Track-Before-Detect Algorithm Using Velocity Bins for Low-SNR Targets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current track-before-detect algorithms are computationally burdensome and complex, making them unsuitable for real-time processing environments, especially when dealing with targets having low signal-to-noise ratios in cluttered backgrounds.
Innovation Solution
The Recursive Estimation of Velocities, Energies, and Locations (REVEAL) technique uses dynamic programming and multiple motion models to estimate target energy, velocity, and location, suppressing noise and clutter through parallel processing and fusion of energy and velocity estimates across frames.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If particle filtering or velocity matched filters are used for track-before-detect, then detection capability for low SNR targets is improved, but computational burden and algorithm complexity increase significantly
Solution Approach 1:
The algorithm segments the search space by dividing velocity hypotheses into discrete velocity bins and spatial locations into grid cells. This segmentation transforms the continuous search problem into a discrete dynamic programming problem, reducing computational complexity while maintaining detection capability for low SNR targets.
Solution Approach 2:
The algorithm implements dynamic programming where the detection threshold and tracking decisions adapt dynamically based on accumulated evidence across multiple frames. The state transition probabilities and detection thresholds are updated recursively, allowing the system to adapt to varying SNR conditions without requiring complex real-time adjustments.
2Measurement precision
If particle filtering or velocity matched filters are used for track-before-detect, then detection capability for low SNR targets is improved, but real-time processing capability deteriorates
Solution Approach 1:
By segmenting the state space into discrete velocity bins and spatial grid cells, the algorithm reduces the computational burden from exponential to polynomial complexity. This enables real-time processing while maintaining the ability to detect low SNR targets through cumulative evidence accumulation across frames.
Solution Approach 2:
The algorithm changes the parameter representation from continuous particle distributions to discrete state probabilities in a reduced state space. This parameter transformation maintains detection sensitivity while enabling efficient computation suitable for real-time processing of video streams.
3Productivity
If traditional detection methods are used, then computational burden is reduced, but ability to detect targets in cluttered backgrounds with low SNR deteriorates
Solution Approach 1:
The algorithm performs preliminary tracking across multiple frames before making detection decisions. By accumulating evidence and maintaining state probabilities over time, the system prepares detection thresholds in advance based on temporal patterns, enabling reliable detection of low SNR targets while maintaining processing efficiency through the reduced state space.
Data Source
AI summary
Apparatus, system and method for tracking an image target in a system, wherein a system receives an image comprising a plurality of pixels. The received image is processed via a plurality of different recursive motion model kernels in parallel to provide a plurality of kernel outputs, wherein each of the motion model kernels may include a respective pixel mask. Per-pixel energy is estimated of at least some of the plurality of kernel outputs. Velocity of at least one of the image pixels may also be estimated by generating a directional energy vector for each motion model kernel. The per-pixel energy and velocity estimates are fused to produce a fused estimate representing at least some of the motion model kernels for the image.