Fourier Analysis Approach to Identify Water Bodies Through Hyperspectral Imagery
Unlike conventional images, which have three channels of information, hyperspectral images are composed of many spectral channels that provide detailed information about the materials present in them. Thus, considering their great potential to monitor changes in the environment and the importance of...
| Main Authors: | , , |
|---|---|
| Format: | Online |
| Language: | eng |
| Published: |
Universidad Pedagógica y Tecnológica de Colombia
2023
|
| Subjects: | |
| Online Access: | https://revistas.uptc.edu.co/index.php/ingenieria/article/view/17232 |
| _version_ | 1801706105339904000 |
|---|---|
| author | Chanchí-Golondrino, Gabriel-Elías Ospina-Alarcón, Manuel-Alejandro Saba, Manuel |
| author_facet | Chanchí-Golondrino, Gabriel-Elías Ospina-Alarcón, Manuel-Alejandro Saba, Manuel |
| author_sort | Chanchí-Golondrino, Gabriel-Elías |
| collection | OJS |
| description | Unlike conventional images, which have three channels of information, hyperspectral images are composed of many spectral channels that provide detailed information about the materials present in them. Thus, considering their great potential to monitor changes in the environment and the importance of freshwater bodies for life and nature, it is relevant to propose and evaluate the effectiveness of different computational methods focused on detecting bodies of water in hyperspectral images; therefore, this research proposes and evaluates a computational method based on Fourier phase similarity. To do so, four methodological phases were defined, namely: exploration and selection of open-source technologies for hyperspectral image analysis, determination of the characteristic pixel of water bodies, calculation of Fourier phase similarity between the representative pixel of water bodies and the 200 sample pixels chosen from water bodies and other materials, and verification of the method on a test hyperspectral image. Spectral, NumPy, and Pandas libraries of Python were used to implement the proposed method, which resulted, for the first 170 bands, on an average phase similarity of 99.46% with respect to water body pixels and a minimum phase similarity with water body pixels of 93.01%. The results show that the proposed method is effective to detect water body pixels and can be used or extrapolated as an alternative to detection methods based on correlation metrics and machine learning. |
| format | Online |
| id | oai:oai.revistas.uptc.edu.co:article-17232 |
| institution | Revista Facultad de Ingeniería |
| language | eng |
| publishDate | 2023 |
| publisher | Universidad Pedagógica y Tecnológica de Colombia |
| record_format | ojs |
| spelling | oai:oai.revistas.uptc.edu.co:article-172322024-03-27T02:22:18Z Fourier Analysis Approach to Identify Water Bodies Through Hyperspectral Imagery Enfoque de análisis de Fourier para la identificación de cuerpos de agua a través de imágenes hiperespectrales Chanchí-Golondrino, Gabriel-Elías Ospina-Alarcón, Manuel-Alejandro Saba, Manuel Computer vision Fourier analysis hyperspectral imaging water bodies detection remote sensing machine learning análisis de Fourier aprendizaje automático detección de cuerpos de agua imágenes hiperespectrales sensado remoto visión por computador Unlike conventional images, which have three channels of information, hyperspectral images are composed of many spectral channels that provide detailed information about the materials present in them. Thus, considering their great potential to monitor changes in the environment and the importance of freshwater bodies for life and nature, it is relevant to propose and evaluate the effectiveness of different computational methods focused on detecting bodies of water in hyperspectral images; therefore, this research proposes and evaluates a computational method based on Fourier phase similarity. To do so, four methodological phases were defined, namely: exploration and selection of open-source technologies for hyperspectral image analysis, determination of the characteristic pixel of water bodies, calculation of Fourier phase similarity between the representative pixel of water bodies and the 200 sample pixels chosen from water bodies and other materials, and verification of the method on a test hyperspectral image. Spectral, NumPy, and Pandas libraries of Python were used to implement the proposed method, which resulted, for the first 170 bands, on an average phase similarity of 99.46% with respect to water body pixels and a minimum phase similarity with water body pixels of 93.01%. The results show that the proposed method is effective to detect water body pixels and can be used or extrapolated as an alternative to detection methods based on correlation metrics and machine learning. A diferencia de las imágenes convencionales, las cuales cuentan con tres canales de información, las imágenes hiperespectrales están conformadas por una gran cantidad de canales espectrales que permiten brindar información detallada sobre diferentes materiales presentes en ellas. De este modo, teniendo en cuenta el gran potencial que tienen estas imágenes en la monitorización de cambios en el ambiente y considerando la importancia de los cuerpos de agua dulce para la vida y la naturaleza, es relevante proponer y evaluar la efectividad de diferentes métodos computacionales enfocados en la detección de cuerpos de agua en imágenes hiperespectrales. Por ende, el objetivo de esta investigación es proponer y evaluar un método computacional basado en la similitud de fase de Fourier para la detección de cuerpos de agua en éstas. Para esto, fueron definidas cuatro fases metodológicas: exploración y selección de tecnologías libres para el análisis de imágenes hiperespectrales, determinación del pixel característico de los cuerpos de agua, cálculo de la similitud de fase de Fourier entre el pixel representativo de los cuerpos de agua y los 200 pixeles de muestra escogidos de cuerpos de agua y otros materiales, y verificación del método en una imagen hiperespectral de prueba. El método propuesto fue implementado mediante el uso de las librerías Spectral, NumPy y Pandas de Python, obteniendo como resultado para las primeras 170 bandas una similitud de fase promedio de 99,46% con respecto a pixeles de cuerpos de agua y una similitud de fase mínima con pixeles de cuerpos de agua de 93,01%. Los resultados permiten concluir que el método propuesto es efectivo para detectar pixeles de cuerpos de agua y puede ser usado o extrapolado como alternativa a los métodos de detección basados en métricas de correlación y machine learning. Universidad Pedagógica y Tecnológica de Colombia 2023-02-28 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf https://revistas.uptc.edu.co/index.php/ingenieria/article/view/17232 10.19053/uptc.01211129.v33.n67.2024.17232 Revista Facultad de Ingeniería; Vol. 33 No. 67 (2024): January-March 2024; e17232 Revista Facultad de Ingeniería; Vol. 33 Núm. 67 (2024): Enero-Marzo 2024; e17232 2357-5328 0121-1129 eng https://revistas.uptc.edu.co/index.php/ingenieria/article/view/17232/13923 Copyright (c) 2023 Gabriel-Elías Chanchí-Golondrino, Manuel-Alejandro Ospina-Alarcón, Manuel Saba http://creativecommons.org/licenses/by/4.0 |
| spellingShingle | Computer vision Fourier analysis hyperspectral imaging water bodies detection remote sensing machine learning análisis de Fourier aprendizaje automático detección de cuerpos de agua imágenes hiperespectrales sensado remoto visión por computador Chanchí-Golondrino, Gabriel-Elías Ospina-Alarcón, Manuel-Alejandro Saba, Manuel Fourier Analysis Approach to Identify Water Bodies Through Hyperspectral Imagery |
| title | Fourier Analysis Approach to Identify Water Bodies Through Hyperspectral Imagery |
| title_alt | Enfoque de análisis de Fourier para la identificación de cuerpos de agua a través de imágenes hiperespectrales |
| title_full | Fourier Analysis Approach to Identify Water Bodies Through Hyperspectral Imagery |
| title_fullStr | Fourier Analysis Approach to Identify Water Bodies Through Hyperspectral Imagery |
| title_full_unstemmed | Fourier Analysis Approach to Identify Water Bodies Through Hyperspectral Imagery |
| title_short | Fourier Analysis Approach to Identify Water Bodies Through Hyperspectral Imagery |
| title_sort | fourier analysis approach to identify water bodies through hyperspectral imagery |
| topic | Computer vision Fourier analysis hyperspectral imaging water bodies detection remote sensing machine learning análisis de Fourier aprendizaje automático detección de cuerpos de agua imágenes hiperespectrales sensado remoto visión por computador |
| topic_facet | Computer vision Fourier analysis hyperspectral imaging water bodies detection remote sensing machine learning análisis de Fourier aprendizaje automático detección de cuerpos de agua imágenes hiperespectrales sensado remoto visión por computador |
| url | https://revistas.uptc.edu.co/index.php/ingenieria/article/view/17232 |
| work_keys_str_mv | AT chanchigolondrinogabrielelias fourieranalysisapproachtoidentifywaterbodiesthroughhyperspectralimagery AT ospinaalarconmanuelalejandro fourieranalysisapproachtoidentifywaterbodiesthroughhyperspectralimagery AT sabamanuel fourieranalysisapproachtoidentifywaterbodiesthroughhyperspectralimagery AT chanchigolondrinogabrielelias enfoquedeanalisisdefourierparalaidentificaciondecuerposdeaguaatravesdeimageneshiperespectrales AT ospinaalarconmanuelalejandro enfoquedeanalisisdefourierparalaidentificaciondecuerposdeaguaatravesdeimageneshiperespectrales AT sabamanuel enfoquedeanalisisdefourierparalaidentificaciondecuerposdeaguaatravesdeimageneshiperespectrales |
