Classification of Corn Diseases from Leaf Images Using Deep Transfer Learning

Abstract

Protecting agricultural crops is essential for preserving food sources. The health of the plants plays a major role in impacting the yield of the agricultural output and results in significant economic loss. This is especially important in small-scale and hobby farming products such as fruits. Grapes are an important and widely cultivated plant especially in the Mediterranean region, with over $189 billion global market value. They are consumed as fruits as well as in other manufactured forms (e.g., drinks and sweet food products). However, much like other plants, grapes are prone to a wide range of diseases that require the application of immediate remedies. Misidentifying these diseases can result in poor disease control and great losses (i.e., 5–80% crop loss). Existing computer-based solutions may suffer from low accuracy, may require high overhead, are poorly deployable, and are prone to changes in image quality. The work in this paper aimed at utilizing ubiquitous technology to help farmers combat plant diseases. Particularly, deep learning artificial intelligence image-based applications were used to classify three common grape diseases—black measles, black rot, and isariopsis leaf spot. In addition, a fourth healthy class was included. A dataset of 3639 grape leaf images (1383 black measles, 1180 black rot, 1076 isariopsis leaf spot, and 423 healthy) was used. These images were used to customize and retrain eleven convolutional network models to classify the four classes. Thorough performance evaluation revealed that it is possible to design pilot and commercial applications that have accuracies that satisfy field requirements. The models achieved consistently high performance values (>99.1%).

Image Quality Enhancement using CLAHlet RetiGaussian Filter for Maize Leaf Images

“…Similarly, Fraiwan et al [35] designed a deep learning-based corn leaf disease detection and classi cation system. They designed a CNN to classify the corn leaf images into four categories i.e.…”

“…MobileNet based CNN was used for the classi cation and identi cation of tomato disease where a bilateral lter was used for image enhancement [35].…”

A lightweight method for maize seed defects identification based on Convolutional Block Attention Module

“…( Xie et al., 2022 ) used Inceptionv3 and DenseNet201 as feature extractors at the same time, and used the dual transfer learning framework to reach the accuracy rate, recall rate and F1 values of 95.11%, 95.33%, and 95.15% on 10 types of seabird data sets. ( Fraiwan et al., 2022 ) applied the transfer learning technology to the convolution neural network model to classify three maize diseases. The average accuracy rate reached 98.6%, which proved that transfer learning can greatly improve the accuracy of classification.…”

Identifying the Growth Status of Hydroponic Lettuce Based on YOLO-EfficientNet

“…Deep learning technology is indeed an important branch in the field of machine learning. It simulates the functioning of the human brain by constructing multi-layer neural network models, enabling automated data analysis and feature extraction such as color, texture, and shape [14,15]. In deep learning, convolutional neural networks (CNN) are widely used for image processing and visual-related tasks.…”