Self-checkout Kiosk Vision Camera Overlap for Shrinkage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Self-checkout kiosks with barcode readers and vision cameras face issues such as users not scanning items, incorrect item scanning, and unauthorized purchases of age-restricted items, leading to inventory shrinkage, as they are not always monitored by employees.
Innovation Solution
A self-checkout kiosk design incorporating a housing with a barcode reader module, multiple vision cameras, and a central processing unit, where the vision cameras provide overlapping fields of view to enhance item recognition and facial recognition, allowing for improved monitoring and security features like scan avoidance detection and age-restricted item control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single vision camera is used above the barcode reader, then the device complexity is reduced, but the measurement precision of item recognition and scan avoidance detection is insufficient
Solution Approach 1:
The vision system is segmented into multiple independent cameras positioned at different locations: a first vision camera above the barcode reader for top-down viewing, and a second vision camera at the front for face-to-face viewing. This segmentation allows each camera to capture specific angles and aspects of scanned items, improving recognition accuracy without requiring a single complex camera system.
Solution Approach 2:
The patent adds a spatial dimension to the vision system by positioning cameras at different vertical and horizontal locations. The first camera provides a top-down perspective while the second camera provides a front-facing perspective, creating multi-dimensional coverage that enhances item recognition and scan avoidance detection capabilities.
2Reliability
If multiple vision cameras with overlapping fields of view are deployed, then the reliability of scan monitoring improves, but the device complexity increases
Solution Approach 1:
Multiple vision cameras are merged into a coordinated system where the first and second cameras work together with the barcode reader. The cameras are positioned to create overlapping fields of view that collectively cover the scanning region, ensuring reliable detection of scan avoidance while distributing the monitoring function across multiple components.
Solution Approach 2:
The system implements feedback mechanisms where the vision cameras continuously monitor the scanning process and provide real-time data to the processor. This feedback loop enables immediate detection of scan avoidance attempts and allows the system to respond by alerting employees or preventing transaction completion, thereby improving monitoring reliability.
3Reliability
If advanced image processing is implemented for facial recognition and item recognition, then the security features improve, but the processing time and computational requirements increase
Solution Approach 1:
The system performs preliminary actions by capturing images continuously as items are brought to the scanning position, before the actual scanning transaction occurs. This allows the processor to pre-process and analyze images in real-time, reducing the computational burden during the critical transaction moment and minimizing processing delays.
Solution Approach 2:
The vision cameras capture more image data than strictly necessary by providing overlapping fields of view and multiple angles. This excessive action ensures that sufficient image data is available for reliable facial recognition and item recognition even if some portions of images are obscured or suboptimal, improving security reliability while allowing the system to process only the most relevant image portions.
Data Source
AI summary
A kiosk includes a barcode reader module, vision camera, and central processing unit positioned within a housing. The housing defines a front surface and has an upper portion that extends outward past the front surface. The barcode reader module includes an imaging assembly with a first field-of-view that defines a product scanning region. The vision camera is positioned within the upper portion of the housing with at least a portion of the vision camera located forward of the front surface of the housing and has a second field-of-view facing generally downward and intersecting the first field-of-view of the barcode reader module within the product scanning region. The central processing unit is operatively connected to the barcode reader module and the vision camera and is configured to decode barcodes captured by the barcode reader module and perform advanced processing of images received from the vision camera.