A Gamma-Ray Identification Algorithm Based on Fisher Linear Discriminant Analysis

Boardman, D.; Flynn, Allen J.

doi:10.1109/tns.2012.2226472

Cited by 6 publications

(5 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are also opportunities for future improvement of our anomaly detection system. We could incorporate methods to detect specific isotopes or ignore known benign source types, such as common medical radioisotopes, building on previous work in isotope detection and identification [8,9]. Additionally, we have not built a spatial model of spectra which allows borrowing of information between spatial cells.…”

Section: Discussionmentioning

confidence: 99%

“…The anomaly detection algorithm should use only the shape of the spectrum, not the total count rate, since observed count rates will vary widely depending on the detector, and a wide area monitoring system may use different sizes of detectors mounted on different vehicles at different heights. Also, like scram, our anomaly detection algorithm does not attempt to discriminate between benign and threatening anomalies, instead searching for any spectral change; users who need to search for specific sources can check anomalies using a source identification algorithm to locate specific isotopes [8,9].…”

Section: Approachmentioning

confidence: 99%

See 1 more Smart Citation

Detecting changes in maps of gamma spectra with Kolmogorov–Smirnov tests

Reinhart

Ventura

Athey

2015

Nuclear Instruments and Methods in Physics Research Section A:

View full text Add to dashboard Cite

Various security, regulatory, and consequence management agencies are interested in continuously monitoring wide areas for unexpected changes in radioactivity. Existing detection systems are designed to search for radioactive sources but are not suited to repeat mapping and change detection. Using a set of daily spectral observations collected at the Pickle Research Campus, we improved on the prior Spectral Comparison Ratio Anomaly Mapping (scram) algorithm and developed a new method based on two-sample Kolmogorov-Smirnov tests to detect sudden spectral changes. We also designed simulations and visualizations of statistical power to compare methods and guide deployment scenarios.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Approachmentioning

confidence: 99%

Detecting changes in maps of gamma spectra with Kolmogorov–Smirnov tests

Reinhart

Ventura

Athey

2015

Nuclear Instruments and Methods in Physics Research Section A:

View full text Add to dashboard Cite

show abstract

“…The total number of entries in the LDB is 92453, which is 226 times larger in comparison to [11] and more than twice the used in [15]. The database entry (γline) with the highest energy belongs to 20 Na with 11258.9 1: Graphical description of the basic process. In general, the RV is criterion dependent which normally involves other peaks.…”

Section: Methodological Approachmentioning

confidence: 99%

“…While automated analysis for this problem is desirable, a study from 2007 concluded that a "secondary analysis of spectra by a trained spectroscopist is frequently necessary" to identify isotopes through their γ-lines [18]. Since then, new techniques have been explored to make automated isotope identification more reliable independently of the field of application, to mention a few: swarm optimization [19], Fischer linear discriminant analysis [20], Bayesian statistics approach [21], [22], neural networks (NN) [23]- [29], hybrid fuzzy-genetic algorithms [30]. Some of these methods have been used to perform automated peak identification [31].…”

mentioning

confidence: 99%

Contextual Isotope Ranking Criteria for Peak Identification in Gamma Spectroscopy Using a Large Database

Aguilar-Arevalo

Bertou

Canet

et al. 2022

IEEE Trans. Nucl. Sci.

View full text Add to dashboard Cite

Isotope identification is a recurrent problem in γ spectroscopy with high purity germanium detectors. In this work, new strategies are introduced to facilitate this type of analysis. Five criteria are used to identify the parent isotopes making a query on a large database of γ-lines from a multitude of isotopes producing an output list whose entries are sorted so that the γ-lines with the highest chance of being present in a sample are placed at the top. A metric to evaluate the performance of the different criteria is introduced and used to compare them. Two of the criteria are found to be superior than the others: one based on fuzzy logic, and another that makes use of the γ relative emission probabilities. A program called histoGe implements these criteria using a SQLite database containing the γ-lines of isotopes which was parsed from WWW Table of Radioactive Isotopes. histoGe is Free Software and is provided along with the database so they can be used to analyze spectra obtained with generic γ-ray detectors.

show abstract

“…In addition, pattern recognition applications have been developed for radioisotope identification from radiation measurements. Several researchers introduced radioisotope identification based on classical pattern recognition methods, such as data matching for silicon detectors [13], data matching [13][14][15] and statistical data analysis [16][17][18] for inorganic scintillation detectors, and data matching [19,20] for plastic scintillation detectors. Machine learning-based radioisotope identifiers have also been studied by several researchers.…”

Section: Introductionmentioning

confidence: 99%

Comparison of Machine Learning-Based Radioisotope Identifiers for Plastic Scintillation Detector

Jeon

Kim

et al. 2021

J Radiat Prot Res

View full text Add to dashboard Cite

Background: Identification of radioisotopes for plastic scintillation detectors is challenging because their spectra have poor energy resolutions and lack photo peaks. To overcome this weakness, many researchers have conducted radioisotope identification studies using machine learning algorithms; however, the effect of data normalization on radioisotope identification has not been addressed yet. Furthermore, studies on machine learning-based radioisotope identifiers for plastic scintillation detectors are limited. Materials and Methods:In this study, machine learning-based radioisotope identifiers were implemented, and their performances according to data normalization methods were compared. Eight classes of radioisotopes consisting of combinations of 22 Na, 60 Co, and 137 Cs, and the background, were defined. The training set was generated by the random sampling technique based on probabilistic density functions acquired by experiments and simulations, and test set was acquired by experiments. Support vector machine (SVM), artificial neural network (ANN), and convolutional neural network (CNN) were implemented as radioisotope identifiers with six data normalization methods, and trained using the generated training set. Results and Discussion:The implemented identifiers were evaluated by test sets acquired by experiments with and without gain shifts to confirm the robustness of the identifiers against the gain shift effect. Among the three machine learning-based radioisotope identifiers, prediction accuracy followed the order SVM > ANN > CNN, while the training time followed the order SVM > ANN > CNN. Conclusion:The prediction accuracy for the combined test sets was highest with the SVM. The CNN exhibited a minimum variation in prediction accuracy for each class, even though it had the lowest prediction accuracy for the combined test sets among three identifiers. The SVM exhibited the highest prediction accuracy for the combined test sets, and its training time was the shortest among three identifiers.

show abstract

A Gamma-Ray Identification Algorithm Based on Fisher Linear Discriminant Analysis

Cited by 6 publications

References 19 publications

Detecting changes in maps of gamma spectra with Kolmogorov–Smirnov tests

Detecting changes in maps of gamma spectra with Kolmogorov–Smirnov tests

Contextual Isotope Ranking Criteria for Peak Identification in Gamma Spectroscopy Using a Large Database

Comparison of Machine Learning-Based Radioisotope Identifiers for Plastic Scintillation Detector

Contact Info

Product

Resources

About