A Three-Dimensional Hough Transform-Based Track-Before-Detect Technique for Detecting Extended Targets in Strong...

Area of Science:

• Signal Processing
• Target Tracking
• Computer Vision

Background:

• Hough Transform (HT) is effective for trajectory detection in cluttered environments, offering robustness against noise and faults.
• Tracking extended targets in 3D space, incorporating positional and temporal data, presents significant challenges.
• Existing methods like GM-PHD filters and 4DHT-TBD algorithms have limitations in handling high noise and sparse measurements.

Purpose of the Study:

• To extend the Hough Transform to three-dimensional data for enhanced trajectory detection.
• To develop a novel track-before-detect technique for simultaneously tracking extended and non-extended targets.
• To improve tracking efficiency and reduce false alarms in cluttered environments with high noise levels.

Main Methods:

• Extension of the Hough Transform to 3D by incorporating the measuring time as a temporal axis.
• Implementation of a 3D accumulator matrix for iterative voting and measurement selection.
• Development of the three-dimensional Hough Transform-based extended target track-before-detect (3DHT-ET-TBD) technique.

Main Results:

• The 3DHT-ET-TBD technique demonstrates suitability for tracking both extended and non-extended targets simultaneously.
• The proposed method effectively suppresses noise and clutters, leading to fewer false alarm trajectories.
• Performance evaluation using real and simulated data shows superior efficiency and lower computational cost compared to existing methods.

Conclusions:

• The 3DHT-ET-TBD is a promising approach for multi-extended target tracking, especially in challenging conditions with high noise and sparse measurements.
• The method offers high efficiency and low computation, making it suitable for real-time applications.
• The 3D extension of HT provides a robust framework for complex target tracking scenarios.