An Ensemble of Transfer Learning based InceptionV3 and VGG16 Models for Paddy Leaf Disease Classification DOI Creative Commons

B Sowmiya,

K. Saminathan,

Chithra Devi M

et al.

ECTI Transactions on Computer and Information Technology (ECTI-CIT), Journal Year: 2024, Volume and Issue: 18(1), P. 89 - 100

Published: Feb. 10, 2024

Paddy is a crucial food crop providing essential nutrients and energy serving more than half the global population. Diagnosing preventing plant diseases at an early stage for health productivity of crops. Automated disease diagnosis eliminates need experts delivers accurate outcomes. This research will diagnose paddy leaf with Deep Learning technology. The such as bacterial blight, blast, tungro, brown spot, healthy classes are diagnosed classified in this study. dataset contains 160 images from each class 800 images. Our proposed model ensemble transfer-learned InceptionV3 VGG16 architectures, which utilizes strength individual models to improve overall performance. use deep learning architectures achieved impressive accuracy rates 97.03%, 94.97%, 98.87% training, validation testing respectively. results indicating that not overfit generalizes well unseen data. model's performance evaluated confusion matrix parameters like precision, recall, F1-score, support. We also tested against other techniques without transfer techniques. Moreover, advances reliable automated detection systems, fostering sustainable agriculture enhancing security.

Language: Английский

Potato leaf disease detection with a novel deep learning model based on depthwise separable convolution and transformer networks DOI
Hatice Çatal Reis, Veysel Turk

Engineering Applications of Artificial Intelligence, Journal Year: 2024, Volume and Issue: 133, P. 108307 - 108307

Published: March 26, 2024

Language: Английский

Citations

22
A hybrid Framework for plant leaf disease detection and classification using convolutional neural networks and vision transformer DOI Creative Commons

Sherihan Aboelenin,

Foriaa Ahmed Elbasheer,

Mohamed Meselhy Eltoukhy

et al.

Complex & Intelligent Systems, Journal Year: 2025, Volume and Issue: 11(2)

Published: Jan. 15, 2025

Recently, scientists have widely utilized Artificial Intelligence (AI) approaches in intelligent agriculture to increase the productivity of sector and overcome a wide range problems. Detection classification plant diseases is challenging problem due vast numbers plants worldwide numerous that negatively affect production different crops. Early detection accurate goal any AI-based system. This paper proposes hybrid framework improve accuracy for leaf significantly. proposed model leverages strength Convolutional Neural Networks (CNNs) Vision Transformers (ViT), where an ensemble model, which consists well-known CNN architectures VGG16, Inception-V3, DenseNet20, used extract robust global features. Then, ViT local features detect precisely. The performance evaluated using two publicly available datasets (Apple Corn). Each dataset four classes. successfully detects classifies multi-class outperforms similar recently published methods, achieved rate 99.24% 98% apple corn datasets.

Language: Английский

Citations

2
The Application of Deep Learning in the Whole Potato Production Chain: A Comprehensive Review DOI Creative Commons

Rui-Feng Wang,

Wen‐Hao Su

Agriculture, Journal Year: 2024, Volume and Issue: 14(8), P. 1225 - 1225

Published: July 25, 2024

The potato is a key crop in addressing global hunger, and deep learning at the core of smart agriculture. Applying (e.g., YOLO series, ResNet, CNN, LSTM, etc.) production can enhance both yield economic efficiency. Therefore, researching efficient models for great importance. Common application areas chain, aimed improving yield, include pest disease detection diagnosis, plant health status monitoring, prediction product quality detection, irrigation strategies, fertilization management, price forecasting. main objective this review to compile research progress various processes provide direction future research. Specifically, paper categorizes applications into four types, thereby discussing introducing advantages disadvantages aforementioned fields, it discusses directions. This provides an overview describes its current stages chain.

Language: Английский

Citations

13
Hybrid Architecture for Crop Detection and Leaf Disease Detection with Improved U-Net segmentation Model and Image Processing DOI
Pramod Chavan, Pratibha Chavan,

A. S. Chavan

et al.

Crop Protection, Journal Year: 2025, Volume and Issue: unknown, P. 107117 - 107117

Published: Jan. 1, 2025

Language: Английский

Citations

1
A customised vision transformer for accurate detection and classification of Java Plum leaf disease DOI Creative Commons

Auvick Chandra Bhowmik,

Md Taimur Ahad, Yousuf Rayhan Emon

et al.

Smart Agricultural Technology, Journal Year: 2024, Volume and Issue: 8, P. 100500 - 100500

Published: July 3, 2024

Vision Transformer (ViT) has recently attracted significant attention for its performance in image classification. However, studies have yet to explore potential detecting and classifying plant leaf disease. Most existing research on diseased detection focused non-transformer convolutional neural networks (CNN). Moreover, the that applied ViT narrowly experimented using hyperparameters such as size, patch learning rate, head, epoch, batch size. these significantly contribute model performance. Recognising gap, this study Java Plum disease optimised hyperparameters. To harness of ViT, presents an experiment detection. diseases threaten agricultural productivity by negatively impacting yield quality. Timely diagnosis are essential successful crop management. The primary dataset collected Bangladesh includes six classes, 'Bacterial Spot', 'Brown Blight', 'Powdery Mildew', 'Sooty Mold', 'healthy', 'dry'. This contributes a thorough understanding diseases. Following rigorous testing refinement, our demonstrated accuracy rate 97.51%. achievement demonstrates possibilities deep-learning tools agriculture inspires further application field. Our offers foundational ensure quality precise detection, instilling confidence global market.

Language: Английский

Citations

8
Potato Harvesting Prediction Using an Improved ResNet-59 Model DOI
Abdelaziz A. Abdelhamid, Amel Ali Alhussan,

Al-Seyday T. Qenawy

et al.

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

Published: Aug. 1, 2024

Language: Английский

Citations

8
Improved tomato leaf disease classification through adaptive ensemble models with exponential moving average fusion and enhanced weighted gradient optimization DOI Creative Commons

V. Pandiyaraju,

A. M. Senthil Kumar,

Praveen Joe I R

et al.

Frontiers in Plant Science, Journal Year: 2024, Volume and Issue: 15

Published: May 17, 2024

Tomato is one of the most popular and important food crops consumed globally. The quality quantity yield by tomato plants are affected impact made various kinds diseases. Therefore, it essential to identify these diseases early so that possible reduce occurrences effect on improve overall crop support farmers. In past, many research works have been carried out applying machine learning techniques segment classify leaf images. However, existing learning-based classifiers not able detect new types more accurately. On other hand, deep with swarm intelligence-based optimization enhance classification accuracy, leading effective accurate detection This paper proposes a method for harnessing power an ensemble model in sample dataset plants, containing images pertaining nine different introduces exponential moving average function temporal constraints enhanced weighted gradient optimizer integrated into fine-tuned Visual Geometry Group-16 (VGG-16) Neural Architecture Search Network (NASNet) mobile training methods providing improved accuracy. used consists 10,000 categorized classes validating additional 1,000 reserved testing model. results analyzed thoroughly benchmarked performance metrics, thus proving proposed approach gives better terms loss, precision, recall, receiver operating characteristic curve, F1-score values 98.7%, 4%, 97.9%, 98.6%, 99.97%, respectively.

Language: Английский

Citations

7
A novel dataset of potato leaf disease in uncontrolled environment DOI Creative Commons
Nabila Husna Shabrina, Siwi Indarti, Rina Maharani

et al.

Data in Brief, Journal Year: 2023, Volume and Issue: 52, P. 109955 - 109955

Published: Dec. 12, 2023

Potatoes are of the utmost importance for both food processing and daily consumption; however, they also prone to pests diseases, which can cause significant economic losses. To address this issue, implementation image computer vision methods in conjunction with machine learning deep techniques serve as an alternative approach quickly identifying diseases potato leaves. Several studies have demonstrated promising results. However, current research is limited by use a single dataset, PlantVillage may not accurately represent diverse conditions real-world settings. Therefore, new dataset that depicts various types crucial. We propose novel offers several advantages over previous datasets, including data obtained uncontrolled environment results range variables such background angles. The proposed comprises 3076 images categorized into seven classes, leaves attacked viruses, bacteria, fungi, pests, nematodes, phytophthora, healthy This aims provide more accurate representation leaf facilitate advancements on disease identification.

Language: Английский

Citations

12
Role of Artificial Intelligence in Agriculture: An Analysis and Advancements With Focus on Plant Diseases DOI Creative Commons
Ruchi Rani, Jayakrushna Sahoo, Sivaiah Bellamkonda

et al.

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 137999 - 138019

Published: Jan. 1, 2023

The increased demand for food is accelerating plant diseases globally. Hence, a manual process of detection almost impossible. Artificial Intelligence (AI) can offer several solutions to many problems farmers. AI facile mitigate farmer's agriculture challenges. With the unpredictable changing climate, plants are often affected by where play an important role. techniques such as Machine learning and deep Learning have been employed in literature detect, predict, design recommendation systems diseases. Significant work has done this area last two decades, which change lives coming years. This paper presents systematic multi-fold survey analysis focusing on recent developed combat article discusses various challenges faced farmers their solutions. It analyzes applications agriculture, current trends, advancements disease detection. will serve researchers valuable document further research solve issues.

Language: Английский

Citations

11
A Comprehensive Approach Toward Wheat Leaf Disease Identification Leveraging Transformer Models and Federated Learning DOI Creative Commons
Md. Fahim-Ul-Islam, Amitabha Chakrabarty,

Sarder Tanvir Ahmed

et al.

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 109128 - 109156

Published: Jan. 1, 2024

Wheat is one of the most extensively cultivated crops worldwide that contributes significantly to global food caloric and protein production grown on millions hectares yearly. However, diseases like brown rust, septoria, yellow other fungus pose notable threats wheat crops, impacting quality. Diagnosing these challenging, especially in areas with limited agricultural experts. Thus, creating computerized disease identification decision-support technologies crucial for safeguarding leaf preservation crop loss mitigation. The traditional approach integrating data gathering model training has substantial challenges terms confidentiality, availability, costs related transmission. To address challenges, federated learning (FL) an appealing effective option. Our study focuses applying FL classify using image analysis. In our study, we conduct experiments high-parameterized transfer (TL) models along proposed architecture based attention mechanism, introducing into a distributed strategy founded FL. leverages beneficial interactions two cutting-edge vision transformer including advanced depthwise incorporating self-attention referred as CoAtNets, enhanced Swin Transformer V2, resulting feature representation. Moreover, introduce weight pruning which further classified by reinforced linear mechanism (LA) lower output dimensions. pruned lightweight (32M parameters) considerably decreases inference time 624.249 ms 644.899 devices low computational power, making it highly efficient FL-based systems. system outperforms all tested models, ConvNeXtBase, ConvNeXtLarge, EfficientNetV2L, InceptionResNetV2, ResNet152, NASNetLarge, achieving accuracies up 98% 99%, precision 98%, recall F-1 scores 95% across multiple input dimensions classification.

Language: Английский

Citations

4