No abstract

No abstract

No abstract

IIi this study, the results of a series of classifications of nine cover types using a texturaVspectral approach are presented. The textu � e extraction method is based on a cooccurrence matnx algorithm.Textura� feature . s were created . fro � a SPOT near-infrared Image usmg four texture mdlces, seven window sizes, and two quantization levels. A supervised classification based on the maximum likelihood algorithm has been conducted on the three SPOT spectral bands combined with the four texture images and the three spectral bands combined with each texture image individually. The classification accuracy is measured by the Kappa coefficient calculated from confusion matrices. A factor analysis has been conducted to evaluate the contribution of each variable to the classification accuracy.The addition of a texture image brings a significant improvement. to the classification accuracy of each cover type over the results obtained from the multispectral analysis alone.The window size accounts for 90% of this improvement, while 7% is explained by the band combination, and 3% by the quantization level. There is a window size which optimizes the discrimination of each cover type. The statistics used as texture measures are reduced to a second-level contribution.

Le milieu côtier du Viêt-nam occupe un rôle primordial pour l'écono-mie et le développement du pays. Pour faciliter la plani cation des mesures de conservation et de développement de ce secteur, il convient de réaliser un suivi continu des changements du trait de côte. Des images HRV de SPOT-1 pour 1986 et de SPOT-2 pour 1991 sont utilisées pour déterminer le bilan accumulation-érosion pour une portion de 14 km de cô te. Des photographies aériennes de 1986 et une campagne de terrain (1994) servent de réalité de terrain. Après des corrections géométriques rigoureuses et un exercice d'harmonisation des capteurs visant à stabiliser la dynamique des images HRVde façon à simuler l'enregistrement des deux images par un même capteur, nous sommes en mesure d'e V ectuer une bonne comparaison des images. L'indice de végétation SAVI (Soil Adjusted Vegetation Index) sert de base pour la réalisation de masque binaire uniquement pour la couverture végétale sur les images de télédétection. La soustraction du masque de 1991 de celui de 1986 fait ressortir la domination de l'érosion dans le secteur d'étude, soit 0.62 km 2 en cinq ans. Cependant, le caractère multidate des données de télédétection permet de prédire la tendance générale de l'évolution de l'environnement cô tier.Abstract. The coastal environment occupies a prime role in the economy and the development of Vietnam. To facilitate the planning of conservation and development measures in this sector, it is necessary to monitor on a continuous basis the changes occurring in the coastline. HRV SPOT-1 and SPOT-2 images, acquired respectively in 1986 and 1991, are used for determining the accumulation-erosion budget for a 14 km portion of coastline. Both aerial photographs acquired in 1986 and ground campaign results (1994 ) are used as ground data. After applying rigorous geometric corrections and sensor harmonization for stabilizing the dynamics of the HRV images in order to simulate the acquisition of the two images as if recorded by the same sensor, we are able to make a good image comparison. The SAVI vegetation index (Soil Adjusted Vegetation Index) is used as the basis for developing a binary mask solely for the vegetation cover on the remote sensing images. The removal of the 1991 mask from the 1986 mask enhances the predominance of erosion in the study area, i.e. 0.62 km over ve years. However, because of the multidate character of remote sensing data it is possible to predict the overall trend in the evolution of the coastal environment.