The codes that accompany the methodology submitted to the IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (JSTAR). This is provided for reproducibility.
Classification of Hyperspectral images (HSI) obtained from aerial sensors to estimate the land cover usage pattern using Machine Learning (ML).
The exercise of using classifying the land per the crop type using HSI, will enable us to understand the growing pattern, land usage, soil con-ditions, water access, human activity mapping and others,are important steps in the goal to meet the challengesin anticipated food production
Hyperspectral Imagery (HSI) refers to hundreds or even thousands of such bands captured in much narrower (10-20 nm) bands, offering very rich spectral information. Each such image is, in reality, a three-dimensional (3-D) volume with each voxel (3-D pixel) having both position information in the X and Y coordinates, and the spectral information content embedded in the Z direction
Data Source: http://www.ehu.eus/ccwintco/index.php?title=Hyperspectral_Remote_Sensing_Scenes
Praveen Pankajakshan
June 16, 2020