Location Tracking Using Feature Matching and Epipolar Geometry
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional location tracking methods using inertial measurement units (IMU) sensors suffer from increasing drift errors over time, leading to inaccurate location values even when the object is stationary, and fail to enhance precision significantly with existing techniques.
Innovation Solution
A location tracking device and method that combines sensing information from an IMU sensor with image information from a camera, using feature matching and epipolar geometry to improve tracking accuracy by detecting and removing outliers, and recalculating the fundamental matrix to reflect image information for more precise object location tracking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If only IMU sensor is used for location tracking, then the device complexity is low, but the measurement precision deteriorates due to drift errors increasing over time
Solution Approach 1:
The patent combines IMU sensor data with camera image data into a unified location tracking system. The controller integrates sensing information (acceleration, gravity, direction) with image information from feature points to calculate fundamental matrices and track object locations, thereby improving measurement precision while maintaining reasonable device complexity through data fusion rather than hardware multiplication
2Productivity
If feature matching is performed without outlier removal, then the processing speed is high, but the measurement precision deteriorates due to inaccurate location values
Solution Approach 1:
The patent applies partial outlier removal by using epipolar geometry to identify and remove only the necessary outliers from feature matching results, rather than removing all matched points or using complex iterative methods. This partial action maintains high processing speed while sufficiently improving measurement precision by eliminating clearly erroneous matches
3Device complexity
If the fundamental matrix is not recalculated, then the device complexity is low, but the measurement precision deteriorates as the initial fundamental matrix becomes inaccurate over time
Solution Approach 1:
The patent implements periodic recalculation of the fundamental matrix at predetermined time intervals or when drift is detected. The controller recalculates the fundamental matrix using updated sensing information and image information, ensuring measurement precision is maintained over time while avoiding continuous recalculation that would increase device complexity and processing burden
Data Source
AI summary
This application relates to a location tracking device and method using a feature matching. The location tracking device may include a sensor, a camera, and a controller. The sensor is provided in a predetermined object and collects sensing information including at least one of speed, direction, gravity, and acceleration of the object. The camera is provided in the predetermined object and collects image information by capturing an image. The controller calculates an initial fundamental matrix (F0) by using the collected sensing information and calibration information of the camera, detects feature points of the image information, performs a feature matching in a fundamental matrix (F) by combining the initial fundamental matrix and the feature points, and tracks a location of the object by using a result of the feature matching.