Evaluation of Cartosat-1 Multi-Scale Digital Surface Modelling Over France

Abstract

On 5 May 2005, the Indian Space Research Organization launched Cartosat-1, the eleventh satellite of its constellation, dedicated to the stereo viewing of the Earth's surface for terrain modeling and large-scale mapping, from the Satish Dhawan Space Centre (India). In early 2006, the Indian Space Research Organization started the Cartosat-1 Scientific Assessment Programme, jointly established with the International Society for Photogrammetry and Remote Sensing. Within this framework, this study evaluated the capabilities of digital surface modeling from Cartosat-1 stereo data for the French test sites of Mausanne les Alpilles and Salon de Provence. The investigation pointed out that for hilly territories it is possible to produce high-resolution digital surface models with a root mean square error less than 7.1 m and a linear error at 90% confidence level less than 9.5 m. The accuracy of the generated digital surface models also fulfilled the requirements of the French Reference 3D®, so Cartosat-1 data may be used to produce or update such kinds of products.

Probabilistic flood inundation mapping of ungauged rivers: Linking GIS techniques and frequency analysis

“…Cartosat-1 carries two high-resolution imaging cameras: the afterward looking camera (Aft) and the foreword looking camera (For), both able to collect panchromatic images with a spatial ground resolution of 2.5 m. The 12,000 pixels, each with 7 Â 7 microns, in the image plane are covering with the focal length of 1.98 m, a swath of 30 km with the forward view and 26.6 km with the backward view (Crespi et al, 2008;Gianinetto, 2009). These characteristics have made Cartosat-1 images a useful instrument for generating ortho-images and high-resolution digital surface models that can improve flow hydraulic simulation, especially in mountainous regions.…”

HOVE-Wedge-Filtering of Geomorphologic Terrestrial Laser Scan Data

“…According to the ASPRS (American Society for Photogrammetry and Remote Sensing) Guidelines [23] and Gianinetto and Fassi [24], the RMSE is often used to assess the accuracy of elevation data and is defined as (10):…”

Hierarchical classification of complex landscape with VHR pan-sharpened satellite data and OBIA techniques

“…On the other hand, with sub-meter multispectral imagery the spatial extent, shape, structure and texture of land features appear fairly clear because the mixture of different land-cover classes in a single pixel is directly proportional to the pixel size. Furthermore, VHR optical images are also useful to retrieve different kind of landscape detail, such as topographyrelated features, which can be used as additional information during the classification process [Gianinetto, 2008[Gianinetto, , 2009. Concerning the thematic classification of VHR data, a limitation of pixel-based approaches is the absence of semantic information on shape and relation to neighbourhood [Webb, 2003;Canty, 2009;Bhaskaran et al, 2010].…”