Geodesic
Recognition of anomalies of an a priori unknown type
Čebyševskij sbornik, Tome 23 (2022) no. 5, pp. 227-240.

Voir la notice de l'article provenant de la source Math-Net.Ru

In the present article we propose a modification of the PaDiM anomaly detection method which maps images to vectors and then calculates the Mahalanobis distance between such vectors and the distribution of the vectors of the training set. Of the coordinate axes of the vectors we choose a subset of such that the distribution along them is close to normal according to the chosen statistical criterion. The uniformization procedure is then applied to those coordinates and the Mahalanobis distance is calculated. This approach is shown to increase the ROCAUC value in comparison with the PaDiM method.
Keywords: anomaly detection, normal distribution, uniformization. 
@article{CHEB_2022_23_5_a18,
     author = {A. O. Ivanov and G. V. Nosovskiy and V. A. Kibkalo and M. A. Nikulin and F. Yu. Popelensky and D. A. Fedoseev and I. V. Gribushin and V. V. Zlobin and S. S. Kuzin and I. L. Mazurenko},
     title = {Recognition of anomalies of an a priori unknown type},
     journal = {\v{C}eby\v{s}evskij sbornik},
     pages = {227--240},
     publisher = {mathdoc},
     volume = {23},
     number = {5},
     year = {2022},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/CHEB_2022_23_5_a18/}
}
TY  - JOUR
AU  - A. O. Ivanov
AU  - G. V. Nosovskiy
AU  - V. A. Kibkalo
AU  - M. A. Nikulin
AU  - F. Yu. Popelensky
AU  - D. A. Fedoseev
AU  - I. V. Gribushin
AU  - V. V. Zlobin
AU  - S. S. Kuzin
AU  - I. L. Mazurenko
TI  - Recognition of anomalies of an a priori unknown type
JO  - Čebyševskij sbornik
PY  - 2022
SP  - 227
EP  - 240
VL  - 23
IS  - 5
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/CHEB_2022_23_5_a18/
LA  - ru
ID  - CHEB_2022_23_5_a18
ER  - 
%0 Journal Article
%A A. O. Ivanov
%A G. V. Nosovskiy
%A V. A. Kibkalo
%A M. A. Nikulin
%A F. Yu. Popelensky
%A D. A. Fedoseev
%A I. V. Gribushin
%A V. V. Zlobin
%A S. S. Kuzin
%A I. L. Mazurenko
%T Recognition of anomalies of an a priori unknown type
%J Čebyševskij sbornik
%D 2022
%P 227-240
%V 23
%N 5
%I mathdoc
%U http://geodesic.mathdoc.fr/item/CHEB_2022_23_5_a18/
%G ru
%F CHEB_2022_23_5_a18
A. O. Ivanov; G. V. Nosovskiy; V. A. Kibkalo; M. A. Nikulin; F. Yu. Popelensky; D. A. Fedoseev; I. V. Gribushin; V. V. Zlobin; S. S. Kuzin; I. L. Mazurenko. Recognition of anomalies of an a priori unknown type. Čebyševskij sbornik, Tome 23 (2022) no. 5, pp. 227-240. http://geodesic.mathdoc.fr/item/CHEB_2022_23_5_a18/

[1] Falkoner K., Fractal geometry: mathematical foundations and applications, John Wiley Sons, Chichester etc., 1990, xxii + 288 pp.

[2] Defard T., Setkov A., Loesch A., Audigier R., PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localization, 2020, arXiv: 2011.08785

[3] Realization of PaDiM approach, https://github.com/xiahaifeng1995/PaDiM-Anomaly-Detection-Localization-master

[4] Bergmann P., Batzner K., Fauser M., Sattlegger D., Steger C., “The MVTec Anomaly Detection Dataset: A Comprehensive Real-World Dataset for Unsupervised Anomaly Detection”, International Journal of Computer Vision, 129 (2021), 1038–1059

[5] Venturini G. M., Statistical Distances and Probability Metrics for Multivariate Data, Ensembles and Probability Distributions, Ph.D. THESIS (advisor: Alberto Munoz Garcia), Dep. of Statistics, Univ. Carlos III, Leganes, Madrid, Spain, June 2015

[6] Grudic G. Z., Mulligan J., “Outdoor Path Labeling Using Polynomial Mahalanobis Distance”, Robotics: Science and Systems II (August 16-19, 2006, University of Pennsylvania, Philadelphia, Pennsylvania, USA)

[7] Shapiro S. S., Wilk M. B., “An analysis of variance test for normality (complete samples)”, Biometrica, 52 (1965), 591–611

[8] Ivanov A., Nosovskiy G., Chekunov A., Fedoseev D., Kibkalo V., Nikulin M., Popelenskiy F., Komkov S., Mazurenko I., Petiushko A., Manifold Hypothesis in Data Analysis: Double Geometrically-Probabilistic Approach to Manifold Dimension Estimation, 2021, arXiv: 2107.03903 [cs.LG]

[9] Stolz B., Tanner J., Harrington H., Nanda V., “Geometric anomaly detection in data”, Proceedings of the National Academy of Sciences, 117 (2020), 202001741 | DOI

[10] Erba V., Gherardi M., Rotondo P., “Intrinsic dimension estimation for locally undersampled data”, Sci. Rep., 9 (2019) | DOI

[11] Fefferman C., Mitter S., Narayanan H., “Testing the manifold hypothesis”, J. Amer. Math. Soc., 29 (2016) | DOI

[12] Bernstein A., Burnaev E., Erofeev P., “Manifold Reconstruction in Dimension Reduction Problem”, International conference “Intelligent Information Processing” IIP-9, 2012

[13] Grassberger P., Procaccia I., “Measuring the strangeness of strange attractors”, Physica D: Nonlinear Phenomena, 9 (1983), 189–208

[14] Eckmann J.-P., Ruelle D., “Fundamental limitations for estimating dimensions and Lyapunov exponents in dynamical systems”, Physica D: Nonlinear Phenomena, 56 (1992), 185–187

[15] Mordohai P., Medioni G., “Dimensionality Estimation, Manifold Learning and Function Approximation using Tensor Voting”, Journal of Machine Learning Research, 11 (2010), 411–450

[16] He J., Jiang L., Ding L., Ii Z., “Intrinsic Dimensionality Estimation based on Manifold Assumption”, Journal of Visual Communication and Image Representation, 25:5 (2014), 740–747

[17] Granata D., Carnevale V., “Accurate Estimation of the Intrinsic Dimension Using Graph Distances: Unraveling the Geometric Complexity of Datasets”, Sci. Rep., 6 (2016) | DOI

[18] Levina E., Bickel P. J., “Maximum likelihood estimation of intrinsic dimension”, Advances in neural information processing systems, 2005, 777–784