Multiscale Anomaly Detection Using Diffusion Maps

@article{Mishne2013MultiscaleAD,
  title={Multiscale Anomaly Detection Using Diffusion Maps},
  author={Gal Mishne and Israel Cohen},
  journal={IEEE Journal of Selected Topics in Signal Processing},
  year={2013},
  volume={7},
  pages={111-123}
}
  • Gal Mishne, I. Cohen
  • Published 1 February 2013
  • Mathematics, Computer Science
  • IEEE Journal of Selected Topics in Signal Processing
We propose a multiscale approach to anomaly detection in images, combining spectral dimensionality reduction and a nearest-neighbor-based anomaly score. We use diffusion maps to embed the data in a low dimensional representation, which separates the anomaly from the background. The diffusion distance between points is then used to estimate the local density of each pixel in the new embedding. The diffusion map is constructed based on a subset of samples from the image and then extended to all… Expand
Multiscale anomaly detection using diffusion maps and saliency score
  • Gal Mishne, I. Cohen
  • Computer Science
  • 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
  • 2014
TLDR
A new anomaly score used in the diffusion map space is proposed, which shows increased performance and enables improved detection when tested on side-scan sonar images of sea-mines and compared with competing algorithms. Expand
Iterative diffusion-based anomaly detection
  • Gal Mishne, I. Cohen
  • Computer Science, Mathematics
  • 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
  • 2017
TLDR
This paper proposes an iterative data-driven approach to improve the sample set and diffusion maps representation by updating the sampleSet with suspicious points detected in the previous iteration, which better separate the anomaly from the normal points in each iteration. Expand
How to Reduce Anomaly Detection in Images to Anomaly Detection in Noise
TLDR
The general and impossible background modeling problem is replaced by a simple noise model, and the calculation of rigorous detection thresholds are allowed, which avoids over-detection by fixing detection thresholds taking into account the multiple tests. Expand
Reducing Anomaly Detection in Images to Detection in Noise
TLDR
The general and impossible background modeling problem is replaced by simpler noise modeling, and the calculation of rigorous thresholds based on the a contrario detection theory is allowed, which works on arbitrary images. Expand
Scale-invariant anomaly detection with multiscale group-sparse models
TLDR
An anomaly-detection algorithm that learns a dictionary that is invariant to a range of scale changes, and overcomes this limitation by use of an appropriate sparse coding stage is presented. Expand
Spectral Embedding Norm: Looking Deep into the Spectrum of the Graph Laplacian
TLDR
The spectral embedding norm which sums the squared values of the first I normalized eigenvectors, where I can be significantly larger than K, is proposed and it is proved that this quantity can be used to separate clusters from the background in unbalanced settings, including extreme cases such as outlier detection. Expand
Image Anomalies: A Review and Synthesis of Detection Methods
TLDR
It is suggested that it is possible to perform automatic anomaly detection on a single image by combining the most general structural assumptions expressing the background’s normality with the proposed generic statistical detection tool. Expand
Graph-Based Supervised Automatic Target Detection
TLDR
A detection method based on data-driven target modeling, which implicitly handles variations in the target appearance, is applied to a data set of side-scan sonar images and achieves impressive results in the detection of sea mines. Expand
Mine-Like Objects detection in Side-Scan Sonar images using a shadows-highlights geometrical features space
We suggest a method to detect mine-like objects in side scan sonar images. First, we use the well known K-means algorithm recognizes which recognizes the presence of mine-like segments, followed byExpand
Auto-adaptative Laplacian Pyramids for high-dimensional data analysis
TLDR
Numerically ALP's behavior on a synthetic problem is illustrated and it is applied to the computation of the DM projection of new patterns and to the extension to them of target function values on a radiation forecasting problem over very high dimensional patterns. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 52 REFERENCES
Anomaly subspace detection based on a multi-scale Markov random field model
  • A. Goldman, I. Cohen
  • Mathematics, Computer Science
  • 2004 23rd IEEE Convention of Electrical and Electronics Engineers in Israel
  • 2004
TLDR
A multi-scale Gaussian Markov random field model and a corresponding anomaly subspace detection algorithm are introduced, based on a multiscale wavelet representation of the image, independent components analysis (ICA), and modeling each independent component as a GMRF. Expand
Multivariate Gaussian MRF for multispectral scene segmentation and anomaly detection
  • G. Hazel
  • Computer Science
  • IEEE Trans. Geosci. Remote. Sens.
  • 2000
TLDR
It is demonstrated that even the simplest first-order isotropic texture models provide significant improvement in image segmentation and anomaly detection over pure spectral clustering for the data sets examined. Expand
Non-Gaussian background modeling for anomaly detection in hyperspectral images
TLDR
Results show that the proposed local-global algorithm performs better than several other local or global anomaly detection techniques, such as the well known RX or its Gaussian Mixture version (GMM-RX). Expand
Detection of Anomaly Trends in Dynamically Evolving Systems
  • N. Rabin, A. Averbuch
  • Computer Science
  • AAAI Fall Symposium: Manifold Learning and Its Applications
  • 2010
TLDR
A learning framework, which is based on diffusion methodology, that performs data fusion and anomaly detection in multi-dimensional time series data that separates between sensors (parameters) that cause stable and instable behavior of the system. Expand
Anomaly Detection Based on Wavelet Domain GARCH Random Field Modeling
  • A. Noiboar, I. Cohen
  • Mathematics, Computer Science
  • IEEE Transactions on Geoscience and Remote Sensing
  • 2007
TLDR
It is shown that the multidimensional GARCH model generalizes the casual Gauss Markov random field (GMRF) model, and a multiscale matched subspace detector (MSD) is developed for detecting anomalies in GARCH clutter. Expand
Diffusion maps for edge-aware image editing
Edge-aware operations, such as edge-preserving smoothing and edge-aware interpolation, require assessing the degree of similarity between pairs of pixels, typically defined as a simple monotonicExpand
Using Diffusion Geometric Coordinates for Hyperspectral Imagery Representation
TLDR
The diffusion geometric coordinates derived from diffusion maps of the hyperspectral data incorporate the intrinsic geometries well where much more details about species-level spatial distributions are revealed in the authors' experiments. Expand
Heterogeneous Datasets Representation and Learning using Diffusion Maps and Laplacian Pyramids
TLDR
A method for representing and learning heterogeneous datasets by using diffusion maps for unifying and embedding heterogeneous dataset and by replacing the geometric harmonics with the Laplacian pyramid extension is proposed. Expand
Defect detection in patterned wafers using anisotropic kernels
TLDR
A defect detection procedure, which avoids image registration and is robust to pattern variations, based on anisotropic kernel reconstruction of the source image using the reference image, which shows that features originating from defect regions are not reconstructible from thereference image, and hence can be identified. Expand
Diffusion Maps, Spectral Clustering and Eigenfunctions of Fokker-Planck Operators
TLDR
A diffusion based probabilistic interpretation of spectral clustering and dimensionality reduction algorithms that use the eigenvectors of the normalized graph Laplacian is presented. Expand
...
1
2
3
4
5
...