Skip to content

DivaniManarandi/Plant_Disease_Detection_Using_Transfer_Learning_with_ResNet50

 
 

Repository files navigation

Plant Disease Detection Using Transfer Learning with ResNet50

This project aims to automate the detection of plant diseases using deep learning techniques. By leveraging transfer learning with the ResNet50 model pretrained on ImageNet, I have developed a robust system capable of identifying diseases in plant leaves from images. The model is fine-tuned and optimized to classify leaves as healthy or diseased with high accuracy.

Screenshot 2024-07-08 080300

Dataset :

https://drive.google.com/file/d/1kA_JWhHQhyzzuzlpzppK2nNTtbiR2N77/view

Key Features:

-Transfer Learning: Utilizes ResNet50 pretrained on ImageNet for feature extraction and fine-tuning.

-Data Augmentation: Enhances model robustness with augmented training data.

-Class Imbalance Handling: Addresses data imbalance using class weights for improved performance.

-Evaluation and Visualization: Provides metrics and visualizations for training/validation accuracy and loss.

Usage:

-Data Preparation: Organize your dataset into directories (e.g., train and validation) with subdirectories for each class (healthy and diseased).

-Model Training: Fine-tune the ResNet50 model using provided scripts or notebooks.

-Evaluation: Evaluate model performance using validation data and visualize training metrics.

-Prediction: Use the trained model for real-time disease detection on new images.

Read more

About

This project uses transfer learning with ResNet50 to automate disease identification from images, refining the model for effective classification of healthy and diseased plant leaves.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 100.0%