Deep Learning-Based Leaf Disease Detection in Crops Using Images for Agricultural Applications

Abstract

The agricultural sector plays a key role in supplying quality food and makes the greatest contribution to growing economies and populations. Plant disease may cause significant losses in food production and eradicate diversity in species. Early diagnosis of plant diseases using accurate or automatic detection techniques can enhance the quality of food production and minimize economic losses. In recent years, deep learning has brought tremendous improvements in the recognition accuracy of image classification and object detection systems. Hence, in this paper, we utilized convolutional neural network (CNN)-based pre-trained models for efficient plant disease identification. We focused on fine tuning the hyperparameters of popular pre-trained models, such as DenseNet-121, ResNet-50, VGG-16, and Inception V4. The experiments were carried out using the popular PlantVillage dataset, which has 54,305 image samples of different plant disease species in 38 classes. The performance of the model was evaluated through classification accuracy, sensitivity, specificity, and F1 score. A comparative analysis was also performed with similar state-of-the-art studies. The experiments proved that DenseNet-121 achieved 99.81% higher classification accuracy, which was superior to state-of-the-art models.

Explainable AI Based framework for Banana Disease Detection

“…Research in banana disease detection often utilizes datasets like PlantVillage but faces challenges due to data variability scarcity [19]. Recent works [20], have explored transfer learning to mitigate the limited data issue in detecting banana diseases. In [21] Ferentinos, highlighted the potential of deep learning in plant disease detection, emphasizing the need for larger, diverse datasets.…”

PlantDet: A Robust Multi-Model Ensemble Method Based on Deep Learning For Plant Disease Detection

“…A recent study shows that DenseNet-121 is preferable for identifying plant diseases and outscored cutting-edge pretrained models like ResNet-50, VGG-16, and Inception-V4 by achieving 99.81% classification accuracy using the PlantVillage dataset. The model's performance was assessed based on classification accuracy, sensitivity, specificity, and F1 score using transfer learning techniques [19]. Another recent study suggested a vigorous method to identify and categorize nine classes of tomato plant diseases and a healthy class using the same PlantVillage dataset called Faster-RCNN based on ResNet-34 and Convolutional Block Atten-tion Module (CBAM) as the foundation of the Faster-RCNN model to specifically extract the deep features from the input samples.…”

CMA-YOLO: A Wine Grape Detection Model Based on YOLOv5x Combining Mixed Attention Mechanism

“…The latter adaptively learns the importance of different spatial positions in the feature map. The channel attention M c and spatial attention M s can be represented as Equations (3)(4):…”

“…In recent years, computer vision has experienced rapid development, and detection algorithms based on deep learning have been widely applied in various aspects of social life. Convolutional neural network (CNN) is a deep learning algorithm that excels in object detection and classification tasks by extracting features from images through multiple layers of convolution, pooling, and fully connected operations [4][5][6]. Wine grape clusters exhibit significant color and shape variations and can be easily affected by cluttered backgrounds.…”