Target detection performance bounds in compressive imaging

@article{Krishnamurthy2012TargetDP,
  title={Target detection performance bounds in compressive imaging},
  author={Kalyani Krishnamurthy and Rebecca M. Willett and Maxim Raginsky},
  journal={EURASIP Journal on Advances in Signal Processing},
  year={2012},
  volume={2012},
  pages={1-19}
}
This article describes computationally efficient approaches and associated theoretical performance guarantees for the detection of known targets and anomalies from few projection measurements of the underlying signals. The proposed approaches accommodate signals of different strengths contaminated by a colored Gaussian background, and perform detection without reconstructing the underlying signals from the observations. The theoretical performance bounds of the target detector highlight… 
Compressive measurement designs for estimating structured signals in structured clutter: A Bayesian Experimental Design approach
TLDR
Experimental results on synthetic data demonstrate that the proposed approach outperforms traditional random compressive measurement designs, which are agnostic to the prior information, as well as several other knowledge-enhanced sensing matrix designs based on more heuristic notions.
Information-Theoretic Compressive Measurement Design for Micro-Doppler Signatures
In this work, we utilise gradient-ascent multi-objective optimisation within an information-theoretic compressive sensing framework to classify micro-Doppler (m-D) signatures in the presence of
Noise Folding in Compressed Sensing
TLDR
For the vast majority of measurement schemes employed in compressed sensing, the two models are equivalent with the important difference that the signal-to-noise ratio (SNR) is divided by a factor proportional to p/n, where p is the dimension of the signal and n is the number of observations.
Compressive phase-only filtering - pattern recognition at extreme compression rates
We introduce a compressive pattern recognition method for non-adaptive WalshHadamard or discrete noiselet-based compressive measurements and show that images measured at extremely high compression
Sparse image measurement with an optical single-pixel detector using various schemes of image sampling
TLDR
The optimal measurement and reconstruction basis is established in order to obtain the best quality of the image reconstruction with possibly few measurements and is introduced into the experimental single-pixel camera set-up for imaging in the visible wavelength range.
Target-detection strategies
TLDR
Eight inventive paradigms, each with deep philosophical underpinnings, are described in relation to their effect on target detector design, including neural networks, deformable templates, and adaptive filtering.
Gradient of Mutual Information in Linear Vector Gaussian Channels in the Presence of Input Noise
TLDR
It is demonstrated that the derived expressions can outperform approximate gradient terms when integrated within a gradient ascent multi-objective optimisation approach.
Sparsity and Structure in Hyperspectral Imaging : Sensing, Reconstruction, and Target Detection
TLDR
Hyperspectral imaging is a powerful technology for remotely inferring the material properties of the objects in a scene of interest with much more accuracy than is possible with conventional three-color images.
Semi-supervised and unsupervised kernel-based novelty detection with application to remote sensing images
TLDR
This thesis addresses the development of methods for novelty detection and one-class classification with few or no labeled information and proposes a method seeking a sparse and low-rank representation of the data mapped in a non-linear feature space.
...
1
2
...

References

SHOWING 1-10 OF 70 REFERENCES
Sparse Signal Detection from Incoherent Projections
TLDR
This paper demonstrates how CS principles can solve signal detection problems given incoherent measurements without ever reconstructing the signals involved, and proposes an incoherent detection and estimation algorithm (IDEA) based on matching pursuit.
Hyperspectral target detection from incoherent projections
TLDR
The theoretical performance bounds highlight fundamental tradeoffs among the number of measurements collected, the spectral resolution of targets, the amount of background signal present, signal-to-noise ratio, and the similarity between potential targets in a dictionary.
Compressive Sampling for Signal Classification
TLDR
This paper investigates the utility of CS projection observations for signal classification (more specifically, m-ary hypothesis testing), and theoretical error bounds are derived and verified with several simulations.
Information Theoretic Bounds for Compressed Sensing
TLDR
This paper derives information theoretic performance bounds to sensing and reconstruction of sparse phenomena from noisy projections by developing novel extensions to Fano's inequality to handle continuous domains and arbitrary distortions and shows that with constant SNR the number of measurements scales linearly with the rate-distortion function of the sparse phenomena.
The smashed filter for compressive classification and target recognition
TLDR
A framework for compressive classification that operates directly on the compressive measurements without first reconstructing the image is proposed, and the effectiveness of the smashed filter for target classification using very few measurements is demonstrated.
Anomaly Detection and Reconstruction From Random Projections
TLDR
The popular RX anomaly detector is derived for the case in which global anomalies are to be identified directly in the random-projection domain, and it is determined via both random simulation and empirical observation that strongly anomalous vectors are likely to be identifiable by the projection-domain RX detector even in low-dimensional projections.
Hyperspectral target detection from incoherent projections: Nonequiprobable targets and inhomogeneous SNR
This paper describes a computationally efficient approach for the detection of spectral targets of different strengths, contaminated by a colored Gaussian background, from relatively few incoherent
Adaptive multiple-band CFAR detection of an optical pattern with unknown spectral distribution
  • I. Reed, Xiaoli Yu
  • Computer Science, Engineering
    IEEE Trans. Acoust. Speech Signal Process.
  • 1990
TLDR
Both theoretical and computer simulation results show that the SNR improvement factor of this algorithm using multiple band scenes over the single scene of maximum SNR can be substantial and illustrates that the generalized SNR of the test using the full data array is always greater than that of using partial data array.
Noise Folding in Compressed Sensing
TLDR
For the vast majority of measurement schemes employed in compressed sensing, the two models are equivalent with the important difference that the signal-to-noise ratio (SNR) is divided by a factor proportional to p/n, where p is the dimension of the signal and n is the number of observations.
A comparative study of target detection algorithms for hyperspectral imagery
In this paper, we review and compare the state-of-the-art target detection algorithms. We introduce a new target detection workflow incorporating a Minimum Noise Fraction (MNF) transform before
...
1
2
3
4
5
...