POS Fraud Detection via Neural Network Image Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for detecting cashier fraud at point of sale (POS) systems using surveillance cameras and artificial neural networks face challenges in accuracy due to high computational complexity and the need for manual updating of product images, which is labor-intensive and inefficient.
Innovation Solution
A POS system that includes a barcode reader, image capture device, and data processing module using artificial neural networks to verify product images in real-time, automatically selecting and updating product images for learning, and dividing products into classes for specific neural networks to enhance fraud detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual selection and updating of product images for neural network learning is performed, then the learning dataset can be maintained, but the process becomes extremely labor consuming and inefficient
Solution Approach 1:
The system automatically selects product images for neural network learning without human intervention. The selection process is self-service, where the system retrieves images from the POS database based on product identifiers and automatically updates the learning dataset, eliminating the need for manual image collection and maintenance
Solution Approach 2:
The system pre-collects and stores product images in the POS database with their corresponding product information. This preliminary action ensures that when images are needed for neural network learning, they are already available and properly organized, eliminating the need for manual image gathering at the time of need
2Measurement precision
If the product database contains a large number of product images for comprehensive fraud detection, then detection accuracy improves, but computational complexity increases significantly
Solution Approach 1:
The system extracts only the necessary product images from the large database for neural network learning. Instead of using all available product images, it selectively retrieves images corresponding to products that have been scanned at the checkout, thereby reducing the computational burden while maintaining detection accuracy for relevant products
Solution Approach 2:
The verification process is segmented into two parts: first, the neural network verifies the scanned product image against learned patterns; second, the system compares the verification result with the scanned barcode data. This segmentation allows efficient processing by dividing the complex verification task into manageable stages
3Quantity of substance
If standard video surveillance methods are used to analyze cashier actions, then coverage of suspicious transactions is improved, but the analysis becomes labor consuming and inefficient
Solution Approach 1:
The system replaces manual video analysis with an automated neural network-based image verification system. Instead of reviewers manually examining video footage of cashier actions, the system automatically captures images at the checkout, verifies them against the neural network, and compares results with POS data, thereby eliminating time-consuming manual analysis while analyzing a larger quantity of transactions
Data Source
AI summary
A group of inventions relates to artificial neural networks and their application for computer vision, in particular for the surveillance camera data processing systems and methods to automatically detect cashier fraud by verifying images using artificial neural networks. To detect cashier fraud, a POS system includes a barcode reader, memory, an image capture device, and a data processing module. The data processing module is configured to receive the data about the scanned product from the product database and to receive the video data from the image capture device. An automatic generation of product image set for artificial neural network learning contains stages when the barcode is read by placing an item against a barcode reader by a cashier, the barcode data signal then provides the data about the scanned product from product database, when the barcode data signal gives the image of the placed product from the image capture device, the received image of the placed item is saved with the data about the scanned product in the product database, then the abovementioned stages are repeated for each item placed against the barcode reader.