Indoor Positioning via Bounding Box Covariance Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image-based indoor positioning systems struggle to provide precise location estimation due to the reliance on visual cues and spatial features, often failing to accurately localize users within smaller indoor spaces, especially when landmarks are missing or subject to variations in object detection.
Innovation Solution
The method involves capturing images, defining bounding boxes around detected objects, calculating mean and covariance to determine distances, and using known object positions to calculate the position of the image capturing device, while also accounting for drift errors through statistical analysis and machine learning algorithms like YOLO for object recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image-based object detection and positioning is performed using bounding boxes, then positioning precision is improved, but sensitivity to detection variations and noise increases
Solution Approach 1:
The system performs preliminary object detection and bounding box definition on multiple images before calculating positioning. By pre-processing multiple images and their corresponding bounding boxes, the system accumulates sufficient data to compute mean and covariance values, which then serve as a robust basis for position determination that is less sensitive to individual detection variations
Solution Approach 2:
The system calculates covariance of bounding box positions and uses this statistical feedback to assess detection reliability. When covariance values indicate high variability, the system can adjust its positioning calculations or request additional images, creating a feedback loop that maintains reliability while achieving precision
2Measurement precision
If multiple images are processed to calculate mean and covariance, then positioning accuracy is improved, but processing time increases
Solution Approach 1:
The system processes multiple images (excessive action) to ensure sufficient statistical data for accurate mean and covariance calculation, but can stop early if the covariance stabilizes or reaches a threshold, avoiding unnecessary processing time while maintaining positioning accuracy
3Adaptability or versatility
If object detection is performed in dynamic indoor environments, then adaptability is improved, but measurement precision deteriorates due to varying lighting and object variations
Solution Approach 1:
The system changes the parameter from individual bounding box coordinates to statistical parameters (mean and covariance) that describe the distribution of detected object positions. This transformation makes the positioning system adaptable to variations in lighting and object appearance while maintaining precision through the statistical central tendency
Data Source
AI summary
There is provided a method performed by an apparatus for positioning comprising receiving images of a video sequence for object detection; searching an object from the received images; defining a bounding box around a detected object in the received images; calculating mean and covariance to the detected object; determining, based on the mean and covariance, a position of at least one of the following: the apparatus, the detected object, or an image capturing device; and determining a drift error of the determined position.