Study on Methods of Extracting New Construction Land Information Based on SPOT6 - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2016

Study on Methods of Extracting New Construction Land Information Based on SPOT6

Lei Guo
  • Fonction : Auteur
  • PersonId : 1012187
Dongling Zhao
  • Fonction : Auteur
  • PersonId : 1012188
Rui Zhang
  • Fonction : Auteur
  • PersonId : 1012189
Meng Du
  • Fonction : Auteur
  • PersonId : 1012190
Zhixiao Li
  • Fonction : Auteur
  • PersonId : 1012191
Xiang Wang
  • Fonction : Auteur
  • PersonId : 1012192
Yaru Wang
  • Fonction : Auteur
  • PersonId : 1012193

Résumé

SPOT6 is a new remote sensing satellite launched in 2012, with high spatial resolution and strong data acquisition ability. However, a complete data preprocessing technology for the regulation of land resources has not yet been formed. According to the characteristics of SPOT6 satellite images, four different image fusion methods – Gram-Schmidt, HPF, PanSharpand PanSharpening were selected to conduct the comparison experiment by using the software platforms of ENVI, ERDAS and PCI. We evaluate the results’ performances from 3 different aspects. First, evaluating the image quality of experiment results qualitatively, then assessed quantitatively by establishing evaluation indexes including mean, standard deviation, information entropy, average gradient and correlation coefficient. Finally, evaluating the applicative effect of fused images based on the classification accuracy. The analysis results shows that the method of PanSharp is best to extract construction land information. Based on the PanSharp fusion image, in order to obtain the texture information under different scales, the authors screened the texture features according to Shannon entropy, and then used distance-based approach J-M to calculate the separation for choosing the optimal texture window. Once got the texture information, combining it with the original image to participate in the multi-scale image classification. The research result showed that multi-window texture participation in classification can improve separation of objects. Finally we extract construction land information with the method of SVM. This study may provide the technical support for application of SPOT6 image in the land resources management.
Fichier principal
Vignette du fichier
434296_1_En_10_Chapter.pdf (633.97 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01557848 , version 1 (06-07-2017)

Licence

Paternité

Identifiants

Citer

Lei Guo, Dongling Zhao, Rui Zhang, Meng Du, Zhixiao Li, et al.. Study on Methods of Extracting New Construction Land Information Based on SPOT6. 9th International Conference on Computer and Computing Technologies in Agriculture (CCTA), Sep 2015, Beijing, China. pp.94-106, ⟨10.1007/978-3-319-48357-3_10⟩. ⟨hal-01557848⟩
90 Consultations
64 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More