Oil Slicks Detection in SLAR Images with Autoencoders
Por favor, use este identificador para citar o enlazar este ítem:
http://hdl.handle.net/10045/71831
Título: | Oil Slicks Detection in SLAR Images with Autoencoders |
---|---|
Autor/es: | Alacid Soto, Beatriz | Gallego, Antonio-Javier | Gil, Pablo | Pertusa, Antonio |
Grupo/s de investigación o GITE: | Reconocimiento de Formas e Inteligencia Artificial | Automática, Robótica y Visión Artificial |
Centro, Departamento o Servicio: | Universidad de Alicante. Departamento de Lenguajes y Sistemas Informáticos | Universidad de Alicante. Departamento de Física, Ingeniería de Sistemas y Teoría de la Señal | Universidad de Alicante. Instituto Universitario de Investigación Informática |
Palabras clave: | Slicks detection | SLAR images | Autoencoders |
Área/s de conocimiento: | Lenguajes y Sistemas Informáticos | Ingeniería de Sistemas y Automática |
Fecha de publicación: | 8-dic-2017 |
Editor: | MDPI |
Cita bibliográfica: | Alacid B, Gallego A-J, Gil P, Pertusa A. Oil Slicks Detection in SLAR Images with Autoencoders. Proceedings. 2017; 1(8):820. doi:10.3390/proceedings1080820 |
Resumen: | In this manuscript, the main aim is to detect candidate regions to be oil slicks in Side-Looking Airborne Radar (SLAR) images using Deep Learning techniques. The proposed approach is based on Autoencoders to allow us to automatically discriminate oil spills without hand-crafted features or other features extracted from traditional computer vision techniques. Each image is acquired from two SLAR sensors located on both sides of an aircraft. The method can work well although the regions have a high probability of having artefacts and noise caused by the aircraft maneuvers. Those artefacts and noise are caused by not only the looking angle of both antennas which generate a blind swath beneath the aircraft, but also by other error measurements obtained when the aircraft turns. The presented method is implemented as a denoising Autoencoder that uses the Convolutional Neural Network (CNN) as encoder and decoder functions, where the decoder function learns to detect only the oil slicks. This approach can be used for detection of oil spills in emergency missions. The behavior method has been tested using SLAR images acquired from a TERMA radar for mapping the sea surface and surveillance of Spanish coasts. The results show that our method works suitably in different weather conditions and with different natural phenomena such as shoals of fish or seaweed. |
Descripción: | Presented at the 5th International Symposium on Sensor Science (I3S 2017), Barcelona, Spain, 27–29 September 2017. |
URI: | http://hdl.handle.net/10045/71831 |
ISSN: | 2504-3900 |
DOI: | 10.3390/proceedings1080820 |
Idioma: | eng |
Tipo: | info:eu-repo/semantics/conferenceObject |
Derechos: | © 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). |
Revisión científica: | si |
Versión del editor: | http://dx.doi.org/10.3390/proceedings1080820 |
Aparece en las colecciones: | INV - GRFIA - Comunicaciones a Congresos, Conferencias, etc. INV - AUROVA - Comunicaciones a Congresos Internacionales INV - HURO - Comunicaciones a Congresos, Conferencias, etc. |
Archivos en este ítem:
Archivo | Descripción | Tamaño | Formato | |
---|---|---|---|---|
2017_Alacid_etal_Proceedings.pdf | 116,82 kB | Adobe PDF | Abrir Vista previa | |
Este ítem está licenciado bajo Licencia Creative Commons