Plant Disease Detection Research Articles

Automated Detection of Plant Diseases and Pests Using Advanced Image Processing and Neural Networks

Pests and viral plant diseases have proliferated in agriculture. Rats spread plant diseases. Hard to detect viral infections or insect infestations that impair plant growth and yield. It decreases crop size, quality, and marketability, incurring major economic losses. Strong, disease-resistant crops are needed for agriculture. Bacterial, fungal, and insect/pest diseases destroy 70% and 15% of crops. Above 85% of disease severity destroys crops. Early detection systems track pests and diseases. High-definition equipment is costly for rural farmers. Correct diagnosis permits quick infection severity and type assessment and control. In precision agriculture, neural networks and image processing eliminate human error. Sustainable agriculture and organic agricultural cultivation must start early. The proposed system classifies diseases and pests for plant safety. A healthy, fast-growing plantation is our goal. Early detection of viral diseases and insect infestations boosts crop productivity. Plant disease diagnostics focuses on leaves and buds. Stem and insect infections are covered. The suggested method classifies plant health and disease automatically. Peanut leaves develop Bacterial Blight from contagious illnesses like Cercospora Leaf Peanut. Boll Weevils, European Corn Borers, and Fall Army Worms attack cotton. The conditions are categorised. Image accuracy depends on lighting, resolution, location, and backdrop complexity. Various low-resolution pictures are enriched to better classification. Complex leaf intensity variations cause the system to misdiagnose illness or insect infestation. To avoid misclassification, the system segments the sample image using various methods to find lesions. Multilevel Segmentation (MS) improves Deep neural network feature extraction using approximate outlines. This method protects important pixels, eliminates unneeded pixels, and amplifies the lesion to control segmentation and reduce insufficient and excessive segmentation.

Comprehensive Study on Plant Disease Detection by using Hybrid Convolution Techniques

Comprehensive Study on Plant Disease Detection by using Hybrid Convolution Techniques

An Improved Weighted Evidence Combination Based on Tangent Similarity and Its Application in Decision-Making

Dempster-Shafer theory (DST) has been widely recognized across multiple disciplines for its superior handling of uncertainty compared to traditional probability theory. Nonetheless, the application of Dempster's rule in the presence of conflicting evidence can lead to outcomes that are sometimes non-intuitive. To mitigate this issue, this paper proposes a new tangent similarity measure within DST to assess the conflict between evidences. The proposed measure adheres to several key properties, enhancing its ability to accurately reflect the similarity between evidences. Additionally, an improved weighted evidence combination framework utilizing the tangent similarity measure is developed. The effectiveness of the proposed method is demonstrated through a decision-making scenario in plant disease detection.

Designing of Lightweight Deep Learning Framework for Plant Disease Detection

Designing of Lightweight Deep Learning Framework for Plant Disease Detection

The impact of fine-tuning paradigms on unknown plant diseases recognition

Plant diseases pose significant threats to agriculture, impacting both food safety and public health. Traditional plant disease detection systems are typically limited to recognizing disease categories included in the training dataset, rendering them ineffective against new disease types. Although out-of-distribution (OOD) detection methods have been proposed to address this issue, the impact of fine-tuning paradigms on these methods has been overlooked. This paper focuses on studying the impact of fine-tuning paradigms on the performance of detecting unknown plant diseases. Currently, fine-tuning on visual tasks is mainly divided into visual-based models and visual-language-based models. We first discuss the limitations of large-scale visual language models in this task: textual prompts are difficult to design. To avoid the side effects of textual prompts, we futher explore the effectiveness of purely visual pre-trained models for OOD detection in plant disease tasks. Specifically, we employed five publicly accessible datasets to establish benchmarks for open-set recognition, OOD detection, and few-shot learning in plant disease recognition. Additionally, we comprehensively compared various OOD detection methods, fine-tuning paradigms, and factors affecting OOD detection performance, such as sample quantity. The results show that visual prompt tuning outperforms fully fine-tuning and linear probe tuning in out-of-distribution detection performance, especially in the few-shot scenarios. Notably, the max-logit-based on visual prompt tuning achieves an AUROC score of 94.8%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\%$$\\end{document} in the 8-shot setting, which is nearly comparable to the method of fully fine-tuning on the full dataset (95.2%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\%$$\\end{document}), which implies that an appropriate fine-tuning paradigm can directly improve OOD detection performance. Finally, we visualized the prediction distributions of different OOD detection methods and discussed the selection of thresholds. Overall, this work lays the foundation for unknown plant disease recognition, providing strong support for the security and reliability of plant disease recognition systems. We will release our code at https://github.com/JiuqingDong/PDOOD to further advance this field.

IMAGE PROCESSING-DRIVEN DEEP LEARNING MODEL FOR PLANT DISEASE DETECTION TO ENHANCE IRRIGATION EFFICIENCY IN SMART AGRICULTURE

In the era of smart agriculture, efficient irrigation management is crucial for optimizing crop yield and resource use. Traditional methods of plant disease detection often rely on manual inspection, which is time-consuming and prone to errors. Smart agriculture leverages technology to improve agricultural practices. Accurate and timely plant disease detection is vital for effective irrigation management and overall crop health. Current methods are limited by their manual nature and inability to process large volumes of data quickly. Manual plant disease detection is labor-intensive and may not provide timely information, leading to inefficient irrigation practices. This inefficiency can result in reduced crop yield and wasted resources. LeNet integrates advanced image processing techniques with a deep learning architecture tailored for plant disease detection. The model utilizes convolutional neural networks (CNNs) to analyze plant leaf images, identifying disease patterns with high precision. LeNet incorporates preprocessing steps such as image normalization and augmentation to enhance model robustness. The network is trained on a comprehensive dataset of plant disease images, employing transfer learning to leverage pre-trained weights for improved accuracy. Evaluation of LeNet on a test dataset comprising 10,000 images demonstrated an impressive accuracy of 92.5%, with a precision of 90.3% and recall of 94.1%. The model significantly outperforms traditional methods, reducing disease detection time by 60% and enhancing irrigation efficiency by 30%. The reduction in water usage and increased crop yield were observed in practical trials.

Maize leaf disease detection using convolutional neural network: a mobile application based on pre-trained VGG16 architecture

ABSTRACT Reliance on visual inspection for Maize leaf disease identification proves unreliable, often resulting in inappropriate pesticide application and associated health hazards. Food security requires precise and automated disease detection methods to save time and prevent crop losses. Although several studies have used deep and machine learning to detect plant leaf diseases from different perspectives, most of them require numerous training parameters or have low classification accuracy. Furthermore, a model that was developed for one region of the world might not be appropriate for another due to distinctions in morphology and other aspects. In this context, an application for mobile phones was developed that recognizes and classifies maize leaf diseases using a CNN-based pretrained VGG16 architecture. The model can detect northern corn leaf blight, common rust, and gray leaf spots in maize leaves in tropical climates. A total of 3024 images were used to generate the underlying model, including publicly available and field-collected images. The established model uses fewer training parameters to attain a training accuracy of 95.16% and a testing accuracy of 93%. The model provides farmers with an early warning system for early detection of plant diseases, enabling them to take preventive measures before significant production deficits occur.

Rapid Detection of Viral, Bacterial, Fungal, and Oomycete Pathogens on Tomatoes with Microneedles, LAMP on a Microfluidic Chip, and Smartphone Device.

Rapid detection of plant diseases before they escalate can improve disease control. Our team has developed rapid nucleic acid extraction methods with microneedles and combined these with loop-mediated amplification (LAMP) assays for pathogen detection in the field. In this work, we developed LAMP assays for early blight (Alternaria linariae, A. alternata, and A. solani) and bacterial spot of tomato (Xanthomonas perforans) and validated these LAMP assays and two previously developed LAMP assays for tomato spotted wilt virus and late blight. Tomato plants were inoculated, and disease severity was measured. Extractions were performed using microneedles, and LAMP assays were run in tubes (with hydroxynaphthol blue) on a heat block or on a newly designed microfluidic slide chip on a heat block or a slide heater. Fluorescence on the microfluidic chip slides was visualized using EvaGreen and photographed on a smartphone. Plants inoculated with X. perforans or tomato spotted wilt virus tested positive prior to visible disease symptoms, whereas Phytophthora infestans and A. linariae were detected at the time of visual disease symptoms. LAMP assays were more sensitive than PCR, and the limit of detection was 1 pg of DNA for both A. linariae and X. perforans. The LAMP assay designed for early blight detected all three species of Alternaria that infect tomato and is thus an Alternaria spp. assay. This study demonstrates the utility of rapid microneedle extraction followed by LAMP on a microfluidic chip for rapid diagnosis of four important tomato pathogens.

Deep autoencoder based image enhancement approach with hybrid feature extraction for plant disease detection using supervised classification

Plant leaf diseases pose significant threats to global agriculture, leading to reduced crop yields and economic losses. Rapid and accurate disease detection is essential for timely interventions and sustainable farming practices. This study presents an innovative approach for plant leaf disease detection by integrating wavelet analysis, color, and texture features, coupled with autoencoder denoising and support vector machine (SVM) classification. Wavelet analysis is employed to extract multi-resolution features, capturing intricate details at different scales. Furthermore, color and texture characteristics are extracted to encompass a broad spectrum of visual information crucial for distinguishing diseases. The Autoencoder model helps to enhance the feature representation that mitigates the impact of noise and irrelevant data. The SVM classifier is utilized to learn complex patterns and accurately classify different disease classes. The combined model of wavelet, color, and texture attributes, in combination with autoencoder denoising and SVM classification, markedly enhances the precision and efficiency of disease detection in contrast to conventional methods. The system's performance is evaluated using a PlantVillage dataset, showcasing its adaptability to different plant species and disease types. The overall performance is obtained as 98.60%, 97.25%, 96.89%, and 97.20% in terms of accuracy, precision, recall, and F-Score, respectively.

Improved FasterViT model for citrus disease diagnosis

Despite advances in deep learning for plant leaf disease recognition, accurately distinguishing morphological features under varying environmental conditions continues to pose significant challenges. Traditional deep learning models often fail to effectively merge local and global information, especially in small-scale datasets, impairing performance and elevating training costs. Focusing on citrus diseases, we propose an improved FasterViT Model, an advanced hybrid CNN-ViT framework that builds upon the FasterViT model. The proposed model seamlessly integrates CNN's rapid local learning capabilities with ViT's global information processing strength, thereby effectively extracting complex textures and morphological features from images. Cross-stage alternating Mixup and Cutout methods are strategically employed to enhance model robustness and generalization capabilities, particularly valuable for fast learning on small-scale datasets by simulating a more diverse training environment. Triplet Attention and AdaptiveAvgPool mechanisms are utilized to reduce training costs and optimize training performance. The proposed model is tested on both our specially constructed small-scale citrus disease dataset called in-field small dataset and the comprehensive PlantVillage dataset. The experimental results demonstrated that the model exhibits the capability of fast learning and adaptation to small sample training in plant disease detection tasks, and demonstrates the effectiveness of our improvement approach in improving model accuracy and reducing training costs. Additionally, its exemplary performance in transfer learning scenarios underscores its adaptability and broad applicability. This study not only highlights the efficacy of the improved FasterViT model in addressing the complexities of plant disease image recognition but also pioneers a new paradigm for developing efficient, scalable, and robust classification systems.

Advancing Agriculture through Artificial Intelligence, Plant Disease Detection Methods, Applications, and Limitations

In recent years, the integration of artificial intelligence (AI) into agriculture has transformed traditional farming practices. One area of significant advancement is in the detection of plant diseases, where AI-driven technologies offer innovative solutions to mitigate crop losses and enhance agricultural productivity. This paper explores the latest methodologies, applications, and challenges in utilizing AI for plant disease detection. We review various AI techniques, including machine learning, computer vision, and deep learning, that have been deployed to accurately identify and diagnose plant diseases. Additionally, we discuss the practical applications of these technologies in real-world agricultural settings, highlighting their potential to revolutionize crop management practices. Despite the promising developments, we also address the limitations and obstacles faced in implementing AI-based plant disease detection systems, including issues related to data quality, model generalization, and scalability. By critically examining the current landscape of AI-driven plant disease detection, this paper aims to provide insights for researchers, practitioners, and policymakers to further advance the integration of AI technologies in agriculture.

Detection of fusarium wilt-induced physiological impairment in strawberry plants using hyperspectral imaging and machine learning

AbstractStrawberry (Fragraria x ananassa) is a crop affected by various soil-borne fungal pathogens with mostly non-specific foliar symptoms and often requiring laboratory isolation for correct diagnosis. Moreover, these nonspecific foliar symptoms, appreciated by the human eye, appear after some time following infection by the pathogen. Early detection of plant diseases is one of the primary objectives in agriculture because it may contribute to identifying more tolerant cultivars in breeding programs and optimise pesticide use in agricultural production with earlier applications in emerging disease foci. New technologies, such as remote sensing and machine learning (ML) algorithms, have arisen as potential tools to improve the ability to detect and classify different crop diseases. The combined use of hyperspectral imagery and ML algorithms were investigated to detect and classify the physiological stress caused by early infections of Fusarium wilt in strawberry plants. Six ML models, namely artificial neural network, decision tree, K-nearest neighbour, support vector machine, multinomial logistic regression and Naïve Bayes were developed to estimate physiological stress associated with Fusarium wilt disease. The results showed that stomatal conductance (gs) and photosynthesis (A) declined even without visual symptoms of the disease. Among the six ML models evaluated, the artificial neural network model showed the highest classification performance with an overall accuracy of 81%, regardless of the physiological parameter utilized for model training. Moreover, the artificial neural network accurately predicted the absolute values of both physiological parameters (gs and A) based on the complete spectral signature from visually healthy foliar tissue, achieving coefficients of determination of 84% and 81%, respectively. Consequently, ML models utilizing physiological response data and hyperspectral imaging exhibited remarkable robustness, facilitating the estimation of Fusarium wilt severity in strawberry plants even without visual symptoms.

Machine Learning Innovations for Precise Plant Disease Detection: A Review

The sustainable agriculture practices demands new innovations identifying plant diseases and instead of crop disease detection and precision and efficacy. An extensive review of the literature found through PubMed searches indicates a gap in the present approaches, which highlights the need for sophisticated machine learning solutions in the field of plant pathology. This study involves a comprehensive review of relevant publications collected via PubMed searches. The methodology involves the analysis of machine learning algorithms, datasets utilized and techniques applied for plant disease detection. Special attention is given to recent advancements in the field, focusing on the development and optimization of models tailored for precise and reliable disease identification. The study reveals compelling results, underscoring the transformative impact of machine learning innovations on plant disease detection accuracy. Specific algorithms exhibit superior performance, with implications for widespread applications in precision agriculture. These outcomes not only enhance current disease identification capabilities but also lay the groundwork for future advancements in automated and high-precision plant pathology diagnostics. The integration of machine learning emerges as a pivotal force in reshaping the landscape of plant disease detection.

Comparative Investigation of Deep Convolutional Networks in Detection of Plant Diseases

Comparative Investigation of Deep Convolutional Networks in Detection of Plant Diseases

Enhanced deep learning model architecture for plant disease detection in Chilli plants

Enhanced deep learning model architecture for plant disease detection in Chilli plants

A Multitask Learning-Based Vision Transformer for Plant Disease Localization and Classification

Plant disease detection is a critical task in agriculture, essential for ensuring crop health and productivity. Traditional methods in this context are often labor-intensive and prone to errors, highlighting the need for automated solutions. While computer vision-based solutions have been successfully deployed in recent years for plant disease identification and localization tasks, these often operate independently, leading to suboptimal performance. It is essential to develop an integrated solution combining these two tasks for improved efficiency and accuracy. This research proposes the innovative Plant Disease Localization and Classification model based on Vision Transformer (PDLC-ViT), which integrates co-scale, co-attention, and cross-attention mechanisms and a ViT, within a Multi-Task Learning (MTL) framework. The model was trained and evaluated on the Plant Village dataset. Key hyperparameters, including learning rate, batch size, dropout ratio, and regularization factor, were optimized through a thorough grid search. Early stopping based on validation loss was employed to prevent overfitting. The PDLC-ViT model demonstrated significant improvements in plant disease localization and classification tasks. The integration of co-scale, co-attention, and cross-attention mechanisms allowed the model to capture multi-scale dependencies and enhance feature learning, leading to superior performance compared to existing models. The PDLC-ViT model evaluated on two public datasets achieved an accuracy of 99.97%, a Mean Average Precision (MAP) of 99.18%, and a Mean Average Recall (MAR) of 99.11%. These results underscore the model's exceptional precision and recall, highlighting its robustness and reliability in detecting and classifying plant diseases. The PDLC-ViT model sets a new benchmark in plant disease detection, offering a reliable and advanced tool for agricultural applications. Its ability to integrate localization and classification tasks within an MTL framework promotes timely and accurate disease management, contributing to sustainable agriculture and food security.

Enhancing Plant Disease Detection through Transfer Learning by Incorporating MemoryAugmented Networks and Meta-Learning Approaches

Transfer learning has revolutionized automated plant disease detection by leveraging pre-trained convolutional neural networks (CNNs) on large-scale datasets like ImageNet. This paper explores advanced methodologies in transfer learning, focusing on the integration of memory-augmented networks and meta-learning approaches. These enhancements aim to improve model adaptation to new disease types and environmental conditions, thereby enhancing accuracy and robustness in agricultural applications. The paper reviews existing literature, discusses methodologies, and suggests future research directions to advance the field of AI-driven plant pathology.

An ensemble deep learning models approach using image analysis for cotton crop classification in AI-enabled smart agriculture

BackgroundAgriculture is one of the most crucial assets of any country, as it brings prosperity by alleviating poverty, food shortages, unemployment, and economic instability. The entire process of agriculture comprises many sectors, such as crop cultivation, water irrigation, the supply chain, and many more. During the cultivation process, the plant is exposed to many challenges, among which pesticide attacks and disease in the plant are the main threats. Diseases affect yield production, which affects the country’s economy. Over the past decade, there have been significant advancements in agriculture; nevertheless, a substantial portion of crop yields continues to be compromised by diseases and pests. Early detection and prevention are crucial for successful crop management.MethodsTo address this, we propose a framework that utilizes state-of-the-art computer vision (CV) and artificial intelligence (AI) techniques, specifically deep learning (DL), for detecting healthy and unhealthy cotton plants. Our approach combines DL with feature extraction methods such as continuous wavelet transform (CWT) and fast Fourier transform (FFT). The detection process involved employing pre-trained models such as AlexNet, GoogLeNet, InceptionV3, and VGG-19. Implemented models performance was analysed based on metrics such as accuracy, precision, recall, F1-Score, and Confusion matrices. Moreover, the proposed framework employed ensemble learning framework which uses averaging method to fuse the classification score of individual DL model, thereby improving the overall classification accuracy.ResultsDuring the training process, the framework achieved better performance when features extracted from CWT were used as inputs to the DL model compared to features extracted from FFT. Among the learning models, GoogleNet obtained a remarkable accuracy of 93.4% and a notable F1-score of 0.953 when trained on features extracted by CWT in comparison to FFT-extracted features. It was closely followed by AlexNet and InceptionV3 with an accuracy of 93.4% and 91.8% respectively. To further improve the classification accuracy, ensemble learning framework achieved 98.4% on the features extracted from CWT as compared to feature extracted from FFT.ConclusionThe results show that the features extracted as scalograms more accurately detect each plant condition using DL models, facilitating the early detection of diseases in cotton plants. This early detection leads to better yield and profit which positively affects the economy.

MLNAS: Meta-learning based neural architecture search for automated generation of deep neural networks for plant disease detection tasks

ABSTRACT Plant diseases pose a significant threat to agricultural productivity worldwide. Convolutional neural networks (CNNs) have achieved state-of-the-art performances on several plant disease detection tasks. However, the manual development of CNN models using an exhaustive approach is a resource-intensive task. Neural Architecture Search (NAS) has emerged as an innovative paradigm that seeks to automate model generation procedures without human intervention. However, the application of NAS in plant disease detection has received limited attention. In this work, we propose a two-stage meta-learning-based neural architecture search system (ML NAS) to automate the generation of CNN models for unseen plant disease detection tasks. The first stage recommends the most suitable benchmark models for unseen plant disease detection tasks based on the prior evaluations of benchmark models on existing plant disease datasets. In the second stage, the proposed NAS operators are employed to optimize the recommended model for the target task. The experimental results showed that the MLNAS system’s model outperformed state-of-the-art models on the fruit disease dataset, achieving an accuracy of 99.61%. Furthermore, the MLNAS-generated model outperformed the Progressive NAS model on the 8-class plant disease dataset, achieving an accuracy of 99.8%. Hence, the proposed MLNAS system facilitates faster model development with reduced computational costs.

PENERAPAN JARINGAN SARAF TIRUAN DENGAN ALGORITMA PERCEPTRON PADA DETEKSI PENYAKIT BAWANG MERAH

This research focuses on advances in the onion (Allium ascalonicum) disease detection system using Artificial Neural Networks (ANN) with the Perceptron Algorithm. Onions are an important horticultural commodity with substantial economic importance in Indonesia. However, shallot production is often hampered by various diseases that can cause significant losses. Timely identification and effective management of the disease is essential to reduce its adverse effects. Expert systems that utilize artificial intelligence, especially ANN, have proven effective in plant disease detection. ANN imitates the way the human brain works in recognizing patterns and making decisions based on the data obtained. This study used data on onion disease symptoms which consisted of 13 symptoms and 7 types of disease. Each symptom and disease is connected in a knowledge base which is then analyzed using the Perceptron Algorithm. Perceptron is a supervised learning algorithm used for classification and pattern recognition. The training process involves initializing weights and biases, calculating output responses, and adjusting weights if errors occur. The final result of this system is the identification of the type of disease that attacks shallots based on the symptoms observed. The implementation of this system includes a home display, diagnostic display, and diagnostic results display. This system allows users to diagnose shallot diseases by answering questions related to the symptoms observed. The diagnosis results will display the type of disease and the recommended solution. This research shows that the Perceptron algorithm can be applied effectively in shallot disease detection systems, helping farmers identify and manage diseases more quickly and accurately.