Enhancing Early Detection of Maize Leaf Diseases: A Deep Learning Framework using SqueezeNet for Rapid Diagnosis

Maize is a staple crop in India and a vital component of the agricultural economy, supporting both food consumption and numerous derived products. However, plant disease infestations pose a significant threat to maize productivity, particularly in Indian agricultural fields. Timely and accurate disease detection is crucial for sustainable crop management and yield enhancement. The aim of this study is to develop a SqueezeNet based light weight deep learning model for the detection of maize leaf diseases. The SqueezeNet based light weight deep learning was trained on a balanced maize leaf disease dataset using the Adam optimizer with a learning rate of 0.0001. The training phase achieved an accuracy of 95.84% and a precision of 92.16%, while the testing phase yielded an accuracy of 94.95% and a precision of 88.96%. The promising performance of the proposed model highlights its potential for real-time deployment in precision agriculture systems aimed at early diagnosis and treatment of plant diseases.