Skip to content

Project to help classify forest around goa, specifically mangroves using aerial mapping via drones

Notifications You must be signed in to change notification settings

bharat787/ForestDeptProj

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

31 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Forest Department Project (WIP)

This repository contains the classification model for the tasks required to be accomplised as a part of the Forest Department Project done by Aerodynamics Club, BITS Goa.

Classification model (WIP)

We performed Transfer Learning on pretrained Resnet18 and VGG models from Pytorch. We use the ensemble of the two for inference.

Model Epochs trained Accuracy F1 Score
Resnet18 30 81.67 0.81
VGG 10 83.165 0.81
Ensemble - 86.334 0.86

Confusion Matrix

Confusion Matrix

Confusion matrix generated using the ensemble model.

Sample Results

The model was tested on real world data collected from our visits to Mangrove forests. More sample results can be found here Sample Sample 2

About

Project to help classify forest around goa, specifically mangroves using aerial mapping via drones

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published