Markerless Object Tracking with Metric-Based Beam Trajectory Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Markerless tracking systems face challenges in selecting suitable beam trajectories for accurate object tracking, especially when human operators are involved, and in compensating for movements due to roll, pitch, and heave, which can lead to inaccurate measurements and increased costs.
Innovation Solution
A metric-based system automatically assesses the suitability of beam trajectories for markerless tracking, allowing for the selection of optimal beam paths without human intervention, and compensates for nuisance degrees of freedom like roll, pitch, and heave by analyzing image variability and using scan-matching techniques to ensure accurate position and orientation measurements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If human operators select beam trajectories for markerless tracking, then the system can adapt to different objects, but the process becomes expensive and time-consuming due to training requirements and human error
Solution Approach 1:
The system enables automatic beam trajectory selection through computational algorithms that evaluate multiple trajectories and select the optimal one based on image quality metrics, eliminating the need for human operators to manually select trajectories while maintaining adaptability to different objects
Solution Approach 2:
The patent replaces the manual human operator selection process with an automated computational system that uses image processing and optimization algorithms to select beam trajectories, substituting mechanical human judgment with automated digital processing
2Adaptability or versatility
If the tracking device moves relative to the object, then the system can operate in dynamic environments, but the beam illuminates different portions of the object making accurate comparison difficult
Solution Approach 1:
The system pre-processes images by identifying and comparing only those portions that correspond to the same physical features across different time points, even when the beam illuminates different regions, thereby maintaining measurement accuracy in dynamic environments
Solution Approach 2:
The patent dynamically adjusts beam trajectories and image processing parameters in response to detected object motion or tracking device movement, maintaining optimal illumination of suitable features while compensating for relative movement through real-time parameter modification
3Measurement precision
If a small beam is used for tracking, then the system achieves higher precision, but the choice of beam trajectory becomes more critical as fewer portions of the object are illuminated
Solution Approach 1:
The system evaluates multiple candidate beam trajectories by simulating or pre-processing images along each trajectory, using image quality metrics and feature analysis to provide feedback on trajectory suitability, then selects the optimal trajectory based on this feedback
Solution Approach 2:
The patent performs preliminary evaluation of beam trajectories by analyzing image data to identify portions with suitable features before finalizing the beam trajectory selection, ensuring that the chosen trajectory illuminates regions with distinguishable features appropriate for markerless tracking
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system improves the reliability and accuracy of markerless tracking by automating beam trajectory selection, reducing human error, and maintaining accurate measurements even under movements caused by roll, pitch, and heave, thus enhancing operational efficiency and cost-effectiveness.
Implementation Method 1
a beam (radar, lidar or sonar) is directed to follow a beam trajectory which illuminates a portion of the object
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A markerless tracking device (100) has a source, a sensor and a processor. The source is configured to direct a beam on a plurality of beam paths (124), where each beam path (124) illuminates a distinct region (142) of an object (170). The sensor is configured to receive beam (126) reflected from each region (142) in order to generate an image of each region (142). A processor is configured to generate a metric which indicates suitability of a beam trajectory (132), where the beam trajectory (132) is a path over which the beam is to be directed by the source during markerless tracking of the object (170). The metric is generated by comparing images of selected regions (142).