People Stream Analysis Using Possession Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing people stream analysis methods struggle to accurately determine the order of stores visited by individuals, especially when face recognition is not precise and possessions carried by customers are not considered, leading to difficulties in analyzing consumer behavior for sales promotion in commercial facilities.
Innovation Solution
A people stream analysis method that captures external appearance images, recognizes individuals, identifies possessions such as bags with store logos, and infers the order of stores visited by analyzing changes in possession lists over time, without the need for cameras in each store, thereby determining the order of store visits with higher accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If face recognition is used to identify customers, then customer identification can be achieved, but accuracy decreases when face recognition is not precise
Solution Approach 1:
The patent introduces possessions (bags, boxes, packages) as an intermediary object to link customers to stores. Instead of relying solely on direct face recognition, the system uses possessions carried by customers as mediating evidence of store visits. The recognition unit identifies possessions in images, and the inference unit connects these possessions to specific stores, thereby indirectly determining store visit orders when direct face recognition is unreliable.
Solution Approach 2:
The patent creates a copy or proxy relationship where possessions serve as substitutes for direct customer identification. Rather than requiring perfect face recognition, the system captures images of possessions (bags, boxes) that customers carry, which act as proxies indicating which stores the customer has visited. This copying approach allows the system to determine store visit orders even when face recognition fails or lacks precision.
2Loss of information
If cameras are installed in each store to monitor customer visits, then store visit information can be collected, but system complexity increases
Solution Approach 1:
The patent makes the image capture system multi-functional by using a single camera to perform multiple tasks: capturing customer images for identification, capturing possession images for store inference, and capturing images across different locations (facility entrance, passages, stores). This universal camera system eliminates the need for separate camera installations in each store while still collecting comprehensive store visit information.
Solution Approach 2:
The patent extracts the essential function of store monitoring from the physical presence of cameras in each store. Instead of requiring cameras in every store, the system extracts and uses only the necessary visual information (possessions carried by customers) that can be captured from strategic locations. This extraction approach reduces the camera system to its core function while maintaining information collection effectiveness.
3Device complexity
If possessions are not considered in analysis, then the system remains simple, but store visit order accuracy decreases
Solution Approach 1:
The patent performs preliminary action by having the recognition unit identify possessions in advance before the inference unit uses them for store determination. The system proactively captures and records possession information (types, characteristics) as customers pass through monitored areas, preparing this data ahead of time. This preliminary identification of possessions enables the subsequent inference unit to accurately determine store visit orders without adding complex real-time analysis requirements.
Data Source
AI summary
A people stream analysis apparatus includes an image information capturer that captures an external appearance image of a person, a person recognizer that recognizes the person from the external appearance image, a store inferrer that identifies from the external appearance image a possession carried by the person, and infers from the identified possession a store from which the possession has been obtained, a memory that stores, in an associated form, person information indicating the recognized person, store information indicating the inferred store, and time information indicating time at which the external appearance image has been captured, and an arrival store order determiner that determines an order of stores in which the person has visited, based on a change in a time sequence of listing of stores indicated by the store information stored on the memory.