Gait-Based Security Event Characterization for Proactive Theft Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic article surveillance (EAS) systems in retail stores primarily provide alarms after theft has occurred, failing to proactively identify potential theft offenders, and facial recognition systems are easily defeated and raise privacy concerns.
Innovation Solution
A security system that utilizes gait and body biometrics, combined with wireless technologies, to create a confidence score for identifying potential theft offenders before they enter a store, using millimeter wave transceivers and cloud-based databases to monitor and compare biometric data across multiple facilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If facial recognition systems are used to identify potential theft offenders, then identification capability is improved, but the system is easily defeated and raises privacy concerns
Solution Approach 1:
The patent replaces facial recognition (optical system) with gait recognition using millimeter wave transceivers (electromagnetic wave system). This substitution makes the system harder to defeat since gait patterns are more difficult to disguise or spoof compared to facial features, while also addressing privacy concerns by not requiring direct facial imaging
Solution Approach 2:
The system changes the biometric parameter being measured from facial features to gait characteristics. By detecting and analyzing walking patterns through millimeter wave transceivers, the system achieves reliable identification that is harder to defeat while maintaining privacy standards
2Reliability
If traditional EAS systems are used to detect theft, then alarm function is provided, but the system cannot proactively identify potential theft offenders
Solution Approach 1:
The system performs preliminary identification of potential theft offenders by analyzing gait patterns before they enter the store. By proactively identifying suspicious individuals in advance and generating alerts, the system enables security personnel to take preventive action rather than merely reacting to alarms after theft has occurred
Solution Approach 2:
The system implements feedback by continuously monitoring gait patterns, comparing them against a database of known offenders, and providing real-time alerts to security personnel. This closed-loop feedback mechanism enables proactive identification and immediate response to potential theft threats
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system achieves a high identification rate of up to 97% for repeat theft offenders by proactively identifying individuals based on gait and body biometrics, without the drawbacks of facial recognition, and can be implemented as a service across disparate retailers.
Implementation Method 1
using millimeter wave transceivers and cloud-based databases to monitor and compare biometric data
Data Source
AI summary
Security event characterization and response includes, for each of a plurality of first security data collection events determining that the first event has started. Collecting, in a database, an event data record associated with the first event determined to have started, the event data record comprising gait information of a person associated with the first event. For a given security data collection event later than the plurality of first events, determining that the given event has started. Detecting given event data comprising gait information of a person associated with the given event determined to have started. Determining a similarity meeting a threshold similarity between the query and one or more sets of records of the database. Initiating an action based on the one or more sets of records of the database determined to meet a threshold similarity with the query.