Self-Checkout Item Verification Using Weight and Computer Vision
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional self-service kiosks require high buyer engagement, leading to inefficiencies and increased physical contact, which can spread contagious diseases, and struggle with transaction accuracy.
Innovation Solution
A self-service kiosk incorporating weight sensing technology and computer vision to facilitate frictionless transactions by using a weight sensor and cameras to calculate and verify item counts, with local object recognition to ensure accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If conventional self-service kiosks are used, then buyer engagement is required for scanning and placing products, but this increases physical contact and transaction time
Solution Approach 1:
The patent replaces manual scanning and item-by-item processing with weight sensing technology and computer vision systems. The weight sensor automatically detects items placed on the platform, and computer vision verifies item counts, eliminating the need for buyers to manually scan barcodes or carefully place each product, thus reducing both physical contact and transaction time
Solution Approach 2:
The system enables automatic item detection and verification without requiring buyer intervention. The kiosk autonomously identifies items through weight changes, counts them using computer vision, and processes transactions, allowing buyers to simply place items on the platform without engaging with scanning devices or following complex procedures
2Measurement precision
If conventional self-service kiosks require manual scanning and careful product placement, then transaction accuracy can be maintained, but physical contact increases spreading contagious diseases
Solution Approach 1:
The patent eliminates physical contact with scanning devices and product handling areas by using weight sensors to detect items and computer vision systems to verify item counts and identify products. This non-contact approach maintains transaction accuracy while preventing the spread of contagious diseases through reduced physical interaction with the kiosk and products
Solution Approach 2:
The system introduces weight sensors and computer vision cameras as intermediaries between the buyer and the transaction processing system. These intermediaries detect and verify items without requiring direct physical contact between the buyer and the kiosk components, thereby maintaining accuracy while reducing disease transmission risk
3Productivity
If weight sensing technology and computer vision are used, then transaction accuracy and speed are improved, but device complexity increases
Solution Approach 1:
The patent combines weight sensing technology and computer vision into a unified system that works together to detect, verify, and identify items. The weight sensor provides initial item detection, while computer vision confirms item counts and product identification, creating an integrated solution that improves transaction speed despite the added complexity of multiple sensing modalities
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
Enables efficient, accurate, and contact-free self-checkout by verifying item counts and weights, reducing transaction time and minimizing physical contact.
Implementation Method 1
A self-service kiosk incorporating weight sensing technology and computer vision to facilitate frictionless transactions by using a weight sensor and cameras to calculate and verify item counts
Implementation Method 2
A self-service kiosk incorporating weight sensing technology and computer vision to facilitate frictionless transactions by using a weight sensor and cameras to calculate and verify item counts
Data Source
AI summary
A self-service computing device detects a first weight for an item positioned on a sensory platform and determines a stabilized weight and a weight-based item count. Responsive to determining the stabilized weight, the device causes image capture devices to capture images of the item, identifies a visual characteristic of the item, and retrieves, from memory, a second weight of the item. To generate an image-based item count, the device compares the identified visual characteristic of the item with the second weight of the item and validates the presence of the item. The device causes or blocks a self-service transaction operation at the self-service computing device depending respectively on whether or not the image-based item count and the weight-based item count correspond.