's lives. Traditional monitoring equipment cannot predict the occurrence of dangerous situations in complex environments. In actual use, such as gas stations, chemical plants, prisons, airports, railway stations and other places where there are many people, it is easy to have fire smoke, criminals escape from prison, pedestrians fall, and open flames are used in flammable areas. Effective identification and alarm measures are required in all scenarios to avoid the deterioration of the situation. In addition, in aspects such as nursing homes and corpor...

AI Technical Summary

Problems solved by technology

In the field of object detection (Object Detection), two series of one-stage methods are usually used: however, the recognition accuracy and speed of object detection (Object Detection) technology will be affected by the size of the detection area and the number of targets. When the target is small or there are many targets, the recognition accuracy will decrease or the speed will slow down. For some specific occasions where people are dense and the area is relatively small, the monitoring equipment based on target detection has low recognition accuracy for abnormal scenes, and the processing speed slower

Images