Real-time flood detection for video surveillance
This paper introduces the real-time flash flood detection method for stationary surveillance cameras. It can be applied for rural and urban areas and capable of working during day time. The background subtraction was used to detect all changes appear in a scene. After this step, many pixel belonging...
Autores Principales: | Cáceres Hernández, Danilo, Hyun Jo, Kang, Filonenko, Alexander, Seo, Dongwook |
---|---|
Formato: | Artículo |
Idioma: | Inglés |
Publicado: |
2018
|
Materias: | |
Acceso en línea: |
https://ieeexplore.ieee.org/abstract/document/7392736/ http://ridda2.utp.ac.pa/handle/123456789/5091 http://ridda2.utp.ac.pa/handle/123456789/5091 |
id |
RepoUTP5091 |
---|---|
recordtype |
dspace |
spelling |
RepoUTP50912021-07-06T15:34:54Z Real-time flood detection for video surveillance Cáceres Hernández, Danilo Hyun Jo, Kang Filonenko, Alexander Seo, Dongwook Image color analysis Cameras Real-time systems Graphics processing units Probability Surveillance Floods Image color analysis Cameras Real-time systems Graphics processing units Probability Surveillance Floods This paper introduces the real-time flash flood detection method for stationary surveillance cameras. It can be applied for rural and urban areas and capable of working during day time. The background subtraction was used to detect all changes appear in a scene. After this step, many pixel belonging to the same moving objects may be divided. They are united by morphological closing. Too small separate objects are then removed form the scene. Color probability was calculated for all the pixels belonging to a foreground mask and connected components with low probability value were filtered out. Finally, results were improved by edge density and boundary roughness. The most time consuming step was implemented in parallel using CUDA. Real-time performance was achieved in this way. The algorithm was tested on publicly accepted video. This paper introduces the real-time flash flood detection method for stationary surveillance cameras. It can be applied for rural and urban areas and capable of working during day time. The background subtraction was used to detect all changes appear in a scene. After this step, many pixel belonging to the same moving objects may be divided. They are united by morphological closing. Too small separate objects are then removed form the scene. Color probability was calculated for all the pixels belonging to a foreground mask and connected components with low probability value were filtered out. Finally, results were improved by edge density and boundary roughness. The most time consuming step was implemented in parallel using CUDA. Real-time performance was achieved in this way. The algorithm was tested on publicly accepted video. 2018-06-29T21:27:03Z 2018-06-29T21:27:03Z 2018-06-29T21:27:03Z 2018-06-29T21:27:03Z 11/09/2015 11/09/2015 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion https://ieeexplore.ieee.org/abstract/document/7392736/ http://ridda2.utp.ac.pa/handle/123456789/5091 http://ridda2.utp.ac.pa/handle/123456789/5091 eng info:eu-repo/semantics/embargoedAccess application/pdf text/html |
institution |
Universidad Tecnológica de Panamá |
collection |
Repositorio UTP – Ridda2 |
language |
Inglés |
topic |
Image color analysis Cameras Real-time systems Graphics processing units Probability Surveillance Floods Image color analysis Cameras Real-time systems Graphics processing units Probability Surveillance Floods |
spellingShingle |
Image color analysis Cameras Real-time systems Graphics processing units Probability Surveillance Floods Image color analysis Cameras Real-time systems Graphics processing units Probability Surveillance Floods Cáceres Hernández, Danilo Hyun Jo, Kang Filonenko, Alexander Seo, Dongwook Real-time flood detection for video surveillance |
description |
This paper introduces the real-time flash flood detection method for stationary surveillance cameras. It can be applied for rural and urban areas and capable of working during day time. The background subtraction was used to detect all changes appear in a scene. After this step, many pixel belonging to the same moving objects may be divided. They are united by morphological closing. Too small separate objects are then removed form the scene. Color probability was calculated for all the pixels belonging to a foreground mask and connected components with low probability value were filtered out. Finally, results were improved by edge density and boundary roughness. The most time consuming step was implemented in parallel using CUDA. Real-time performance was achieved in this way. The algorithm was tested on publicly accepted video. |
format |
Artículo |
author |
Cáceres Hernández, Danilo Hyun Jo, Kang Filonenko, Alexander Seo, Dongwook |
author_sort |
Cáceres Hernández, Danilo |
title |
Real-time flood detection for video surveillance |
title_short |
Real-time flood detection for video surveillance |
title_full |
Real-time flood detection for video surveillance |
title_fullStr |
Real-time flood detection for video surveillance |
title_full_unstemmed |
Real-time flood detection for video surveillance |
title_sort |
real-time flood detection for video surveillance |
publishDate |
2018 |
url |
https://ieeexplore.ieee.org/abstract/document/7392736/ http://ridda2.utp.ac.pa/handle/123456789/5091 http://ridda2.utp.ac.pa/handle/123456789/5091 |
_version_ |
1796209746568544256 |
score |
12.041272 |