Rice leaf Nutrient Deficiency Classification System using CAR-Capsule network

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 169518 - 169532

Published: Jan. 1, 2024

Language: Английский

---

Pattern Classification of an Onion Crop (Allium Cepa) Field Using Convolutional Neural Network Models

Agronomy, Journal Year: 2024, Volume and Issue: 14(6), P. 1206 - 1206

Published: June 3, 2024

Agriculture is an area that currently benefits from the use of new technologies and techniques, such as artificial intelligence, to improve production in crop fields. Zacatecas one states producing most onions northeast region Mexico. Identifying determining vegetation, soil, humidity zones could help solve problems irrigation demands or excesses, identify spaces with different levels soil homogeneity, estimate yield health crop. This study examines application intelligence through deep learning, specifically convolutional neural networks, patterns can be found a field, this case, zones. To extract mentioned patterns, K-nearest neighbor algorithm was used pre-process images taken using unmanned aerial vehicles form dataset composed 3672 (1224 for each class). A total six network models were classify namely Alexnet, DenseNet, VGG16, SqueezeNet, MobileNetV2, Res-Net18. Each model evaluated following validation metrics: accuracy, F1-score, precision, recall. The results showed variation performance between 90% almost 100%. Alexnet obtained highest metrics accuracy 99.92%, while MobileNetV2 had lowest 90.85%. Other models, ResNet18, 92.02% 98.78%. Furthermore, our highlights importance adopting agriculture, particularly management onion fields Zacatecas, findings farmers agronomists make more informed efficient decisions, which lead greater sustainability local agriculture.

Language: Английский

---

Quality Grading of Dried Abalone Using an Optimized VGGNet

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(13), P. 5894 - 5894

Published: July 5, 2024

As living standards have improved, consumer demand for high-quality dried abalone has increased. Traditional grading is achieved through slice analysis (sampling analysis) combined with human experience. However, this method several issues, including non-uniform standards, low detection accuracy, inconsistency between internal and external quality, high loss rate. Therefore, we propose a deep-learning-aided approach leveraging X-ray images that can achieve efficient non-destructive quality of abalone. To the best our knowledge, first work to use image structure The was divided into three phases. First, database constructed, containing 644 samples, relationship analyzed. Second, augmented by rotation, mirroring, noise superposition. Subsequently, model selection evaluation process carried out. results showed that, in comparison models such as VGG-16, MobileNet (Version 1.0), AlexNet, Xception, VGG-19 demonstrated performance Finally, modified network based on CBAM proposed classify show effective, achieving score 95.14%, outperformed competitors, i.e., alone squeeze-and-excitation block (SE) attention mechanism.

Language: Английский

---

Enhancing deep convolutional neural network models for orange quality classification using MobileNetV2 and data augmentation techniques

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: July 19, 2024

This study introduces significant improvements in the construction of Deep Convolutional Neural Network (DCNN) models for classifying agricultural products, specifically oranges, based on their shape, size, and color. Utilizing MobileNetV2 architecture, this research leverages its efficiency lightweight nature, making it suitable mobile embedded applications. Key techniques such as Depthwise Separable Convolutions, Linear Bottlenecks, Inverted Residuals help reduce number parameters computational load while maintaining high performance feature extraction. Additionally, employs comprehensive data augmentation methods, including horizontal vertical flips, grayscale transformations, hue adjustments, brightness noise addition to enhance model's robustness generalization capabilities. The proposed model demonstrates superior performance, achieving an overall accuracy 100% with nearly perfect precision, recall, F1-score both "orange_good" "orange_bad" classes, significantly outperforming previous which typically achieved accuracies between 70–90%. confusion matrix shows that has sensitivity specificity, very few misclassifications. Finally, empresentasizes practical applicability model, particularly easy deployment resource-constrained devices effectiveness product quality control processes. These findings affirm a reliable highly efficient tool classification, surpassing capabilities traditional field.

Language: Английский

---

Convolution Self-Guided Transformer for Diagnosis and Recognition of Crop Disease in Different Environments

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 165903 - 165917

Published: Jan. 1, 2024

Language: Английский

---