Share Email Print
cover

Proceedings Paper

Identification and characterization of agro-ecological infrastructures by remote sensing
Author(s): D. Ducrot; S. Duthoit; A. d'Abzac; C. Marais-Sicre; V. Chéret; C. Sausse
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Agro-Ecological Infrastructures (AEIs) include many semi-natural habitats (hedgerows, grass strips, grasslands, thickets…) and play a key role in biodiversity preservation, water quality and erosion control. Indirect biodiversity indicators based on AEISs are used in many national and European public policies to analyze ecological processes. The identification of these landscape features is difficult and expensive and limits their use. Remote sensing has a great potential to solve this problem. In this study, we propose an operational tool for the identification and characterization of AEISs. The method is based on segmentation, contextual classification and fusion of temporal classifications. Experiments were carried out on various temporal and spatial resolution satellite data (20-m, 10-m, 5-m, 2.5-m, 50-cm), on three French regions southwest landscape (hilly, plain, wooded, cultivated), north (open-field) and Brittany (farmland closed by hedges).

The results give a good idea of the potential of remote sensing image processing methods to map fine agro-ecological objects. At 20-m spatial resolution, only larger hedgerows and riparian forests are apparent. Classification results show that 10-m resolution is well suited for agricultural and AEIs applications, most hedges, forest edges, thickets can be detected. Results highlight the multi-temporal data importance. The future Sentinel satellites with a very high temporal resolution and a 10-m spatial resolution should be an answer to AEIs detection. 2.50-m resolution is more precise with more details. But treatments are more complicated. At 50-cm resolution, accuracy level of details is even higher; this amplifies the difficulties previously reported. The results obtained allow calculation of statistics and metrics describing landscape structures.

Paper Details

Date Published: 14 October 2015
PDF: 15 pages
Proc. SPIE 9637, Remote Sensing for Agriculture, Ecosystems, and Hydrology XVII, 96372H (14 October 2015); doi: 10.1117/12.2195077
Show Author Affiliations
D. Ducrot, Ctr. d'Etudes Spatiales de la Biosphère (France)
S. Duthoit, Ecole d'Ingénieurs de PURPAN (France)
A. d'Abzac, Ctr. d'Etudes Spatiales de la Biosphère (France)
C. Marais-Sicre, Ctr. d'Etudes Spatiales de la Biosphère (France)
V. Chéret, Ecole d'Ingénieurs de PURPAN (France)
C. Sausse, Terres Inovia (France)


Published in SPIE Proceedings Vol. 9637:
Remote Sensing for Agriculture, Ecosystems, and Hydrology XVII
Christopher M. U. Neale; Antonino Maltese, Editor(s)

© SPIE. Terms of Use
Back to Top