https://github.com/aminkhavari78/geoai-challenge-estimating-soil-parameters-from-hyperspectral-images-

Can you predict soil parameters from hyperspectral earth observation data?
https://github.com/aminkhavari78/geoai-challenge-estimating-soil-parameters-from-hyperspectral-images-

dee keras matplotlib neural-network numpy pandas seaborn

Last synced: about 21 hours ago
JSON representation

Can you predict soil parameters from hyperspectral earth observation data?

• Host: GitHub
• URL: https://github.com/aminkhavari78/geoai-challenge-estimating-soil-parameters-from-hyperspectral-images-
• Owner: AminKhavari78
• License: mit
• Created: 2023-10-23T11:39:58.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2023-11-13T10:15:36.000Z (about 1 year ago)
• Last Synced: 2023-11-13T11:29:53.369Z (about 1 year ago)
• Topics: dee, keras, matplotlib, neural-network, numpy, pandas, seaborn
• Language: Jupyter Notebook
• Homepage:
• Size: 15.9 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# GeoAI-Challenge-Estimating-Soil-Parameters-from-Hyperspectral-Images-

Can you predict soil parameters from hyperspectral earth observation data?

The objective of this challenge is to automate the process of estimating the soil parameters, specifically, potassium (K), phosphorus pentoxide (P2O5), magnesium (Mg) and pH, through extracting them from the airborne hyperspectral images captured over agricultural areas in Poland (the exact locations are not revealed). To make the solution applicable in real-life use cases, all the parameters should be estimated as precisely as possible.

https://zindi.africa/competitions/geoai-challenge-estimating-soil-parameters-from-hyperspectral-images/data

Experiment1----> use basic dense layers and relu activation, get same output for all input --> not working

Experiment2---> use conv,pooling,dropout multiple times, imporove a liitle --> keep working 


Experiment3---> use LSTM,GRU RNN, run it for 20 epochs --> this model work better 


Experiment4---> use basic Machine Learning algorithms(SVM, RandomForest,...) and HP tuning