International Workshop on Analysis of Multi-temporal Remote Sensing Images 

About: International Workshop on Analysis of Multi-temporal Remote Sensing Images is an academic conference. The conference publishes majorly in the area(s): Change detection & Land cover. Over the lifetime, 206 publications have been published by the conference receiving 1516 citations.

Robust Satellite Techniques (RST) for Natural and Environmental Hazards Monitoring and Mitigation: Theory and Applications

Abstract: Several algorithms and data analysis techniques have been proposed using satellite observations (within atmospheric spectral windows) for cloud and surface parameters studies and for human environment monitoring applications. Quite all these algorithms are difficult to extend to different geographical, seasonal conditions, generally offering poor performances and uncertain reliability especially when applied in environmental risk prevision, monitoring and/or mitigation. In this paper the generalized satellite data analysis approach RST (Robust Satellite Technique) is described which extend the use of RAT (Robust AVHRR Techniques) approach -previously proposed by the same author in 1998 -to whatever satellite sensors. Successful RST applications are also described with reference to results so far achieved by using optical and microwaves passive sensors for volcanic eruption monitoring and prediction, forest fire detection, floods mapping, monitoring and early warning, possible earthquake precursors monitoring, oil spill detection and pipeline networks monitoring.

Development and application of multi-temporal colorimetric transformation to monitor vegetation in the desert locust habitat

Abstract: The Desert Locust (Schistocerca gregaria) is the most feared of all the locusts worldwide. Satellite imagery can provide a continuous overview of ecological conditions (i.e., vegetation, soil moisture) suitable for the Desert Locust at the continental scale and in near real time. To monitor green vegetation, most remote sensing techniques are based on vegetation indices (e.g., NDVI). However, several limitations have been observed for this index based approaches in sparsely vegetated areas. To guarantee a more robust and reliable image-independent discrimination between vegetation and non-vegetated surface types, an innovative multi-temporal and multi-spectral image analysis method was developed based on a combination of MIR, NIR and Red reflectance measurements. The proposed approach is based on a transformation of the RGB color space into HSV that decouples chromaticity and luminance. A complete automatic processing chain combining the daily observations of MODIS and SPOT VEGETATION, was designed to provide user-friendly vegetation dynamic maps at 250 m resolution over the entire locust area every 10 days. This new product informs users about the location of green vegetation and its temporal evolution. The methodology is currently implemented at the Vlaamse instelling voor technologisch onderzoek (VITO) to provide vegetation dynamic maps every dekade to the Desert Locust Information Service at FAO.

Land-cover characterization and change detection using multitemporal MODIS NDVI data

Abstract: Land-cover (LC) composition and conversions are important factors that affect ecosystem condition and function. These data are frequently used as a primary data source to generate landscape-based metrics to assess landscape condition at multiple assessment scales. The use of satellite-based remote sensor data has been widely applied to provide a cost-effective means to develop LC coverages over large geographic regions. Past and ongoing efforts for generating LC data for the United States have been implemented using an interagency consortium to share the substantial costs associated satellite data acquisition, processing and analysis. The first moderate resolution National Land-Cover Data (NLCD) set was developed for the conterminous United States using Landsat Thematic Mapper (TM) imagery collected between1991-1992 (Vogelmann et al., 1998). Currently, the 2001 NLCD is under development for all 50 States and the Commonwealth of Puerto Rico (Homer et al., 2004). The 2001 effort, building on the lessons learned from the 1991 NLCD, promises to provide a relatively high quality baseline LC product.

Unsupervised linear unmixing for change detection in multitemporal airborne hyperspectral imagery

Abstract: The linear unmixing technique is investigated for change detection in multitemporal airborne hyperspectral imagery. Several practical implementation issues are discussed. The preliminary study using the CASI data shows its feasibility when the noise level is moderate and some prior information about endmembers is known. Keywords— linear mixture model; unsupervised linear unmixing; change detection; multitemporal airborne hyperspectral imagery.