Published on Thu May 26 2016

Suppressing Background Radiation Using Poisson Principal Component Analysis

P. Tandon, P. Huggins, A. Dubrawski, S. Labov, K. Nelson

Nuclear threat detection systems based on gamma-raySpectrometry often strongly depends on the ability to identify the part of a signal that can be attributed to background radiation. We have successfully applied a method based on Principal Component Analysis to a compact null-space model of background spectra.

0
0
0
Abstract

Performance of nuclear threat detection systems based on gamma-ray spectrometry often strongly depends on the ability to identify the part of measured signal that can be attributed to background radiation. We have successfully applied a method based on Principal Component Analysis (PCA) to obtain a compact null-space model of background spectra using PCA projection residuals to derive a source detection score. We have shown the method's utility in a threat detection system using mobile spectrometers in urban scenes (Tandon et al 2012). While it is commonly assumed that measured photon counts follow a Poisson process, standard PCA makes a Gaussian assumption about the data distribution, which may be a poor approximation when photon counts are low. This paper studies whether and in what conditions PCA with a Poisson-based loss function (Poisson PCA) can outperform standard Gaussian PCA in modeling background radiation to enable more sensitive and specific nuclear threat detection.

Sat Jun 02 2012
Machine Learning
Poisson noise reduction with non-local PCA
Photon limitations are an important concern for many applications such as night vision, nuclear medicine, and astronomy. This paper introduces a novel denoising algorithm which combines elements of dictionary learning and patch-based representations of images.
0
0
0
Mon Dec 07 2015
Computer Vision
Hyperspectral Chemical Plume Detection Algorithms Based On Multidimensional Iterative Filtering Decomposition
Hyperspectral images can be used to identify chemical plumes. However, due to noise and sensors fault, the accurate identification of chemical pixels is not easy. In this paper we present a tool that, in a completely adaptive and data driven fashion, allows to improve the performance of any classification methods.
0
0
0
Mon May 21 2018
Computer Vision
A Nonconvex Projection Method for Robust PCA
Robust principal component analysis (RPCA) is a well-studied problem with the goal of decomposing a matrix into the sum of low-rank and sparse components. We are the first to propose a method that solves RPCA problem without considering any objective function or convex relaxation.
0
0
0
Mon Jun 29 2015
Machine Learning
Variational Inference for Background Subtraction in Infrared Imagery
We propose a Gaussian mixture model for background subtraction in infraredimagery. Following a Bayesian approach, our method automatically estimates the number of Gaussian components as well as their parameters.
0
0
0
Tue Jul 04 2017
Computer Vision
Weighted Low Rank Approximation for Background Estimation Problems
Classical principal component analysis (PCA) is not robust to the presence of sparse outliers in the data. In this paper, by sticking a simple weight to the Frobenius norm, we propose a weighted low rank (WLR) method.
0
0
0
Thu Oct 24 2019
Machine Learning
Robust Principal Component Analysis Based On Maximum Correntropy Power Iterations
Principal component analysis (PCA) is recognised as a quintessential data analysis technique. The sensitivity of PCA to non-Gaussian samples and/or outliers often makes it unreliable in practice. The proposed solution reduces PCA's sensitivity to a generalised power iteration.
0
0
0