Portable Terminal Acceleration Data for Crowd Density Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating congestion degree and number of people, such as those using GPS data and surveillance cameras, face challenges in accuracy and cost-effectiveness, particularly in small regions and tourist areas where camera installation is restricted.
Innovation Solution
A congestion degree estimation method utilizing acceleration data from portable terminal devices to calculate the standard deviation of acceleration, which serves as an index value for determining congestion levels, allowing for accurate estimation without the need for surveillance cameras.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If GPS-based estimation methods are used to estimate the number of people, then the estimation can be performed without surveillance cameras, but the accuracy is insufficient especially in small regions
Solution Approach 1:
The patent combines GPS positional information from multiple portable terminals with acceleration sensor data to create a composite estimation method. By merging location data with motion state data, the system achieves higher estimation accuracy in small regions compared to using GPS alone, while maintaining the advantage of not requiring surveillance camera infrastructure.
Solution Approach 2:
The patent introduces acceleration sensor data as an intermediary parameter to improve the relationship between GPS positional information and crowd density estimation. The acceleration data serves as a mediator that captures motion patterns and behavioral characteristics, enabling more accurate inference of crowd conditions without direct visual observation from cameras.
2Measurement precision
If surveillance cameras are installed to estimate the number of people, then high accuracy can be achieved, but the cost increases significantly
Solution Approach 1:
The patent uses data from portable terminals that users already carry as a 'copy' or alternative source of information, replacing the need for expensive surveillance camera infrastructure. The acceleration sensor data and GPS information from mobile devices serve as substitutes for direct camera-based observation, achieving cost-effective crowd estimation.
Solution Approach 2:
The patent leverages the portable terminals themselves as the measurement devices. Users' own mobile devices, which they already possess and use for navigation, automatically collect and transmit the necessary data. This self-service approach eliminates the need for external surveillance infrastructure, significantly reducing installation and maintenance costs.
3Quantity of substance
If estimation is performed for large regions using GPS data, then sufficient data can be collected, but the required time from current location to destination is not accurately estimated
Solution Approach 1:
The patent divides the estimation process into segments: first determining the user's motion state (walking, standing, etc.) using acceleration data, then applying appropriate estimation algorithms based on the segment. This segmentation allows for more precise local estimation in small regions rather than relying on averaged data from large regions, improving time prediction accuracy for specific destinations.
Solution Approach 2:
The patent dynamically adjusts the estimation approach based on real-time acceleration data and motion state changes. By continuously monitoring whether the user is walking, standing, or moving, the system adapts its estimation methodology to match current conditions, enabling more accurate prediction of arrival times compared to static large-region averaging.
Data Source
AI summary
A method of estimating the congestion degree and the number of people with high accuracy and at a low cost, without requiring a surveillance camera, is provided. Provided are a congestion degree estimating step of estimating a congestion degree based on sensor information (such as acceleration information) of a portable terminal device, an average congestion degree calculating step of calculating an average congestion degree for each mesh (reference range) set in advance, based on the congestion degree estimated in the congestion degree estimating step for a plurality of portable terminal devices, and an estimated number-of-people calculating step of calculating an estimated number of people for each mesh by multiplying the average congestion degree calculated in the average congestion degree calculating step by a base number of people (maximum number of accommodated people) determined in advance for each mesh.