Skip to content

Saathi - Crop recommendation using ML and plant disease identification using CNN and transfer-learning approach

Notifications You must be signed in to change notification settings

lostmartian/crop-recommendation-and-plant-disease-detection

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Caution

The requirements.txt file is missing, and I do not have the project on my local machine. The project should run if the following packages are installed:

pip3 install flask tensorflow werkzeug pickle numpy

Saathi

An AI powered agriculture utility platform

Features

  • Plant disease detection(Detects upto 33 classes)
  • Crop recommendation based on soil quality and environmental factors
  • Information about different crops

Tech Stack

ML/DL: Tensorflow, Keras, Scikit

Web: Flask, Bootstrap

Installation

There are 2 parts in this project

ML/DL

In the models directory there are two folders 'recommender-models' & 'cnn'. Recommender-models has all the models related to crop recommendation system and cnn folder contains all the notebooks and models related to plant disease classification.

Web

  python3 -m venv venv
  cd Saathi/webapp
  pip3 install -r requirements.txt
  python3 setup.py

Results

Crop Recommendation

Algorithm Accuracy Precision Recall F1-Score
Logistic Regression 94.54 0.95 0.95 0.94
Decision Tree 97.72 0.98 0.98 0.98
SVM 9.09 0.59 0.09 0.11
Multilayer Perceptron 95.22 0.96 0.95 0.95
Random Forest 99.31 0.99 0.99 0.99

Plant Disease Detection

Architectures Training Accuracy Testing Accuracy Validation Accuracy
VGG16 92.18 91.33 91.78
ResNet50 96.02 95.41 95.53
EfficientNetV2 96.06 95.53 95.83

About

Saathi - Crop recommendation using ML and plant disease identification using CNN and transfer-learning approach

Topics

Resources

Stars

Watchers

Forks