Portable Checkout Labeling for Automated Product Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional checkout systems in stores are inefficient, leading to long wait times and high costs due to the need for multiple cashiers, while automated systems face challenges in accurately identifying products due to occlusion and lack of trained data, requiring expensive human-labeled images for machine-learning models.
Innovation Solution
A portable checkout unit that automatically generates training data by scanning items and capturing images, using cameras and sensors to create labeled image data, which can be used to train automated checkout systems, reducing development costs and improving product identification accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional cashiers manually scan each item, then product identification accuracy is maintained, but checkout speed and customer service efficiency deteriorate due to long wait times
Solution Approach 1:
The system enables automated self-service checkout where the shopping cart itself performs product identification and checkout functions. Sensors and cameras in the cart automatically detect products as they are placed in the cart, eliminating the need for cashiers to manually scan each item, thereby dramatically improving checkout speed and reducing customer wait time
Solution Approach 2:
The patent replaces the manual mechanical scanning process with automated sensor-based detection systems. Optical sensors, cameras, and other detection devices automatically identify products through image recognition and sensor data, substituting the cashier's manual scanning action and enabling parallel processing of multiple items simultaneously
2Productivity
If multiple cashiers and POS systems are deployed to reduce wait times, then checkout speed improves, but operational cost and system complexity worsen
Solution Approach 1:
The shopping cart system performs multiple functions including product detection, image capture, data processing, and checkout execution all in one device. This multi-functional cart replaces the need for separate cashiers and multiple POS systems, achieving high checkout throughput while reducing overall system complexity
Solution Approach 2:
Each shopping cart operates as an independent automated checkout station, eliminating the need for human cashiers. The self-service capability of each cart allows parallel processing of multiple customers simultaneously, achieving high throughput without increasing staff or system complexity
3Measurement precision
If automated checkout systems use machine-learned models with human-labeled images, then product identification accuracy improves, but training data generation cost and time worsen
Solution Approach 1:
The system automatically generates its own training data through the sensors and cameras embedded in the shopping carts. As customers shop, the system captures real-world images and sensor data, automatically labeling them through the detection process itself, eliminating the need for manual human labeling and dramatically reducing training data generation time
Solution Approach 2:
The system performs preliminary data collection and labeling during normal shopping operations before actual deployment. Training data is accumulated in advance through continuous operation, allowing the machine learning models to be trained with extensive real-world data without requiring separate data collection campaigns
4Measurement precision
If cameras capture more detailed product images, then product identification accuracy improves, but system complexity and data processing requirements worsen
Solution Approach 1:
The camera system is divided into multiple independent cameras positioned at different locations on the shopping cart. Each camera captures images from its specific viewpoint, and the system processes these segmented views separately before combining them. This segmentation reduces the complexity of any single camera system while maintaining comprehensive product identification capability
Data Source
AI summary
A portable checkout unit automatically generates training data for an automatic checkout system as a customer collects items in a store. A customer uses an item scanner of portable checkout unit to generate a virtual shopping list of items collected in the shopping cart. When the customer adds a new item to the shopping cart or on some regular interval, the portable checkout unit captures images of the items contained by the shopping cart and can generate bounding boxes for each product in each image. The bounding boxes can be associated with item identifiers from previously-generated bounding boxes to identify the items captured by the bounding boxes. Each bounding box paired with an item identifier can then be used as training data for an automated checkout system.