Maize Leaf Disease Detection using Manta-Ray Foraging Optimization with Deep Learning Model

Abstract

Maize (corn) is a major and high yield crop, cultivated worldwide although diseases may cause severe yield reductions. Monitoring and identifying maize diseases throughout the growth cycle are crucial tasks. Accurately detecting diseases is an issue for farmers who need expertise in plant pathology, while professional diagnosis can be time-consuming and expensive. Meanwhile, conventional Deep Learning (DL) and image recognition models are slowly entering the field of plant disease detection. This paper proposes the Intelligent Maize Leaf Disease Detection design using the Manta-Ray Foraging Optimization with a DL (IMLDD-MRFODL) model. The aim of the IMLDD-MRFODL method is to detect and categorize maize leaf diseases. The IMLDD-MRFODL method applies Median Filtering (MF) for image preprocessing, a densely connected network (DenseNet) for feature extraction, and the MRFO technique for hyperparameter tuning. The IMLDD-MRFODL technique exploits a Long Short-Term Memory (LSTM) network for maize leaf disease classification. Experimental evaluation was conducted to validate the IMLDD-MRFODL approach and the comparative analysis exhibited the superior accuracy of the proposed method.