Bogner, C; Kühnel, A; Huwe, B: Predicting with limited data - Increasing the accuracy in VIS-NIR diffuse reflectance spectroscopy by SMOTE, Proc. 6th IEEE Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS) (2014)

Diffuse reflectance spectroscopy is a powerful technique to predict soil properties. It can be used in situ to provide data inexpensively and rapidly compared to the standard laboratory measurements. Because most spectral data bases contain air-dried samples scanned in the laboratory, field spectra acquired in situ are either absent or rare in calibration data sets. However, when models are calibrated on air-dried spectra, prediction using field spectra are often inaccurate. We propose a framework to calibrate partial least squares models when field spectra are rare using synthetic minority oversampling technique (SMOTE). We calibrated a model to predict soil organic carbon content using air-dried spectra spiked with synthetic field spectra. The root mean-squared error of prediction decreased from 6.18 to 2.12 mg/g and R² increased from -0.53 to 0.82 compared to the model calibrated on air-dried spectra only.

Upcoming ...

BayCEER Colloquium:
Th. 2017-12-21
Silicon in biogeochemistry
Th. 2018-01-11
Biodiversity conservation under complex socio-ecological settings - the Colombian case
Th. 2018-01-18
Effects of climate change on plants and ecosystem functioning of managed temperate grasslands
Ökologisch-Botanischer Garten:
Su. 2018-01-07
Stärk´ anlaufen: Winterspaziergang im ÖBG
Su. 2018-01-07
Musikalischer Jahresbeginn mit den Rockin´ Dinos
Weather research site
Luftdruck (356m): 955.4 hPa
Lufttemperatur: 1.5 °C
Niederschlag: 6.8 mm/24h
Sonnenschein: <1 h/d
Wind (Höhe 17m): 5.8 km/h
Wind (Max.): 13.0 km/h
Windrichtung: S