Tools for multitemporal analysis and classification of multisource satellite imagery

Abstract

As acquisition technology progresses, remote sensing data contains an ever increasing amount of information. Future projects in remote sensing will give high repeatability of acquisition like Venμs (CNES1) which may provide data every 2 days with a resolution of 5.3 meters on 12 bands (420nm–900nm) and Sentinel−2 (ESA) 13 bands, 10–60m resolution and 5 days. With such data, process automation appears crucial. For that purpose, we develop several algorithms to automate image processing (classification, segmentation, interpretation, etc.). In this paper, we present an algorithm of automatic analysis which selects the best dataset of dates maximizing classification quality indices. We create two indices to evaluate jointly accuracy and precision. We present tests performed on Formosat-2 images which are similar to Venμs and Sentinel−2 for temporal repetitiveness. These tests allow validating the presented process for temporal discrimination improvement.