Localizing hot spots in Poisson radiation data matrices: nonnegative tensor factorization and phase congruency
Abstract Detecting and delineating hot spots in data from radiation sensors is required in applications ranging from monitoring large geospatial areas to imaging small objects in close proximity. This paper describes a computational method for localizing potential hot spots in matrices of independen...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2021-09-01
|
| Series: | Journal of Big Data |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s40537-021-00510-1 |
| _version_ | 1818909997821591552 |
|---|---|
| author | Michael G. Thomason Benjamin S. Jordan |
| author_facet | Michael G. Thomason Benjamin S. Jordan |
| author_sort | Michael G. Thomason |
| collection | DOAJ |
| description | Abstract Detecting and delineating hot spots in data from radiation sensors is required in applications ranging from monitoring large geospatial areas to imaging small objects in close proximity. This paper describes a computational method for localizing potential hot spots in matrices of independent Poisson data where, in numerical terms, a hot spot is a cluster of locally higher sample mean values (higher Poisson intensity) embedded in lower sample mean values (lower background intensity). Two numerical algorithms are computed sequentially for a 3D array of 2D matrices of gross Poisson counts: (1) nonnegative tensor factorization of the 3D array to maximize a Poisson likelihood and (2) phase congruency in pertinent matrices. The indicators of potential hot spots are closed contours in phase congruency in these matrices. The method is illustrated for simulated Poisson radiation datasets, including visualization of the phase congruency contours. The method may be useful in other applications in which there are matrices of nonnegative counts, provided that a Poisson distribution fits the dataset. |
| first_indexed | 2024-12-19T22:35:48Z |
| format | Article |
| id | doaj.art-694d5a015fd948eb9e5d3b9a95797b0d |
| institution | Directory Open Access Journal |
| issn | 2196-1115 |
| language | English |
| last_indexed | 2024-12-19T22:35:48Z |
| publishDate | 2021-09-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | Journal of Big Data |
| spelling | doaj.art-694d5a015fd948eb9e5d3b9a95797b0d2022-12-21T20:03:13ZengSpringerOpenJournal of Big Data2196-11152021-09-018111610.1186/s40537-021-00510-1Localizing hot spots in Poisson radiation data matrices: nonnegative tensor factorization and phase congruencyMichael G. Thomason0Benjamin S. Jordan1EECS Dept., Tickle College of Engineering, University of TennesseeEECS Dept., Tickle College of Engineering, University of TennesseeAbstract Detecting and delineating hot spots in data from radiation sensors is required in applications ranging from monitoring large geospatial areas to imaging small objects in close proximity. This paper describes a computational method for localizing potential hot spots in matrices of independent Poisson data where, in numerical terms, a hot spot is a cluster of locally higher sample mean values (higher Poisson intensity) embedded in lower sample mean values (lower background intensity). Two numerical algorithms are computed sequentially for a 3D array of 2D matrices of gross Poisson counts: (1) nonnegative tensor factorization of the 3D array to maximize a Poisson likelihood and (2) phase congruency in pertinent matrices. The indicators of potential hot spots are closed contours in phase congruency in these matrices. The method is illustrated for simulated Poisson radiation datasets, including visualization of the phase congruency contours. The method may be useful in other applications in which there are matrices of nonnegative counts, provided that a Poisson distribution fits the dataset.https://doi.org/10.1186/s40537-021-00510-1Poisson dataHot spot localizationNonnegative tensor factorizationPhase congruency |
| spellingShingle | Michael G. Thomason Benjamin S. Jordan Localizing hot spots in Poisson radiation data matrices: nonnegative tensor factorization and phase congruency Journal of Big Data Poisson data Hot spot localization Nonnegative tensor factorization Phase congruency |
| title | Localizing hot spots in Poisson radiation data matrices: nonnegative tensor factorization and phase congruency |
| title_full | Localizing hot spots in Poisson radiation data matrices: nonnegative tensor factorization and phase congruency |
| title_fullStr | Localizing hot spots in Poisson radiation data matrices: nonnegative tensor factorization and phase congruency |
| title_full_unstemmed | Localizing hot spots in Poisson radiation data matrices: nonnegative tensor factorization and phase congruency |
| title_short | Localizing hot spots in Poisson radiation data matrices: nonnegative tensor factorization and phase congruency |
| title_sort | localizing hot spots in poisson radiation data matrices nonnegative tensor factorization and phase congruency |
| topic | Poisson data Hot spot localization Nonnegative tensor factorization Phase congruency |
| url | https://doi.org/10.1186/s40537-021-00510-1 |
| work_keys_str_mv | AT michaelgthomason localizinghotspotsinpoissonradiationdatamatricesnonnegativetensorfactorizationandphasecongruency AT benjaminsjordan localizinghotspotsinpoissonradiationdatamatricesnonnegativetensorfactorizationandphasecongruency |