Optimization of image parameters using a hyperspectral library application to soil identification and moisture estimation
The growing number of sensors raises questions about the image parameters required for the application, soil identification and moisture estimation. Hyperspectral images are also known to contain highly redundant information. Hence not all the spectral bands are needed for the satisfactory classification of the soil types. Hence, the work was aimed at obtaining these optimal spectral bands for identifying the soil types and to use these spectral bands to estimate the moisture content of the soils using the method proposed by Whiting et.al.
Quantification of LAI interannual anomalies by adjusting climatological patterns
International audience; Scaling variations and shifts in the timing of seasonal phenology are central features of global change research. In this study, we propose a novel climatology fitting approach to quantify inter-annual anomalies in LAI seasonality. A consistent archive of daily LAI estimates was first derived from historical AVHRR satellite data for the 1981-2000 period over a globally representative sample of sites. The climatology values were then computed by averaging multi-year LAI profiles, gap filling and smoothing to eliminate possible high temporal frequency residual artifacts. The inter-annual variations in LAI were finally quantified by scaling and shifting the seasonal cli…
Optimal band selection for future satellite sensor dedicated to soil science
Hyperspectral imaging systems could be used for identifying the different soil types from the satellites. However, detecting the reflectance of the soils in all the wavelengths involves the use of a large number of sensors with high accuracy and also creates a problem in transmitting the data to earth stations for processing. The current sensors can reach a bandwidth of 20 nm and hence, the reflectance obtained using the sensors are the integration of reflectance obtained in each of the wavelength present in the spectral band. Moreover, not all spectral bands contribute equally to classification and hence, identifying the bands necessary to have a good classification is necessary to reduce …