I've come across a couple of publicly available hyperspectral image datasets from the website http://www.ehu.eus/ccwintco/index.php/Hyperspectral_Remote_Sensing_Scenes
Now, my concern is regarding whether or not we can use these datasets directly for building machine learning based models.
Are these datasets geometrically and atmospherically correct? Or do these datasets further require atmospheric and geometric corrections prior to building any machine learning based model?
Also, how can one be assured that a captured hyperspectral image scene is atmospherically and geometrically correct?
Nasir Hameed thank you for your response to my doubt.
The purpose is to design a machine learning based land cover classification framework out of it. How can it be assured that atmospheric correction is unnecessary for a particular HSI dataset?
And what about geometric corrections?
- Badges
- Science topic