CNN-Based Classification of Nutrient Stress in Rice Using Leaf Images

This project presents a deep learning-based approach for classifying the phosphorus status of rice leaves using digital images. Convolutional neural networks were used to detect subtle visual patterns associated with phosphorus deficiency, enabling non-invasive diagnosis at the leaf level. The workflow was implemented entirely in Python, combining image preprocessing (OpenCV), model training (TensorFlow), and performance evaluation (scikit-learn). Preliminary results showed that CNNs can accurately distinguish between healthy and phosphorus-deficient leaves, demonstrating the potential of AI-powered tools in agricultural diagnostics.

The poster will show how deep learning with Python can detect phosphorus deficiency in rice plants using leaf images. It presents a practical workflow combining image preprocessing and CNN-based classification to support plant health monitoring. Tools used include OpenCV, TensorFlow, and scikit-learn.

As an engineer, I'm interested in innovative applications of AI. This project was a chance to apply computer vision to agriculture by using deep learning to detect phosphorus deficiency in rice leaves.

I expect they will learn from my experience in applying CNNs to real-world problems, including data preparation, model training, and evaluation in the context of agricultural diagnostics.

Basic Python and generally familiar with machine learning ideas like classification and neural networks.