An Intelligent Early Warning Method for Video Surveillance Based on Target Spatial Relationship Constraints

Abstract

The invention discloses a video monitoring intelligent early-warning system and a video monitoring intelligent early-warning method on the basis of target space relation constraint. The system comprises a video acquisition module, a target component intelligent analysis functional module, a video content visual analysis module, a general behavior rule sample database and an alarm module. On the basis of behavior comprehension on target components, aiming at spatial correlation between targets, the system carries out behavior characteristic analysis, evolution and classification, establishes a spatial correlation mode sample library, implements detection and identification on abnormal behaviors, can more effectively and accurately carry out identification and judgment on emergencies and can lock an abnormity trigger point to provide evidence for post-mordem forensics. The system introduces target space relation constraint, divides a video scene region to complete defining related concepts of the target components of a video, eliminates a certain fuzziness and faults in space semantic information, solves the problems of incompletion and inaccuracy in video content analysis and is more accurate and more efficient to detect abnormal events in a video monitoring range.

Description

technical field [0001] The invention belongs to the field of intelligent early warning and relates to a video monitoring system and an intelligent early warning method for video monitoring, especially for public monitoring systems. Background technique [0002] The current development trend of video surveillance is to change from the passive mode of manual monitoring to the active mode of intelligent early warning. The key is the automation and intelligence of the video content analysis process. At present, video content analysis technology mainly adopts a data-driven idea. Faced with massive data and complex types, it uses itself and its context information to mine information, learn knowledge, and intelligently process videos. Usually, machine learning and content-based video retrieval and indexing methods are used to improve the problems of multi-instance, diversity and multi-modality that are difficult to solve by model-based methods, and can obtain information more effe...

AI Technical Summary

Problems solved by technology

Therefore, a "semantic gap" is formed between the low-level information and the high-level semantics, which makes it insufficient to analyze the high-level semantics of the video, and cannot fully meet the needs of intelligent early warning for real-time monitoring

[0006] On the other hand, due to the lack of exploration of space constraints, the analysis of video is usually uncertain and incomplete, which can only reflect part of the knowledge of the observation space, and there are technical "blind spots"

Images