Commodity Trajectory Analysis in Intelligent Carts for Stacked Items
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing intelligent shopping carts lack accuracy in detecting abnormal shopping behaviors, such as missed or wrong scanning, over-picking, and shielding, due to commodities being stacked and shielded, which affects detection results.
Innovation Solution
The method involves acquiring code scanning data and video-frame image data, segmenting commodities in each frame, tracking their trajectories, determining motion direction and distance, and detecting abnormal behaviors based on these data using machine learning algorithms and deep learning models to accurately identify whether commodities are put in or taken out of the cart.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional detection methods are used for intelligent shopping carts, then the system is simple and easy to operate, but the detection accuracy of abnormal shopping behaviors is low
Solution Approach 1:
The detection system is segmented into multiple independent modules: video acquisition module, image processing module (with segmentation unit), trajectory analysis module, and behavior detection module. Each module handles a specific aspect of the detection process, improving overall accuracy while maintaining manageable complexity through modular design.
Solution Approach 2:
The system transitions from traditional single-point detection to multi-dimensional analysis by tracking commodity trajectories across multiple video frames. This adds temporal and spatial dimensions to the detection process, enabling accurate identification of abnormal behaviors despite increased system complexity.
2Quantity of substance
If commodities are stacked in the shopping cart basket, then the shopping capacity is increased, but the detection of individual commodities becomes difficult due to shielding
Solution Approach 1:
The system performs preliminary segmentation of commodities in each video frame before trajectory tracking. By pre-processing the image data to separate individual commodity regions, the system maintains detection accuracy even when commodities are stacked, allowing the basket to hold more items without compromising detection quality.
Solution Approach 2:
The image processing module acts as an intermediary between the video camera and the behavior detection system. It processes the raw video data to extract and separate individual commodity information, enabling accurate tracking even when commodities are physically stacked and visually shielded from each other.
3Productivity
If self-service shopping is implemented, then the shopping efficiency is improved and cashier interaction is reduced, but abnormal behaviors such as missed scanning and over-picking are more likely to occur
Solution Approach 1:
The system continuously monitors commodity trajectories and provides real-time feedback on shopping behavior compliance. By comparing actual commodity movements with expected scanning patterns, the system can identify and alert abnormal behaviors like missed scanning or over-picking, maintaining reliability while preserving self-service efficiency.
Solution Approach 2:
The abnormal behavior detection system operates autonomously without requiring cashier intervention. It automatically analyzes video data, tracks commodities, and detects violations, enabling the shopping process to remain efficient and self-service oriented while still ensuring behavioral compliance through automated monitoring.
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
An abnormal shopping behavior detection method and apparatus for an intelligent shopping cart, and the shopping cart are disclosed in the present disclosure. The method includes: acquiring code scanning data of commodities and video-frame image data of a basket area during a shopping behavior of a user; segmenting the commodities in an image of each frame in the video-frame image data of the basket area, to obtain image data of each target commodity in the basket area in the image of each frame; tracking a trajectory of each target commodity according to the image data of each target commodity in the basket area among images of a plurality of frames; determining a motion direction and a motion distance of the tracked trajectory of each target commodity according to the trajectory tracking data; determining an indicative state of whether each target commodity is put in or taken out of the shopping cart, according to the motion direction and the motion distance; and detecting an abnormal shopping behavior of the user, according to the code scanning data, the trajectory tracking data and the indicative state of whether each target commodity is put in or taken out of the shopping cart. The present disclosure can accurately detect the abnormal shopping behavior for the intelligent shopping cart.