This project focuses on classifying potato diseases using Convolutional Neural Networks (CNN) in TensorFlow. The dataset used is from the PlantVillage Dataset on Kaggle.
- Deep Learning with TensorFlow and Keras
- Convolutional Neural Networks
- FastAPI for API development
- Postman for API testing
- Data Augmentation and Preprocessing
- Model Evaluation and Visualization
Potato diseases can significantly impact crop yield and quality. This project aims to accurately classify various potato diseases using image data to help farmers and agricultural experts take timely action.
By automating the detection of potato diseases, this project can help in:
- Reducing crop losses
- Improving agricultural productivity
- Assisting farmers in early disease diagnosis
- Data Collection: Used the PlantVillage dataset.
- Data Preprocessing: Image resizing, rescaling, and augmentation.
- Model Building: Created a CNN model with TensorFlow and Keras.
- Training: Trained the model on the dataset.
- Evaluation: Evaluated model performance on test data.
Visualizations include:
- Training and Validation Accuracy
- Training and Validation Loss
- Sample Predictions
The model successfully classifies potato diseases with 99% accuracy, demonstrating the potential of deep learning in agricultural applications.
- Further data augmentation could improve model robustness.
- Experiment with different model architectures for potentially better performance.
- Deploy the model as a web or mobile application for real-time disease detection.
Watch the final presentation video here.
- PlantVillage Dataset
- TensorFlow Documentation
- Code Basics YouTube Channel by Code Basics
[Prakash.P] is a [Data Scientist] with expertise in [Modeling]. Feel free to reach out for any questions or collaborations: [prakash2822001@gmail.com].