Leaf Disease Research Articles

Enhanced detection of tomato leaf diseases using ensemble deep learning: INCVX-NET model

<p align="justify"><span lang="EN-US">Automated leaf disease detection quickly identifies early symptoms, and saves time on large farms. Traditional methods like visual inspection and laboratory detection are prevalent despite being labor-intensive, time-consuming, and susceptible to human error. Recently, deep learning (DL) has emerged as a promising alternative for crop disease recognition. However, these models usually demand extensive training data and face problems in generalization due to the diverse features among different crop diseases. This complexity makes it difficult to achieve optimal recognition performance across all scenarios. To solve this issue, a novel ensemble approach INCVX-Net is proposed to integrate the three DL models, ‘Inception, visual geometry group (VGG)-16, and Xception’ using weighted averaging ensemble for tomato crop leaf disease detection. This approach utilizes the strengths of three DL models to recognize a wide range of disease patterns and captures even slight changes in leaf characteristics. INCVX-Net achieves an impressive 99.5% accuracy in disease detection, outperforming base models such as InceptionV2 (93.4%), VGG-16 Net (92.7%), and Xception (95.2%). This significant leap in accuracy demonstrates the growing power of ensemble DL models in disease detection compared to standalone DL models. The research paves the groundwork for future advancements in disease detection, enhancing precision agriculture through ensemble models.</span></p>

WE-DeepLabV3+: A lightweight segmentation model for Panax notoginseng leaf diseases

Panax notoginseng plays an important role in traditional Chinese medicine. However, diseases pose a significant threat to the quality and yield of P. notoginseng. The main challenge related to the identification of P. notoginseng leaf diseases is how to achieve good performance in the case of small diseased spots on P. notoginseng leaf, overlapping edges of diseased leaf and the difficulty in mobile deployment. A lightweight semantic segmentation model Window Efficient-DeepLabv3+ is proposed for segmentation and quantification of P. notoginseng leaf diseases. We propose the Window Attention-ASPP module and present a hierarchical stacking of features, which improves model accuracy for minor target lesions while reducing parameters. In addition, a lightweight backbone network MobileNetV2 is utilized as a feature extraction module. The decoding stage introduces the Efficient Channel Attention module, which effectively improves the accuracy of the segmentation of the blade contour. Experimental results yielded the Mean Intersection Over Union, Mean Precision, and Mean Recall metrics of WE-DeepLabV3+ network to be 82.0 %, 87.6 %, and 92.4 % respectively, outperforming other segmentation models such as UNet, PSPNet, CaraNet, SegNet, and BiSeNetV2. Moreover, the number of parameters has been reduced by 90.6 % with only 5.1 M parameters. Finally, the method is used to quantify the disease of P. notoginseng leaf, the error is only 1.15 % and 0.82 %, which proves that it can quantify the disease severity accurately. Thus, the proposed method holds great significance for raising the yield and quality of P. notoginseng, also providing reliable guidance for precise fertilization and drug control.

Transfer learning based leaf disease detection model using convolution neural network

The plants are attacked from various micro-organisms, bacterial illnesses, and pests. The signs are normally identified via leaves, stem, or fruit inspection. Illnesses that generally appeared on vegetation are from leaves and causes big harm if not managed in the early ranges. To stop this huge harm and manipulate the unfold of disorder this work implements a software system. This research work customs deep neural network to gain knowledge of probable illnesses on leaves within the early phases so it can be stopped early. Deep neural network (DNN) used for image classification. This work mainly focuses a neural network model of leaves ailment detection. The commonly available plant leaves dataset is undertaken with a dataset having special training of disease detection. In this work VGG16, ResNet50, Inception V3 and Inception ResNetV2 architectural techniques are implemented to generate and compare the results. Results are compared on the factors like precision, accuracy, recall and F1-Score. The results lead to the conclusion, that the convolution neural network (CNN) is more impactful technique to perceive and predict plant diseases.

Tomato leaf disease recognition system using Faster R-CNN

Tomato leaf disease recognition system using Faster R-CNN

Automatic Detection and Classification of Grape Leaf Diseases based on Deep Learning and Enhanced Chameleon Swarm Algorithm

Automatic Detection and Classification of Grape Leaf Diseases based on Deep Learning and Enhanced Chameleon Swarm Algorithm

Development of an IoT based smart potato leaf diseases monitoring and controlling system with image processing

Development of an IoT based smart potato leaf diseases monitoring and controlling system with image processing

Efficient detection of tomato leaf diseases using optimized Compact Convolutional Transformers (CCT) Model

Tomato crops are highly susceptible to various leaf diseases, posing a significant threat to agricultural yield and economic viability. Traditional disease detection methods, reliant on expert visual inspection, are time-intensive, inconsistent, and impractical on a large scale. This study addresses these limitations by developing an optimized Compact Convolutional Transformer (CCT) model tailored to efficiently and accurately classify tomato leaf diseases using image data. Leveraging a dataset of over 30,000 images spanning multiple disease classes and augmented through advanced techniques, we trained and tested the CCT model alongside popular transfer learning architectures, including VGG16, ResNet50, and Vision Transformers (ViTs). Our methodology involved extensive hyperparameter tuning and comparative analysis to maximize model accuracy and robustness. Results demonstrate that the optimized CCT model outperforms competing architectures, achieving an impressive accuracy of 98.87%, significantly higher than baseline models. The analysis further includes learning curves, confusion matrices, and ROC-AUC evaluations, which validate the model's reliability and ability to generalize across diverse image conditions. This work underscores the potential of hybrid transformer models in agriculture, offering a scalable, high-performance solution for the real-time detection of tomato leaf disease. The scalability of our solution makes it adaptable to various agricultural settings, ensuring its forward-thinking nature.

FRRSA: Fractional Remora Reptile Search Algorithm-based LeNet for rice leaf disease classification

ABSTRACT Rice is considered as a chief food in most of the areas across India, simultaneously rice plants are infected by diverse diseases strongly, which causes vital losses in the agricultural field. So, earlier and more precise prediction of diverse rice crop diseases becomes a main challenge to farmers and research persons. Hence, a Fractional Remora Reptile search algorithm-LeNet (FRRSA-LeNet) is proposed for rice leaf disease classification. At first, the input leaf image is passed to the image pre-processing stage. In image pre-processing, a bilateral filter is utilized for the quality improvement of an image. Afterwards, segmentation is carried out, where the GrabCut algorithm is employed for segmenting disease-affected areas. Then, the image augmentation phase is conducted and in the feature extraction, CNN features and statistical features are extracted for the following process. Lastly, rice leaf disease classification is conducted utilizing LeNet, which is tuned by the proposed FRRSA. Moreover, FRRSA is introduced by assimilating Fractional Calculus (FC), Remora Optimization Algorithm (ROA), and Reptile Search Algorithm (RSA). FRRSA-LeNet acquired maximal testing accuracy of 0.903, TPR of 0.883, TNR of 0.904, and MSE of 0.800. The proposed model accurately identifies rice leaf diseases, enhancing the overall effectiveness of the classification process.

Image-based rice leaf disease detection using CNN and generative adversarial network

Image-based rice leaf disease detection using CNN and generative adversarial network

Rose Plant Leaf Disease Recognition Using Machine Learning Methodologies

The most popular flowers in the world are roses, not only cheer people up but also support livelihoods. Diseases, however, can harm these priceless flowers' health and negatively affect both their quality and the growers' livelihoods. The increased occurrence of ailments in rose plants poses a severe danger to the ornamental flower industry and agricultural productivity. In this paper, we describe a novel deep learning-based method for the automated diagnosis of leaf diseases in rose plants. A big dataset containing images of both healthy and damaged rose leaves was carefully picked to illustrate different disease types and stages. To analyze and identify the visual characteristics that correspond to various illnesses, we used a Convolutional Neural Network architecture, Support Vector Machine, and K-Nearest Neighbors architectures specifically intended for picture classification tasks. We address the interpretability and explainability of the model's predictions in addition to performance indicators, offering insights into the decision-making process. This work addresses a fundamental requirement for effective and long-lasting disease management in rose cultivation by bridging the gap between deep learning and plant pathology. CNNs are often the preferred choice due to their ability to automatically learn relevant features from raw pixel values.

HCAR-AM ground nut leaf net: Hybrid convolution-based adaptive ResNet with attention mechanism for detecting ground nut leaf diseases with adaptive segmentation

ABSTRACT Estimating the optimal answer is expensive for huge data resources that decrease the functionality of the system. To solve these issues, the latest groundnut leaf disorder identification model by deep learning techniques is implemented. The images are collected from traditional databases, and then they are given to the pre-processing stage. Then, relevant features are drawn out from the preprocessed images in two stages. In the first stage, the preprocessed image is segmented using adaptive TransResunet++, where the variables are tuned with the help of designed Hybrid Position of Beluga Whale and Cuttle Fish (HP-BWCF) and finally get the feature set 1 using Kaze Feature Points and Binary Descriptors. In the second stage, the same Kaze feature points and the binary descriptors are extracted from the preprocessed image separately, and then obtain feature set 2. Then, the extracted feature sets 1 and 2 are concatenated and given to the Hybrid Convolution-based Adaptive Resnet with Attention Mechanism (HCAR-AM) to detect the ground nut leaf diseases very effectively. The parameters from this HCAR-AM are tuned via the same HP-BWCF. The experimental outcome is analysed over various recently developed ground nut leaf disease detection approaches in accordance with various performance measures.

Comparison of the performance of classical and quantum machine‐learning methods on the detection of sugar beet Cercospora leaf disease

AbstractImage processing and machine‐learning (ML) techniques are essential for the detection of diseases and pests in plants. This study explored the application of quantum ML (QML) algorithms for the early detection of Cercospora beticola leaf disease in sugar beet, which causes significant impact on global sugar production. Using a dataset of 1065 images (739 diseased and 326 healthy), we extracted 70 ML statistical features, including 10 from the grey‐level co‐occurrence matrix (GLCM) and 60 colour‐related features. Performance evaluations of classical ML algorithms, such as random forest (RF; 91.95% accuracy) and extreme gradient boosting (91.95% accuracy), demonstrated strong results compared to quantum approaches. Notably, the quantum support vector classifier (QSVC) achieved an accuracy of 85% with perfect recall of 1.00, while the variational quantum classifier (VQC) recorded an accuracy of 88.73%. Dimensionality reduction via principal component analysis reduced features from 70 to 5, enabling effective classification with competitive results: ML (RF) 91.41%, VQC with limited‐memory Broyden–Fletcher–Goldfarb–Shanno with box constraints (L_BFGS_B) 88.73% and QSVC 85%. These findings highlight the potential of QML algorithms in improving agricultural disease identification and aiding in the advancement of more efficient, sustainable farming techniques.

Hyperspectral Imaging Combined with Deep Learning for the Early Detection of Strawberry Leaf Gray Mold Disease

The presence of gray mold can seriously affect the yield and quality of strawberries. Due to their susceptibility and the rapid spread of this disease, it is important to develop early, accurate, rapid, and non-destructive disease identification strategies. In this study, the early detection of strawberry leaf diseases was performed using hyperspectral imaging combining multi-dimensional features like spectral fingerprints and vegetation indices. Firstly, hyperspectral images of healthy and early affected leaves (24 h) were acquired using a hyperspectral imaging system. Then, spectral reflectance (616) and vegetation index (40) were extracted. Next, the CARS algorithm was used to extract spectral fingerprint features (17). Pearson correlation analysis combined with the SPA method was used to select five significant vegetation indices. Finally, we used five deep learning methods (LSTMs, CNNs, BPFs, and KNNs) to build disease detection models for strawberries based on individual and fusion characteristics. The results showed that the accuracy of the recognition model based on fused features ranged from 88.9% to 96.6%. The CNN recognition model based on fused features performed best, with a recognition accuracy of 96.6%. Overall, the fused feature-based model can reduce the dimensionality of the classification data and effectively improve the predicting accuracy and precision of the classification algorithm.

Plant Disease Classification Using Convolutional Neural Networks

The agricultural sector faces significant losses due to plant diseases, particularly in major crops such as potatoes, tomatoes, and bell peppers. This paper presents a machine learning-based approach to classify diseases in these crops using leaf images. A Convolutional Neural Network (CNN) model was constructed and trained on datasets of healthy leaf images and diseased leaf images from potato, tomato, and bell pepper plants. The model successfully classifies diseases such as Bacterial Spot (for bell peppers), Early Blight, Late Blight, Mosaic Virus, Leaf Mold (for tomatoes), and with a classification accuracy of 93%, this system provides early detection, helping farmers take timely action to reduce disease impact and increase crop yield.

2024/00264 Harnessing Transdisciplinary Knowledge: Integrated Deep Learning Techniques for Accurate Tomato Leaf Disease Classification

The study proposes a transdisciplinary approach integrating knowledge from fields such ascomputer science, botany, and data science to classifying leaf diseases. We integrated two deep-learningmodels that combine the strengths of the Inception network and the ResNet architecture to address thechallenge of accurately classifying tomato leaf diseases. The Inception network’s ability to quickly pick upvisual features on multiple scales is used to pull out fine-grained details that are needed to tell the differencebetween small changes in the shape of tomato leaves and disease symptoms. The ResNet architecture isgood at learning deep representations and getting around the vanishing gradient problem. This lets themodel learn the high-level concepts and complicated connections between different tomato leaf diseasepatterns. The integration of these two powerful deep-learning techniques results in a robust and highlyperformant tomato leaf classification model. Extensive tests on a 10-class dataset of tomato leaves, with 9disease categories and 1 healthy class, show that the proposed model works better than others, with a testset accuracy of 98.07%. The findings of this research contribute to the advancement of automated andefficient tomato leaf disease detection systems, which can aid in the early identification and management oftomato diseases, leading to improved crop yields and quality.

Multi-kernel inception aggregation diffusion network for tomato disease detection.

Tomato leaf diseases significantly impact the yield and quality of tomatoes during cultivation, the main of which are septoria leaf spot, leaf curl virus, verticillium wilt, and early blight. These diseases necessitate prompt detection and management strategies to mitigate their deleterious effects on crop productivity. Due to the considerable scale variations in diseased tomato leaves, accurate and rapid detection and diagnosis remain challenging. To address the detection of tomato leaf diseases at different scales, we propose a real-time detection model incorporating a Multi-kernel Inception Aggregation Diffusion Network. In this paper, (1) We present a Multi-kernel Inception Aggregation Diffusion Network (MIADN) for the feature processing stage, which facilitates the aggregation and diffusion of multi-scale features across hierarchical levels, benefiting the detection of targets at various scales. (2) We present the Multi-kernel Inception Module (MKIM), designed to extract multi-scale object features using diverse convolution kernels, thereby enhancing the model's feature fusion and representation capabilities. (3) We incorporate the efficient FasterNet network at the feature extraction stage to preserve feature diversity and improve the model's ability to extract complex target features. (4) Extensive comparative and ablation experiments demonstrate that our method achieves the mean average precision (mAP50) of 96.6%, surpassing the baseline model by 4.1% and the advanced YOLOv9s model by 2.0%. This method provides an effective solution for high-quality tomato cultivation.

Spinach leaf disease identification based on deep learning techniques

Spinach leaf disease identification based on deep learning techniques

DeepEMPR: coffee leaf disease detection with deep learning and enhanced multivariance product representation

Plant diseases threaten agricultural sustainability by reducing crop yields. Rapid and accurate disease identification is crucial for effective management. Recent advancements in artificial intelligence (AI) have facilitated the development of automated systems for disease detection. This study focuses on enhancing the classification of diseases and estimating their severity in coffee leaf images. To do so, we propose a novel approach as the preprocessing step for the classification in which enhanced multivariance product representation (EMPR) is used to decompose the considered image into components, a new image is constructed using some of those components, and the contrast of the new image is enhanced by applying high-dimensional model representation (HDMR) to highlight the diseased parts of the leaves. Popular convolutional neural network (CNN) architectures, including AlexNet, VGG16, and ResNet50, are evaluated. Results show that VGG16 achieves the highest classification accuracy of approximately 96%, while all models perform well in predicting disease severity levels, with accuracies exceeding 85%. Notably, the ResNet50 model achieves accuracy levels surpassing 90%. This research contributes to the advancement of automated crop health management systems.

Calculation Method Based on Flower Petal Area and Other Plant Leaf Spot Area

In order to calculate the area of irregular shapes such as plant leaves and diseased spots or flower petals, this paper presents a new method to calculate them by using flood fill algorithm, HSV color space, improved k-means algorithm and morphological operation. First, 501 butterfly petal images and pathological leaf images of Bauhinia and phyllotaxus were collected, and then flood was used Fill algorithm selects the disease-free area and records the selected pixel value. HSV color space conversion is applied to the image to facilitate the segmentation of leaves. Then, the improved k-means algorithm is used to extract the binary image of leaves and record the pixel value of the outer contour with morphological closed operation. Finally, the proportion and truth of the disease spots of plant leaves are obtained by calculating the pixel value and the real value of the rectangle in the sampling area Real area. Compared with the results of artificial labeling, the average accuracy of petal area and lesion area of Phalaenopsis was 96.3% and 96.61%, respectively. It can be seen that the program can calculate the area of irregular shape of plant surface accurately. In conclusion, this method can replace the artificial grid method to calculate the information of plant leaf area and disease proportion, and effectively reduce the work intensity of experimental personnel.

An intelligent agriculture monitoring framework for leaf disease detection using YOLOv7

Agriculture is one of the most important economic sectors on which societies have relied since ancient times. With the recent development of technology, agriculture has also been incorporating modern techniques such as the Internet of Things and Artificial Intelligence to improve productivity and monitor the farming process. One of agriculture’s most prominent issues is the spread of plant diseases and the lack of real-time monitoring. Various systems and operations have recently been developed to predict and diagnose plant diseases. However, current operations have been selective, focusing on a specific aspect without addressing other important aspects, resulting in either partial or compound application of results, rendering the desired outcomes ineffective. To deal with such challenges, we propose an intelligent framework for real-time agriculture monitoring and disease detection, namely a system for monitoring plant diseases using YOLOv7. In the proposed framework, a rule-based policy has been designed for detecting plant diseases using online plant leaf monitoring, sensors, and surveillance cameras. Images of plant leaves captured by different cameras are sent in real-time to central cloud servers for disease detection. The improved YOLOv7 technology is utilized for plant disease detection, and the proposed system has been evaluated using a dataset of diseased tomato leaves, comparing it with different models based on various performance metrics to demonstrate its effectiveness, achieving an accuracy of 96%.