Security System Occlusion Detection for Suspicious Behavior Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current security systems fail to effectively identify suspicious behavior, such as sudden changes in appearance, in camera images captured by surveillance cameras, which can hinder crime prevention efforts.
Innovation Solution
A security system that uses a neural network model to track individuals across time-series camera images, detects occlusions, and determines suspicious behavior by assessing the reliability of tracking processes, thereby identifying and alerting authorities to potential threats.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If tracking processing is performed to identify persons in time-series camera images, then the ability to detect suspicious behavior is improved, but the system may produce false positives when occlusion occurs
Solution Approach 1:
The system performs preliminary occlusion detection before executing suspicious behavior determination. By detecting occlusion states in advance and prohibiting suspicious behavior determination when occlusion is detected, the system avoids false positives while maintaining reliable tracking functionality.
2Measurement precision
If the system prohibits suspicious behavior determination when occlusion occurs, then false positive rates are reduced, but the system cannot detect actual suspicious behavior during occlusion periods
Solution Approach 1:
The system applies partial action by selectively prohibiting suspicious behavior determination only during occlusion periods while continuing tracking processing. This partial restriction maintains detection accuracy for visible periods while accepting temporary monitoring gaps during occlusion, balancing precision with continuous surveillance capability.
3Measurement precision
If the system uses neural network models for person identification, then tracking accuracy is improved, but processing time and computational resources increase
Solution Approach 1:
The system extracts and separates occlusion detection as an independent preliminary processing step before suspicious behavior determination. This extraction allows the neural network to focus on identification tasks while occlusion detection handles the preliminary filtering, improving overall processing efficiency without sacrificing identification accuracy.
Data Source
AI summary
A security system monitors a behavior of a person included in a plurality of camera images captured continuously in time series by a surveillance camera. The security system performs tracking processing of detecting each person region of a person included in a plurality of camera images, and identifying a person included in the plurality of camera images in time series based on person image data included in the person region. When the identification by the tracking processing transitions to failure in the middle, it is determined that a person having suspicious behavior is included in the camera image. When it is determined that the suspicious behavior is included, alert information may be notified from an output device.