Classification and Detection of the Region of Interest in Arachis Hypogaea Diseases using NASNetMobile

Agriculture is a crucial sector in India, facing numerous challenges such as detecting and managing crop diseases that can significantly affect yield. This paper introduces an approach for identifying and categorizing diseases in Arachis Hypogaea plant. Manual inspection is the traditional method often used, which is time-consuming and prone to errors.Automated systems, utilising computer vision and machine learning, show promise in overcoming these drawbacks. However, accurately classifying and segmenting diseases, especially in complex agricultural settings, remains a challenge.To address this challenge, a method is proposed that combines background removal with NASNetMobile for disease classification and segmentation of affected areas.Removing the background allows for precise isolation of groundnut plant images, enhancing the quality of input data for disease classification. NASNetMobile utilises its deep learning capabilities to accurately classify diseases based on features extracted from leaf images.This approach provides several advantages over existing methods, achieving an improved 97% accuracy in disease detection and segmentation by integrating background removal and disease classification techniques.