Nutrient Deficiency Detection System for Black Pepper Leaf using Machine Learning

This project aims to detect plant nutrient deficiencies in black pepper plants using machine learning algorithms. The system analyzes images of black pepper leaves to identify healthy and unhealthy plants. It uses a wide data set to train the model, labeling plants with specific deficiency symptoms. The system has a user-friendly interface and uses the Xception model of Convolutional Neural Network (CNN) as the neural network architecture when implementing machine learning models. Also, the flask is used to implement the backend of the system. TensorFlow in Python is used to train the model. Real-time recommendations and predictions are provided, allowing for adjustments in crop management decisions. This innovative approach helps ensure crop health and productivity. The machine learning model consists of several steps pre-processing including resizing, image rotation and three blur image filters. The accuracy of the Nutrient Deficiency Detection model was $\mathbf{8 9. 4 \%}$.