Target Tracking in Water Areas Using Neural Network Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing boat safety systems face challenges in accurately and efficiently tracking objects in water areas due to limited computing capacity on board, leading to potential collisions with close or dangerous objects.
Innovation Solution
A tracking system comprising a pan-tilt unit, a camera module, and a control module implementing a neural network, which continuously generates a video stream, detects objects, and calculates pan and tilt angles to center the camera on selected objects, thereby enhancing tracking accuracy and frequency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If distance and trajectory calculations are performed for all detected objects, then tracking accuracy is improved, but computing load increases
Solution Approach 1:
The patent extracts and focuses computational resources on a specific target object rather than processing all detected objects. The system identifies a particular object of interest and concentrates tracking calculations on this selected object, thereby reducing overall computing load while maintaining high tracking accuracy for the critical target.
Solution Approach 2:
The patent applies different processing qualities to different objects: high-precision tracking calculations are applied locally to the selected target object, while other objects receive standard detection processing. This localized quality approach optimizes computational resource allocation by intensifying processing only where necessary.
2Reliability
If tracking calculations are performed at high frequency, then collision risk is reduced, but computing capacity requirements increase
Solution Approach 1:
The patent extracts high-frequency tracking calculations from the general object detection process and applies them specifically to the selected target object. This allows the system to perform frequent distance and trajectory updates for the critical object without requiring high-frequency processing for all detected objects, thus reducing overall computing capacity requirements.
3Loss of information
If all detected objects are tracked, then detection completeness is improved, but system complexity increases
Solution Approach 1:
The patent extracts the complex tracking functionality and applies it selectively to one selected object rather than implementing universal tracking for all detected objects. This maintains detection completeness by still detecting all objects, but reduces system complexity by concentrating sophisticated tracking resources on a single target.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
The invention relates to a method for tracking an object (5) in a water area (2) using a tracking system (30), comprising a tracking phase including controlling a pan-tilt unit to orientate a camera module around the pan axis and the tilt axis to centre the field of view of said camera module on a selected object (5) and thus track said selected object (5).