Remote Sensing

Remote Sensing is the research area in which the Earth's surface is studied, usually using the reflectance spectrum of the Sun. Vision Lab has built expertise in the processing and analysis of multispectral and hyperspectral remote sensing images. Topics of research include the development of techniques for image denoising, restoration, fusion, segmentation, classification and spectral unmixing. Recently, research is shifted to more close-range applications, for which we have our own acquisition equipement (spectrometer 400-2500 nm; VNIR camera 400-900 nm and SWIR camera 1000-1700 nm). Applications are material characterization (corrosion, soil moisture, powder material composition).

Journal publications


B. Koirala, Rasti, B., and Scheunders, P., A Supervised Approach for Estimating Fractional Abundances of Binary Intimate Mixtures, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, pp. 1-11, 2024.PDF icon extensive_analysis_of_intimate_mixtures.pdf (4.49 MB)



B. Koirala, Zahiri, Z., and Scheunders, P., A Robust Supervised Method for Estimating Soil Moisture Content From Spectral Reflectance, IEEE Transactions on Geoscience and Remote Sensing , vol. 60, pp. 1-13, 2022.PDF icon soil_moisture_content.pdf (7.95 MB)
P. Ghosh, Roy, S. Kumar, Koirala, B., Rasti, B., and Scheunders, P., Hyperspectral Unmixing Using Transformer Network, IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1-16, 2022.PDF icon transformer_unmixing.pdf (6.32 MB)
B. Rast, Koirala, B., Scheunders, P., and Chanussot, J., MiSiCNet: Minimum Simplex Convolutional Network for Deep Hyperspectral Unmixing, IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1-15, 2022.PDF icon ieee_journal_misicnet.pdf (11.02 MB)