Tea leaf disease detection and identification based on YOLOv7 (YOLO-T)

Using a multi-convolutional neural network to automatically identify small-sample tea leaf diseases

<p>The detection of leaf diseases using modern technology has significant importance in agriculture and artificial intelligence. Deep learning, specifically, plays a crucial role in this field, as it enables accurate and efficient disease classification. Early detection of leaf diseases is vital to implementing timely treatments and preventing widespread damage to leaves. Leaf diseases can be caused by various factors, including bacteria, fungi, viruses, and other pathogens. Among them, bacteria and viruses are the most invasive and can lead to substantial yield losses if not identified and treated promptly. Bacterial and viral infections are common in agricultural settings, affecting leaves of all types and ages. Our research aims to propose a transfer learning-based model for predicting leaf diseases using a dataset of leaf images. The images will be classified into healthy or diseased leaves based on extracted features. The proposed model, named Leaf Disease Transfer Learning Algorithm (LDTLA), demonstrates promising results with an average accuracy of 97.37% on the dataset. Utilizing convolutional neural networks (CNN) and deep learning techniques, our LDTLA model outperforms previous quantitative and qualitative research studies in leaf disease detection. This advanced approach to leaf disease identification holds the potential to revolutionize agriculture by enabling farmers to make informed decisions, implement targeted treatments, and minimize leaf losses caused by diseases.<strong></strong></p>

Agriculture is the most important sector in Indian economy. India occupies the second highest rank in farm outputs in the world. Its contribution to the development of Indian economy has immense potential. So agriculture products may play vital role for economic growth. But the different kind of diseases in plant decreases the production of crops and growth rate of farmers. To identify and monitor the leaf diseases manually by farmers is very difficult. This is one of the reasons to develop an automatic leaf diseases detection model. The proposed model helps in automatic detection of different plant diseases at early stages. Thus, the production will increase in many folds. The main aim of this study is to identify different types of leaf diseases. Different feature extraction techniques have been used to enhance the classification accuracy. Support Vector Machine (SVM), Random Forest and Logistic Regression have been applied to classify different types of leaf diseases. When obtained results are compared SVM outperforms other two classifiers. Results show that, the model can be used in real life applications.

Tea cultivation holds significant economic and agricultural value across many regions, particularly in developing countries, where it supports millions of livelihoods. However, the productivity and quality of tea leaves are severely threatened by a wide range of foliar diseases, which, if undetected or misdiagnosed, can lead to substantial crop losses. Existing manual and traditional image-based detection methods are time-consuming and often lack the precision required for early-stage identification across multiple disease types. To address these limitations, this study proposes TeaLeafNet-GWO, an intelligent deep learning framework that integrates a hybrid CNN–Transformer architecture with Gray Wolf Optimization (GWO) for automated tea leaf disease classification. The model combines the spatial feature extraction power of CNNs with the global contextual awareness of Transformer blocks (MobileViT), enhanced by channel-wise recalibration using a squeeze-and-excitation network (SENet). GWO is employed to fine-tune critical hyperparameters, ensuring optimal performance and generalization. The dataset comprises 999 high-resolution images representing eight distinct tea leaf conditions, including both healthy and diseased classes. Preprocessing techniques such as contrast-limited adaptive histogram equalization (CLAHE) and median filtering were applied to enhance image quality and suppress noise. The model was trained and evaluated using a stratified split (70% training, 30% testing). TeaLeafNet-GWO achieved an average accuracy of 97.72%, F1-score of 86.66%, and Matthews correlation coefficient (MCC) of 84.44% in the training phase, with similarly high performance in testing, outperforming several state-of-the-art hybrid models in comparative analysis. These results demonstrate TeaLeafNet-GWO’s effectiveness as a lightweight, scalable, and accurate solution for real-time disease monitoring, offering tangible benefits to precision agriculture and smart farming systems.

The identification of plant leaf diseases is crucial in precision agriculture, playing a pivotal role in advancing the modernization of agriculture. Timely detection and diagnosis of leaf diseases for preventive measures significantly contribute to enhancing both the quantity and quality of agricultural products, thereby fostering the in-depth development of precision agriculture. However, despite the rapid development of research on plant leaf disease identification, it still faces challenges such as insufficient agricultural datasets and the problem of deep learning-based disease identification models having numerous training parameters and insufficient accuracy. This paper proposes a plant leaf disease identification method based on improved SinGAN and improved ResNet34 to address the aforementioned issues. Firstly, an improved SinGAN called Reconstruction-Based Single Image Generation Network (ReSinGN) is proposed for image enhancement. This network accelerates model training speed by using an autoencoder to replace the GAN in the SinGAN and incorporates a Convolutional Block Attention Module (CBAM) into the autoencoder to more accurately capture important features and structural information in the images. Random pixel Shuffling are introduced in ReSinGN to enable the model to learn richer data representations, further enhancing the quality of generated images. Secondly, an improved ResNet34 is proposed for plant leaf disease identification. This involves adding CBAM modules to the ResNet34 to alleviate the limitations of parameter sharing, replacing the ReLU activation function with LeakyReLU activation function to address the problem of neuron death, and utilizing transfer learning-based training methods to accelerate network training speed. This paper takes tomato leaf diseases as the experimental subject, and the experimental results demonstrate that: (1) ReSinGN generates high-quality images at least 44.6 times faster in training speed compared to SinGAN. (2) The Tenengrad score of images generated by the ReSinGN model is 67.3, which is improved by 30.2 compared to the SinGAN, resulting in clearer images. (3) ReSinGN model with random pixel Shuffling outperforms SinGAN in both image clarity and distortion, achieving the optimal balance between image clarity and distortion. (4) The improved ResNet34 achieved an average recognition accuracy, recognition precision, recognition accuracy (redundant as it's similar to precision), recall, and F1 score of 98.57, 96.57, 98.68, 97.7, and 98.17%, respectively, for tomato leaf disease identification. Compared to the original ResNet34, this represents enhancements of 3.65, 4.66, 0.88, 4.1, and 2.47%, respectively.

Efficiently managing plant leaf conditions in large agricultural areas requires automated detection of leaf diseases. This research presents an integrated approach utilizing Content-Based Image Retrieval (CBIR) systems for the automatic identification of plant leaf diseases. The CBIR system employs preprocessing, segmentation using the graph cut superpixel method and feature extraction via 2D Convolutional Neural Network (CNN) models, which is focused primarily on detecting colorful plant flake complaints prevalent in blueberry, cherry and apple crops. Additionally, Gradient Boosting Trees (GBT) serve as classifiers to point to database stores of class predictions made by the classifiers for plant leaf images. Comparable deep learning models such as InceptionV3 and ResNet50 in the proposed CBIR system extract high-level features from images, predict leaf classes and store them in the database. Euclidean similarity measures between feature representations further assess the CBIR system's overall performance, with deep learning models successfully supporting retrieval and image classification tasks when compared with 2D CNN models.

Plant leaves and crops play a crucial role as a primary food source globally, making significant contributions to dietary iron intake (9%) and energy consumption (23%) per capita in the Asian region. Bacterial, yeast, and other microbial diseases pose significant challenges to farmers as they detrimentally impact plant health and reduce crop productivity. The manual diagnosis of these diseases poses a considerable challenge, particularly in regions with a scarcity of professionals specializing in leaves and crop protection. Automating leaf disease detection and providing easily accessible decision-support resources are crucial for facilitating efficient leaf protection strategies and mitigating crop damage. Despite multiple classification methods for diagnosing leaf diseases, a secure and accurate approach that fulfills these requirements has not yet been identified. This paper presents an architectural framework called Lightweight Federated Transfer Learning (LFTL) that addresses the challenge of Leaf Disease Detection and Classification (LDDC) while ensuring data privacy limitations are upheld. A dataset consisting of leaf disease images has been compiled, characterized by an imbalance in the distribution of the diseases. The collection includes four conditions: bacterial decay, brown spot, blast, and tungro, corresponding image counts of 1695, 1551, 1711, and 1419, respectively. Following the preprocessing stage, the LFTL framework was tested using both Independent and Identically Distributed (IID) and non-IID datasets. The study commenced with an efficacy evaluation of the Convolutional Neural Network (CNN) and eight TL models in the LDDC. The framework’s performance was evaluated across different circumstances and compared to conventional and federated learning models. The study’s findings revealed that the LFTL framework outperformed traditional distributed deep-learning classifiers, thus demonstrating its efficacy in individual and multiple client scenarios.

The economic gainful of the country largely join on the crop production quality and quantity. The production profit can be improved by leaf disease detection in correct time. Many image processing methods have been developed for leaf disease detection. Technological improvement allows the task easier and faster at very early stage. Leaf disease is one of the major issue in the agriculture field. The Cotton leaves are affected by the disease named Cercospora, Bacterial blight, Ascochyta blight, and Target spot. General observation by farmers may time consume, expensive and sometimes inaccurate. For this, we introduce Deep Convolutional Neural Network based approach for identifying Cotton leaf diseases automatically. As per our understanding, Deep Convolutional Neural Network is used for the first time to detect cotton leaf disease. We train our data and assess the experimental result which displays the average accuracy of 96 %.

Sugarcane is a vital cash crop grown in many tropical and subtropical regions. Its productivity, however, is significantly affected by various leaf diseases, which, if not detected early, can lead to major yield losses. Traditional disease identification methods are manual, time-consuming, and prone to inaccuracies. To overcome these limitations, this project proposes a deep learning-based system for automatic detection of sugarcane leaf diseases. Using Convolutional Neural Networks (CNNs), the model is trained on a dataset of sugarcane leaf images to accurately identify diseases such as red rot, leaf scald, and mosaic virus. The system provides efficient, real-time detection and can assist farmers in timely decision-making and disease management. This approach improves agricultural practices and contributes to sustainable crop production by integrating modern AI technologies into farming. INDEX TERMS: Sugarcane, Leaf Disease Detection, Deep Learning, CNN, Image Classification, Smart Agriculture, Crop Monitoring, AI in Farming.

Leaf disease detection is a crucial task in modern agriculture, aiding in early diagnosis and prevention of crop infections. In this research paper, authors present a comprehensive study comparing nine widely used pre-trained models, namely DenseNet201, EfficientNetB3, EfficientNetB4, InceptionResNetV2, MobileNetV2, ResNet50, ResNet152, VGG16, and Xception, with our newly developed custom CNN (Convolutional Neural Network) for leaf disease detection. The objective is to determine if our custom CNN can match the performance of these established pre-trained models while maintaining superior efficiency. The authors trained and fine-tuned each pre-trained model and our custom CNN on a large dataset of labeled leaf images, covering various diseases and healthy states. Subsequently, the authors evaluated the models using standard metrics, including accuracy, precision, recall, and F1-score, to gauge their overall performance. Additionally, the authors analyzed computational efficiency regarding training time and memory consumption. Surprisingly, our results indicate that the custom CNN performs comparable to the pre-trained models, despite their sophisticated architectures and extensive pre-training on massive datasets. Moreover, our custom CNN demonstrates superior efficiency, outperforming the pre-trained models regarding training speed and memory requirements. These findings highlight the potential of custom CNN architectures for leaf disease detection tasks, offering a compelling alternative to the commonly used pre-trained models. The efficiency gains achieved by our custom CNN can be beneficial in resource-constrained environments, enabling faster inference and deployment of leaf disease detection systems. Overall, our research contributes to the advancement of agricultural technology by presenting a robust and efficient solution for the early detection of leaf diseases, thereby aiding in crop protection and yield enhancement.

Text recognition in natural scene images is a challenging problem in computer vision. Different than the optical character recognition (OCR), text recognition in natural scene images is more complex due to variations in text size, colors, fonts, orientations, complex backgrounds, occlusion, illuminations and uneven lighting conditions. In this paper, we propose a segmentation-free method based on a deep convolutional recurrent neural network to solve the problem of cursive text recognition, particularly focusing on Urdu text in natural scenes. Compared to the non-cursive scripts, Urdu text recognition is more complex due to variations in the writing styles, several shapes of the same character, connected text, ligature overlapping, stretched, diagonal and condensed text. The proposed model gets a whole word image as an input without pre-segmenting into individual characters, and then transforms into the sequence of the relevant features. Our model is based on three components: a deep convolutional neural network (CNN) with shortcut connections to extract and encode the features, a recurrent neural network (RNN) to decode the convolutional features, and a connectionist temporal classification (CTC) to map the predicted sequences into the target labels. To increase the text recognition accuracy further, we explore deeper CNN architectures like VGG-16, VGG-19, ResNet-18 and ResNet-34 to extract more appropriate Urdu text features, and compare the recognition results. To conduct the experiments, a new large-scale benchmark dataset of cropped Urdu word images in natural scenes is developed. The experimental results show that the proposed deep CRNN network with shortcut connections outperform than other network architectures. The dataset is publicly available and can be downloaded from <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://data.mendeley.com/datasets/k5fz57zd9z/1</uri> .

Agriculture remains the backbone of the Indian economy, yet farmers continue to face significant challenges including unpredictable crop prices, rampant plant diseases, and limited access to timely decision-support tools. This paper presents AgriPredict, an integrated machine learning framework designed to assist farmers and village officials (Talathis) through two primary modules: crop price forecasting and leaf disease identification with treatment recommendations. The crop price forecasting module leverages machine learning regression models trained on historical market data, weather patterns, and regional crop information to predict future prices, enabling farmers to plan sales strategies effectively. The disease identification module employs Convolutional Neural Networks (CNNs) to classify leaf diseases from uploaded images and recommends appropriate pesticide treatments, facilitating early intervention and reduced crop loss. The platform further integrates real-time weather forecasts and links to government agricultural schemes, providing a holistic decision-support environment. Experimental evaluation demonstrates high accuracy in both price forecasting and disease classification tasks. The system is designed with a user-friendly web interface accessible to users with limited technical expertise. AgriPredict contributes toward bridging the technological gap in Indian agriculture, promoting data-driven decisions that improve crop yield, reduce financial losses, and enhance overall agricultural productivity.

Agriculture impacts life and economic status of the people. It employs many people and accounts for major part of Gross Domestic Product. Diseases are quite natural and are a major factor for crop losses. Improper management of diseases results in annual loss of agricultural yield which will have serious effects on the quality, quantity and productivity if no proper care is taken. By using some automatic technique such as image processing, detection of leaf disease is quite significant and beneficial. The use of the most utilized deep learning classification mechanism, Convolutional Neural Network, helps in this regard.

Apples are susceptible to infection by various pathogens during growth, which induces various leaf diseases and thus affects apple quality and yield. The timely and accurate identification of apple leaf diseases is essential to ensure the high-quality development of the apple industry. In practical applications in orchards, the complex background in which apple leaves are located poses certain difficulties for the identification of leaf diseases. Therefore, this paper suggests a novel approach to identifying and classifying apple leaf diseases in complex backgrounds. First, we used a bilateral filter-based MSRCR algorithm (BF-MSRCR) to pre-process the images, aiming to highlight the color and texture features of leaves and to reduce the difficulty of extracting leaf disease features with subsequent networks. Then, BAM-Net, with ConvNext-T as the backbone network, was designed to achieve an accurate classification of apple leaf diseases. In this network, we used the aggregate coordinate attention mechanism (ACAM) to strengthen the network’s attention to disease feature regions and to suppress the interference of redundant background information. Then, the multi-scale feature refinement module (MFRM) was used to further identify deeper disease features and to improve the network’s ability to discriminate between similar disease features. In our self-made complex background apple leaf disease dataset, the proposed method achieved 95.64% accuracy, 95.62% precision, 95.89% recall, and a 95.25% F1-score. Compared with existing methods, BAM-Net has higher disease recognition accuracy and classification results. It is worth mentioning that BAM-Net still performs well when applied to the task of the leaf disease identification of other crops in the PlantVillage public dataset. This indicates that BAM-Net has good generalization ability. Therefore, the method proposed in this paper can be helpful for apple disease control in modern agriculture, and it also provides a new reference for the disease identification of other crops.

Accurate identification of crop leaf diseases is essential for sustainable agriculture, yet variations in leaf morphology and disease symptoms across species remain challenging for traditional machine learning approaches. This study proposes a meta-learning–based stacking ensemble that integrates Convolutional Neural Networks (CNN), Support Vector Machines (SVM), and Random Forest Classifiers (RFC) for multi-species leaf pathology recognition, evaluated on apple and tomato datasets. Each base model independently learns discriminative visual features, and their class-probability outputs are fused through an RBF-SVM meta-learner. Experimental results show that individual models achieve accuracies of 89% (CNN), 88% (SVM), and 85% (RFC), while the proposed stacking ensemble significantly outperforms them with an accuracy of 98% and macro and weighted F1-scores of 0.975. These results demonstrate the effectiveness of heterogeneous meta-learning ensembles in enhancing robustness and generalization for multi-species crop disease classification.