Maize Disease classification using Deep Learning Techniques: A Review

Abstract

Recent years have seen a significant increase in interest from both academic and commercial premises due to the benefits of autonomous learning and features extraction. Natural language processing, voice processing, picture and video processing all make extensive use of it. In addition, it has developed into a hub for research in the field of agricultural plant protection, including the identification of plant diseases and the evaluation of pest ranges. To increase agricultural productivity in a sustainable way, it’s critical to identify crop leaf diseases quickly and precisely. In this paper, we present a comprehensive assessment of recent work on crop leaf disease prediction using machine learning, image processing and deep learning techniques. Deep learning (DL) techniques, particularly those built on convolutional neural networks (CNN), are now widely used to classify plant diseases. The research articles that presented the various techniques are surveyed in this article, which assesses them in terms of the dataset, the quantity of images, the quantity of classes, the techniques applied, the convolutional neural networks (CNN) models employed, and the final results obtained. Modified DL techniques outperform conventional ML techniques in terms of performance. In order to expand the real-time autonomous system for identifying maize leaf disease, we addressed the performance measurements that were employed as well as some of the limits and future work that needs to be focused on.