Enhanced Plant Health Recognition via Humanoid Robot Advanced Sensing and AI

Abstract

Accurate and timely detection of plant diseases is crucial for protecting crop yields and promoting sustainable agriculture. This study introduces a deep learning-based approach for plant health detection by integrating a Convolutional Neural Network (CNN) with a Humanoid robot for real-time monitoring. The approach leverages advanced tools such as TensorFlow for model development, OpenCV for image processing, and YOLOv5 for object detection. A dataset comprising 1,530 images, labeled as “Healthy,” “Powdery,” and “Rust,” was used to train, validate, and test the model. Through pre-processing techniques like rescaling, data augmentation, and feature extraction, the model achieved impressive results, with a training accuracy of 98.4%, validation accuracy of 98.2%, and testing accuracy of 99.3%. This approach marks a significant improvement in precision agriculture, offering a scalable and highly accurate solution for early plant health detection.