The Effects of GLCM parameters on LAI estimation using texture values from Quickbird Satellite Imagery
Abstract When the leaf area index (LAI) of a forest reaches 3, the problem of spectrum saturation becomes the main limitation to improving the accuracy of the LAI estimate. A sensitivity analysis of the Grey Level Co-occurrence Matrix (GLCM) parameters which can be applied to satellite image process...
Guardado en:
Autores principales: | , , , , , , |
---|---|
Formato: | article |
Lenguaje: | EN |
Publicado: |
Nature Portfolio
2017
|
Materias: | |
Acceso en línea: | https://doaj.org/article/ce697c94121a43bf99974af5f0445b3d |
Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
id |
oai:doaj.org-article:ce697c94121a43bf99974af5f0445b3d |
---|---|
record_format |
dspace |
spelling |
oai:doaj.org-article:ce697c94121a43bf99974af5f0445b3d2021-12-02T12:31:50ZThe Effects of GLCM parameters on LAI estimation using texture values from Quickbird Satellite Imagery10.1038/s41598-017-07951-w2045-2322https://doaj.org/article/ce697c94121a43bf99974af5f0445b3d2017-08-01T00:00:00Zhttps://doi.org/10.1038/s41598-017-07951-whttps://doaj.org/toc/2045-2322Abstract When the leaf area index (LAI) of a forest reaches 3, the problem of spectrum saturation becomes the main limitation to improving the accuracy of the LAI estimate. A sensitivity analysis of the Grey Level Co-occurrence Matrix (GLCM) parameters which can be applied to satellite image processing and analysis showed that the most important parameters included orientation, displacement and moving window size. We calculated the values of Angular Second Moment (ASM), Entropy (ENT), Correlation (COR), Contrast (CON), Dissimilarity (DIS) and Homogeneity (HOM) from Quickbird panchromatic imagery using a GLCM method. Four orientations, seven displacements and seven window sizes were considered. An orientation of 90° was best for estimating the LAI of black locust forest, regardless of moving window size, displacement and texture parameters. Displacements of 3 pixels appeared to be best. The orientation and window size had only a little influence on these settings. The highest adjusted r2 values were obtained using a 3 × 3 moving window size for ASM and ENT. The tendency of CON, COR, DIS and HOM to vary with window size was significantly affected by orientation. This study can help with parameter selection when texture features from high resolution imagery are used to estimate broad-leaved forest structure information.Jingjing ZhouRui Yan GuoMengtian SunTajiguli Tu DiShan WangJiangyuan ZhaiZhong ZhaoNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 7, Iss 1, Pp 1-12 (2017) |
institution |
DOAJ |
collection |
DOAJ |
language |
EN |
topic |
Medicine R Science Q |
spellingShingle |
Medicine R Science Q Jingjing Zhou Rui Yan Guo Mengtian Sun Tajiguli Tu Di Shan Wang Jiangyuan Zhai Zhong Zhao The Effects of GLCM parameters on LAI estimation using texture values from Quickbird Satellite Imagery |
description |
Abstract When the leaf area index (LAI) of a forest reaches 3, the problem of spectrum saturation becomes the main limitation to improving the accuracy of the LAI estimate. A sensitivity analysis of the Grey Level Co-occurrence Matrix (GLCM) parameters which can be applied to satellite image processing and analysis showed that the most important parameters included orientation, displacement and moving window size. We calculated the values of Angular Second Moment (ASM), Entropy (ENT), Correlation (COR), Contrast (CON), Dissimilarity (DIS) and Homogeneity (HOM) from Quickbird panchromatic imagery using a GLCM method. Four orientations, seven displacements and seven window sizes were considered. An orientation of 90° was best for estimating the LAI of black locust forest, regardless of moving window size, displacement and texture parameters. Displacements of 3 pixels appeared to be best. The orientation and window size had only a little influence on these settings. The highest adjusted r2 values were obtained using a 3 × 3 moving window size for ASM and ENT. The tendency of CON, COR, DIS and HOM to vary with window size was significantly affected by orientation. This study can help with parameter selection when texture features from high resolution imagery are used to estimate broad-leaved forest structure information. |
format |
article |
author |
Jingjing Zhou Rui Yan Guo Mengtian Sun Tajiguli Tu Di Shan Wang Jiangyuan Zhai Zhong Zhao |
author_facet |
Jingjing Zhou Rui Yan Guo Mengtian Sun Tajiguli Tu Di Shan Wang Jiangyuan Zhai Zhong Zhao |
author_sort |
Jingjing Zhou |
title |
The Effects of GLCM parameters on LAI estimation using texture values from Quickbird Satellite Imagery |
title_short |
The Effects of GLCM parameters on LAI estimation using texture values from Quickbird Satellite Imagery |
title_full |
The Effects of GLCM parameters on LAI estimation using texture values from Quickbird Satellite Imagery |
title_fullStr |
The Effects of GLCM parameters on LAI estimation using texture values from Quickbird Satellite Imagery |
title_full_unstemmed |
The Effects of GLCM parameters on LAI estimation using texture values from Quickbird Satellite Imagery |
title_sort |
effects of glcm parameters on lai estimation using texture values from quickbird satellite imagery |
publisher |
Nature Portfolio |
publishDate |
2017 |
url |
https://doaj.org/article/ce697c94121a43bf99974af5f0445b3d |
work_keys_str_mv |
AT jingjingzhou theeffectsofglcmparametersonlaiestimationusingtexturevaluesfromquickbirdsatelliteimagery AT ruiyanguo theeffectsofglcmparametersonlaiestimationusingtexturevaluesfromquickbirdsatelliteimagery AT mengtiansun theeffectsofglcmparametersonlaiestimationusingtexturevaluesfromquickbirdsatelliteimagery AT tajigulitudi theeffectsofglcmparametersonlaiestimationusingtexturevaluesfromquickbirdsatelliteimagery AT shanwang theeffectsofglcmparametersonlaiestimationusingtexturevaluesfromquickbirdsatelliteimagery AT jiangyuanzhai theeffectsofglcmparametersonlaiestimationusingtexturevaluesfromquickbirdsatelliteimagery AT zhongzhao theeffectsofglcmparametersonlaiestimationusingtexturevaluesfromquickbirdsatelliteimagery AT jingjingzhou effectsofglcmparametersonlaiestimationusingtexturevaluesfromquickbirdsatelliteimagery AT ruiyanguo effectsofglcmparametersonlaiestimationusingtexturevaluesfromquickbirdsatelliteimagery AT mengtiansun effectsofglcmparametersonlaiestimationusingtexturevaluesfromquickbirdsatelliteimagery AT tajigulitudi effectsofglcmparametersonlaiestimationusingtexturevaluesfromquickbirdsatelliteimagery AT shanwang effectsofglcmparametersonlaiestimationusingtexturevaluesfromquickbirdsatelliteimagery AT jiangyuanzhai effectsofglcmparametersonlaiestimationusingtexturevaluesfromquickbirdsatelliteimagery AT zhongzhao effectsofglcmparametersonlaiestimationusingtexturevaluesfromquickbirdsatelliteimagery |
_version_ |
1718394306362867712 |