Predictive Theft Notification via Biometric Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Retail stores face challenges in preventing Organized Retail Crime (ORC) and shoplifting, as existing Electronic Article Surveillance (EAS) systems often trigger alarms too late to be effective, and there is a need to distinguish between repeat offenders and genuine shoppers to minimize false enrollments in facial recognition databases.
Innovation Solution
Implementing a predictive theft notification system that uses imaging devices to capture and process facial images, comparing them to biometric models in a database, and applying a rules-based enrollment process to select and deselect images for enrollment based on distance, size, and other criteria, while using machine learning to predict potential theft events and reduce false notifications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If EAS alarm is triggered when thieves leave the store, then theft detection is achieved, but the alarm is too late to prevent theft
Solution Approach 1:
The system captures facial images and performs biometric matching before the theft is completed. When an EAS alarm is triggered, the system has already recorded the offender's facial image and can immediately identify them by comparing against the database, enabling preventive action rather than reactive response.
Solution Approach 2:
The system divides the monitoring process into distinct stages: continuous facial image capture, real-time biometric matching against the database, and separate alert generation. This segmentation allows the system to identify offenders at different points in the shopping journey, not just at the exit.
2Reliability
If all captured facial images are enrolled in the database, then repeat offenders can be identified, but false enrollments of genuine shoppers increase
Solution Approach 1:
The system uses feedback from EAS alarm events to selectively enroll only those facial images associated with actual theft incidents. When an EAS alarm triggers, the system captures the offender's image and enrolls it in the database, using the alarm event as feedback to distinguish genuine offenders from regular shoppers.
Solution Approach 2:
The system applies different enrollment criteria to different contexts: images captured during EAS alarm events are enrolled, while images from normal shopping activities are not. This localized approach to image enrollment ensures that only relevant offender images are stored, minimizing false positives.
Data Source
AI summary
Predictive theft notifications are used to coordinate appropriate responses to persons who are likely to commit acts of theft. Image data is generated and processed in a computer processing device to recognize the presence of a facial image comprising a face of a person. An analysis is performed of data representative of the facial image to determine a biometric match relative to one or more biometric models of facial images stored in a database. Based on this analysis, at least one predictive notification is generated with regard to a future potential theft of merchandise from the secured facility. The predictive notification is generated based upon a determination of the biometric match.