Rice Disease Classification Using a Stacked Ensemble of Deep Convolutional Neural Networks DOI Open Access
Zhibin Wang, Yana Wei,

Cuixia Mu

et al.

Sustainability, Journal Year: 2024, Volume and Issue: 17(1), P. 124 - 124

Published: Dec. 27, 2024

Rice is a staple food for almost half of the world’s population, and stability sustainability rice production plays decisive role in security. Diseases are major cause loss crops. The timely discovery control diseases important reducing use pesticides, protecting agricultural eco-environment, improving yield quality Deep convolutional neural networks (DCNNs) have achieved great success disease image classification. However, most models complex network structures that frequently problems, such as redundant parameters, low training efficiency, high computational costs. To address this issue improve accuracy classification, lightweight deep (DCNN) ensemble method classification proposed. First, new DCNN model (called CG-EfficientNet), which based on an attention mechanism EfficientNet, was designed base learner. Second, CG-EfficientNet with different optimization algorithms parameters were trained datasets to generate seven CG-EfficientNets, resampling strategy used enhance diversity individual models. Then, sequential least squares programming algorithm calculate weight each model. Finally, logistic regression meta-classifier stacking. verify effectiveness, experiments performed five classes tissue images: bacterial blight, kernel smut, false brown spot, healthy leaves. proposed 96.10%, higher than results classic CNN VGG16, InceptionV3, ResNet101, DenseNet201 four integration methods. experimental show not only capable accurately identifying but also computationally efficient.

Language: Английский

The Role of Genetic Resistance in Rice Disease Management DOI Open Access
Andrews Danso Ofori, Tengda Zheng, John Kwame Titriku

et al.

International Journal of Molecular Sciences, Journal Year: 2025, Volume and Issue: 26(3), P. 956 - 956

Published: Jan. 23, 2025

Rice (Oryza sativa) is a crucial staple crop for global food security, particularly in Asia. However, rice production faces significant challenges from various diseases that can cause substantial yield losses. This review explores the role of genetic resistance disease management, focusing on molecular mechanisms underlying plant–pathogen interactions and strategies developing resistant varieties. The paper discusses qualitative quantitative resistance, emphasizing importance (R) genes, defense-regulator trait loci (QTLs) conferring broad-spectrum resistance. Gene-for-gene relationships rice–pathogen are examined, Xanthomonas oryzae pv. Magnaporthe oryzae. also covers recent advancements breeding techniques, including marker-assisted selection, engineering, genome editing technologies like CRISPR-Cas. These approaches offer promising avenues enhancing while maintaining potential. Understanding exploiting durable disease-resistant varieties, essential ensuring sustainable security face evolving pathogen threats changing environmental conditions.

Language: Английский

Citations

0
A new multi-objective hyperparameter optimization algorithm for COVID-19 detection from x-ray images DOI Creative Commons
Burak Gülmez

Soft Computing, Journal Year: 2024, Volume and Issue: 28(19), P. 11601 - 11617

Published: July 23, 2024

Abstract The coronavirus occurred in Wuhan (China) first and it was declared a global pandemic. To detect X-ray images can be used. Convolutional neural networks (CNNs) are used commonly to illness from images. There lots of different alternative deep CNN models or architectures. find the best architecture, hyper-parameter optimization In this study, problem is modeled as multi-objective (MOO) problem. Objective functions multi-class cross entropy, error ratio, complexity network. For solutions objective functions, made by NSGA-III, NSGA-II, R-NSGA-II, SMS-EMOA, MOEA/D, proposed Swarm Genetic Algorithms (SGA). SGA swarm-based algorithm with cross-over process. All six algorithms run give Pareto optimal solution sets. When figures obtained analyzed hypervolume values compared, outperforms MOEA/D algorithms. It concluded that better than others for COVID-19 detection Also, sensitivity analysis has been understand effect number parameters on model success.

Language: Английский

Citations

3
Algorithms for Plant Monitoring Applications: A Comprehensive Review DOI Creative Commons
Giovanni Paolo Colucci, Paola Battilani, Marco Camardo Leggieri

et al.

Algorithms, Journal Year: 2025, Volume and Issue: 18(2), P. 84 - 84

Published: Feb. 5, 2025

Many sciences exploit algorithms in a large variety of applications. In agronomy, amounts agricultural data are handled by adopting procedures for optimization, clustering, or automatic learning. this particular field, the number scientific papers has significantly increased recent years, triggered scientists using artificial intelligence, comprising deep learning and machine methods bots, to process crop, plant, leaf images. Moreover, many other examples can be found, with different applied plant diseases phenology. This paper reviews publications which have appeared past three analyzing used classifying agronomic aims crops applied. Starting from broad selection 6060 papers, we subsequently refined search, reducing 358 research articles 30 comprehensive reviews. By summarizing advantages applying analyses, propose guide farming practitioners, agronomists, researchers, policymakers regarding best practices, challenges, visions counteract effects climate change, promoting transition towards more sustainable, productive, cost-effective encouraging introduction smart technologies.

Language: Английский

Citations

0
Predictive modelling employing machine learning, convolutional neural networks (CNNs), and smartphone RGB images for non-destructive biomass estimation of pearl millet (Pennisetum glaucum) DOI Creative Commons
Faten Dhawi, Abdul Ghafoor,

Noura Almousa

et al.

Frontiers in Plant Science, Journal Year: 2025, Volume and Issue: 16

Published: May 6, 2025

Digital tools and non-destructive monitoring techniques are crucial for real-time evaluations of crop output health in sustainable agriculture, particularly precise above-ground biomass (AGB) computation pearl millet ( Pennisetum glaucum ). This study employed a transfer learning approach using pre-trained convolutional neural networks (CNNs) alongside shallow machine algorithms (Support Vector Regression, XGBoost, Random Forest Regression) to estimate AGB. Smartphone-based RGB imaging was used data collection, Shapley additive explanations (SHAP) methodology evaluated predictor importance. The SHAP analysis identified Normalized Green-Red Difference Index (NGRDI) plant height as the most influential features AGB estimation. XGBoost achieved highest accuracy (R 2 = 0.98, RMSE 0.26) with comprehensive feature set, while CNN-based models also showed strong predictive ability. Regression performed best two important features, whereas Support least effective. These findings demonstrate effectiveness CNNs non-invasive estimation cost-effective imagery, supporting automated prediction growth monitoring. can aid small-scale carbon inventories smallholder agricultural systems, contributing climate-resilient strategies.

Language: Английский

Citations

0
A Comprehensive Review of Convolutional Neural Networks based Disease Detection Strategies in Potato Agriculture DOI Creative Commons
Burak Gülmez

Potato Research, Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 26, 2024

Abstract This review paper investigates the utilization of Convolutional Neural Networks (CNNs) for disease detection in potato agriculture, highlighting their pivotal role efficiently analyzing large-scale agricultural datasets. The datasets used, preprocessing methodologies applied, specific data collection zones, and efficacy prominent algorithms like ResNet, VGG, MobileNet variants classification are scrutinized. Additionally, various hyperparameter optimization techniques such as grid search, random genetic algorithms, Bayesian examined, impact on model performance is assessed. Challenges including dataset scarcity, variability symptoms, generalization models across diverse environmental conditions addressed discussion section. Opportunities advancing CNN-based detection, integration multi-spectral imaging remote sensing data, implementation federated learning collaborative training, explored. Future directions propose research into robust transfer deployment CNNs real-time monitoring systems proactive management agriculture. Current knowledge consolidated, gaps identified, avenues future strategies to sustain farming effectively proposed by this review. study paves way advancements AI-driven potentially revolutionizing practices enhancing food security. Also, it aims guide development efforts leading improved crop practices, increased yields, enhanced

Language: Английский

Citations

2
Advancements in maize disease detection: A comprehensive review of convolutional neural networks DOI Creative Commons
Burak Gülmez

Computers in Biology and Medicine, Journal Year: 2024, Volume and Issue: 183, P. 109222 - 109222

Published: Oct. 10, 2024

Language: Английский

Citations

1
Rice Disease Classification Using a Stacked Ensemble of Deep Convolutional Neural Networks DOI Open Access
Zhibin Wang, Yana Wei,

Cuixia Mu

et al.

Sustainability, Journal Year: 2024, Volume and Issue: 17(1), P. 124 - 124

Published: Dec. 27, 2024

Rice is a staple food for almost half of the world’s population, and stability sustainability rice production plays decisive role in security. Diseases are major cause loss crops. The timely discovery control diseases important reducing use pesticides, protecting agricultural eco-environment, improving yield quality Deep convolutional neural networks (DCNNs) have achieved great success disease image classification. However, most models complex network structures that frequently problems, such as redundant parameters, low training efficiency, high computational costs. To address this issue improve accuracy classification, lightweight deep (DCNN) ensemble method classification proposed. First, new DCNN model (called CG-EfficientNet), which based on an attention mechanism EfficientNet, was designed base learner. Second, CG-EfficientNet with different optimization algorithms parameters were trained datasets to generate seven CG-EfficientNets, resampling strategy used enhance diversity individual models. Then, sequential least squares programming algorithm calculate weight each model. Finally, logistic regression meta-classifier stacking. verify effectiveness, experiments performed five classes tissue images: bacterial blight, kernel smut, false brown spot, healthy leaves. proposed 96.10%, higher than results classic CNN VGG16, InceptionV3, ResNet101, DenseNet201 four integration methods. experimental show not only capable accurately identifying but also computationally efficient.

Language: Английский

Citations

1