Skip to content

Leaf Disease Detection using Image Processing and Deep Learning

Notifications You must be signed in to change notification settings

Aakash1822/Projects

Repository files navigation

Project Name

Leaf Disease Detection using Image Processing

Collection of Datasets from online resources

https://www.plant-image-analysis.org/dataset

https://imagedatabase.apsnet.org/

Description:

This project is about collecting images of various infected, good and seems to be infected plant leafs. Then apply image processing on the images and predict the infected plant leafs using Deep Learning+ImageProcessing.

Steps Involved in Image Processing:-

  1. Image Acquisition
  2. Image Enhancement
  3. Image Restoration
  4. Color Image Processing
  5. Compression
  6. Segmentation

@@Using Deep Learning for Prediction along with Image processing:-

Convolution Neural network are mostly prefer neural network for Image analysis.

For more:- https://en.wikipedia.org/wiki/Convolutional_neural_network

Installation:

  1. Tensorflow:-Install tensorflow in your local environment or anaconda environment using command-line by using command: ✔$ pip install tensorflow 💻
  2. Keras:- For detailed installation of keras with tensorflow read this:- https://keras.io/#installation
  3. Scikit Learn:- If you are using anaconda it is previously installed. https://scikit-learn.org/0.15/install.html
  4. Pickle:- If you are using Anaconda its previously installed.
  5. OpenCv:- pip install opencv-python

Better to use Google Collabs or Cloud for model training and evaluation.(Because datasets are complex and huge)

Usage:

After installing all these softwares in your local machine/PC

  1. Import the libraries and forks the source code
  2. Set the path of datasets in jupyter notebook and proceed accordingly
  3. After running you seen various analysis along with accuracy, validation and loss calculation.

Contributing:

Working on open source projects have lots of learning and fun. If anyone wants to contribute on this project or wants to add any exciting features or convert it into a android application are most welcome!!