Mulberry Leaf Disease Detectionand Classification using Deep Learning Models

Abstract

Background: Silkworms are produced from mulberry trees since they are prone to several diseases that might significantly lower the output. Early identification and accurate disease classification that affects mulberry leaves will help to guarantee the best possible crop health and production. Since hand inspection is usually time-consuming and prone to errors, automated systems are absolutely essential.

Problem: Even if agricultural technology has advanced rather much, the identification and classification of diseases on mulberry leaves remains a difficult task. For large mulberry farms, conventional methods are especially difficult to scale-up given their limitations in terms of accuracy, speed, and scalability.

Method: This paper recognizes and classifies diseases affecting mulberry leaves using deep learning models. Image recognition tasks are common uses for convolutional neural networks (CNNs), with exceptional performance. The current work is compiling, preprocessing, and feeding a mulberry leaf image collection into CNN-based architectures for disease detection and classification. The model is trained and validated on these using a set of 3,000 images spanning several disease categories.

Results: The proposed deep learning model was able to achieve a classification accuracy of 92.5%, together with a precision of 91.2%, recall of 90.3%, and an F1 score of 90.7%. The model exceeded both traditional image processing techniques and conventional machine learning models. The deep learning method showed better accuracy and processing times when compared to other methods now in use. This highlights the agricultural applications including real-time implementation possibilities.