Shopping System Gesture Recognition for Order Determination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional checkout processes in physical stores are inefficient due to inaccurate determination of order information, often caused by multiple individuals selecting items from lower shelves or items with high similarity being misidentified, leading to lengthy waiting times for consumers.
Innovation Solution
A shopping management system utilizing facial recognition, RFID labels, and gesture recognition technology to automatically identify selected commodities and associate them with consumers, enabling self-service payment without queuing at checkout counters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If computer vision technology is used to analyze consumer selection and purchase processes in real time, then payment efficiency is improved, but determination accuracy deteriorates due to factors like multiple people taking commodities from lower shelves or shielded identification devices
Solution Approach 1:
The system divides the monitoring area into multiple zones with different camera angles and perspectives. Multiple cameras capture images from various positions, allowing the system to segment the observation task across multiple viewing angles. This enables accurate identification of commodities even when some views are blocked, as other cameras provide unobstructed perspectives.
Solution Approach 2:
The system introduces an intermediary verification mechanism where multiple camera views act as intermediate observation points. When one camera's view is shielded or ambiguous, the system uses images from other cameras as intermediary evidence to verify commodity identification, thereby maintaining high determination accuracy while enabling real-time payment processing.
2Measurement precision
If multiple cameras are deployed to improve determination accuracy, then measurement precision is improved, but device complexity increases
Solution Approach 1:
Each camera in the system is designed to serve multiple functions: primary commodity identification, backup verification for other cameras, and crowd behavior monitoring. This multi-functionality allows the system to achieve high determination accuracy with a moderate number of cameras, as each camera contributes to multiple objectives simultaneously, reducing overall system complexity.
Solution Approach 2:
The system merges the functions of multiple cameras into a unified image processing pipeline. Instead of treating each camera independently, the system combines images from multiple cameras and processes them through a single analysis algorithm, simplifying the control architecture and reducing system complexity while maintaining high determination accuracy through combined visual information.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present disclosure provides an order information determination method and apparatus. The method is used to determine an association between a user and a commodity selected and purchased by the user and includes: performing human gesture recognition on a user to obtain gesture data of the user; positioning a commodity to obtain location information of the commodity; determining whether a gesture of the user is taking the commodity based on the gesture data and the location information; and adding the commodity to an order of the user if a determination result is that the user takes the commodity. In the present disclosure, shopping efficiency is improved and fairly good shopping experience is provided.