Crop signatures will generally vary more in time-series than spectrally. Look at phenology, e.g., when a crop is planted grows, flowers, etc., than at its spectrum. Green vegetation often looks alike spectrally regardless of species.
Crop signatures will generally vary more in time-series than spectrally. Look at phenology, e.g., when a crop is planted grows, flowers, etc., than at its spectrum. Green vegetation often looks alike spectrally regardless of species.
Most of the available libraries are for minerals and rocks. Phenology is more important for crops as Guy Serbin mentoned. However, if you are using hyperspectral imagery, spectral signatures might be useful as well.
thanks for the answers yes I will be using hyperspectral imagery. Phenology is also very important as you mentioned, looking at both the reflectance derived map as well as the phenology based method would be a plus. however it requires images of longer time series or at known temporal points for specific phenological events are required which makes it expensive.. hmm... if you know of any sites for spectral signature databases, not necessarily open or free please share them with me.
Senait, I have a paper in press that details a procedure that can identify broad phrenological stages without having a high temporal data set. The current form is not of great quality but we just got the proofs yesterday and hopefully the online version will be up in a couple of weeks.
Anthony, this is great thank you! I didn't check back for more answers as there were no additions for a while. Congratulations on your paper, I saw that it is published. I will have a read. yes timing is really everything :)