Skip to content
forked from ESIPFed/Ag-Net

Ag-Net: customized deep neural network for crop recognition

License

Notifications You must be signed in to change notification settings

ZihengSun/Ag-Net

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Ag-Net

Introduction

Ag-Net is a bundle of customized deep neural networks for recognizing crops based on spectral characteristics. The main idea is to customize the current neural networks like U-Net to make it more suitable for extracting and recognizing the season-changing features of various crop over the growing season. The ultimate goal is to generate high-quality accurate crop maps at any stage of crops rather than waiting for next year. The training and validation ground truth data comes from USGS EarthExplorer and USDA NASS.

Training Datasets

We have created a free sample dataset for direct training and testing of neural network. Ag-Net-Dataset

Ag-Net Models

Ayush Patel, Jaypee University of Information Technology : Agriculture Net.ipynb

SANKALP MITTAL, BITS Pilani : Ag-Net.ipynb Training Accuracy: 91% Validation Accuracy: 84%

Results

AgNet_NE_v1 dataset: Agricultural Map Archive of Nebraska, 1985 ~ 2000, Download here

Contributor

We have several talented students voluntarily working on this project as their summer projects.

If you would like to contribute to this project or list existing jupyterbooks or github repos in this page, please feel free to contact Dr. Ziheng Sun ([email protected]) or Dr. Annie Burgess ([email protected]).

About

Ag-Net: customized deep neural network for crop recognition

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 100.0%