This repository is about an end to end implemetation of deep learning cotton plant disease classification web application using flask.
The dataset is downloaded from Kaggle.
- total size - 152 MB.
- train - 1951 images
- validation - 253 images
- test - 160 images
- diseased cotton plant (Fusarium Wilt)
- diseased cotton leaf (Leaf Curl Disease)
- fresh cotton plant (Healthy Plant)
- fresh cotton leaf (Healthy Leaf).
Image resolution - 694x694 pixels
Image format - JPG format
Pretrained DenseNet121 model on ImageNet dataset is used. With the help of transfer learning, the last 8 layers of the model are tuned to solve the problem. The model is trained for 20 epoches and the accuracy is 97% on test data.
pip install -r requirements.txt
For model implementation and training, run
densenet121cottondisease.ipynb
Download model weight - DenseNet121.h5 and store inside /model
folder.
For running Flask app,run:
python app.py
To build docker image, run:
docker build -t cotton .
To run docker image, run:
docker run --name cotton-app cotton
To stop docker:
docker stop cotton-app
- Sakshi (@GitHub Sakshi053)
- Prachi Gupta (@GitHub Prachigupta0305)