People-flow analysis using differential tracking and parameter adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image analysis methods for people-flow analysis are costly and require multiple camera devices, with limited accuracy in identifying and tracking moving pedestrians.
Innovation Solution
A people-flow analysis system that includes an image source, a computing device, and a host, using different neural network models for detecting and tracking pedestrians by generating position boxes in images and adjusting parameters based on detection and tracking results to improve analysis accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional image analysis methods are used with multiple camera devices at different visual angles, then detection coverage is improved, but system cost and complexity increase
Solution Approach 1:
A single camera device performs multiple functions: initial pedestrian detection, tracking across frames, and people-flow analysis. The system makes the camera device universal by integrating detection and tracking capabilities in one unit rather than requiring separate devices for each function.
Solution Approach 2:
The patent introduces an intermediary tracking image generated from differential processing between consecutive frames. This tracking image acts as a mediator that bridges the gap between initial detection and final people-flow analysis, enabling continuous tracking without additional camera devices.
2Measurement precision
If conventional image analysis methods are used, then basic pedestrian detection is achieved, but tracking accuracy and people-flow analysis precision are insufficient
Solution Approach 1:
The patent segments the image processing task into distinct components: initial pedestrian detection from the first image, tracking using differential images between consecutive frames, and final people-flow analysis. This segmentation allows each component to be optimized independently, improving overall tracking accuracy and analysis efficiency.
Solution Approach 2:
The system performs preliminary detection in the first image to identify pedestrians before proceeding to tracking in subsequent frames. This preliminary action establishes baseline position boxes that guide the tracking process, improving both accuracy and efficiency by focusing computational resources on relevant areas.
3Measurement precision
If parameter adjustment is performed manually in conventional systems, then detection results can be optimized, but time consumption and operational complexity increase
Solution Approach 1:
The system performs self-service by automatically adjusting detection and tracking parameters based on feedback from the tracking process. The computing device monitors detection results and dynamically optimizes parameters without manual intervention, maintaining high detection accuracy while eliminating time-consuming manual adjustments.
Solution Approach 2:
The patent implements a feedback mechanism where detection results from position boxes are fed back to automatically adjust detection and tracking parameters. This closed-loop feedback system continuously optimizes detection accuracy by learning from actual tracking performance, eliminating the need for manual parameter tuning.
Data Source
AI summary
A people-flow analysis system includes an image source, a computing device, and a host. The image source captures a first image and a second image. The computing device is connected to the image source. The computing device identifies the first image according to a data set to generate a first detecting image. The first detecting image has a position box corresponding to a pedestrian in the first image. The computing device generates a tracking image according to the data set and a difference between the first detecting image and the second image. The tracking image has another position box corresponding to a pedestrian in the second image. The host is connected to the computing device and generates a people-flow list according to the first detecting image and the tracking image.