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.

Machine Learning Innovations for Precise Plant Disease Detection: A Review

Background: Agriculture is very important for human existence since time immemorial. Approximately sixty percent of the world&rsquo;s population is dependent on agriculture and its allied activities. Due to plant diseases every year farmers bear heavy economic loss which can be reduced by having an early detection system for plant disease. Conventional plant disease detection techniques are laborious and based on chemical and analytical testing. In this work, we suggest a deep learning method to precisely detect plant diseases by applying the VGG19 convolutional neural network. 
 
Methods: Modifications specifically designed for the classification of plant diseases were applied to the VGG19 model. By first freezing the bottom layers of the pre-trained VGG19 model-which had been trained on the ImageNet dataset-and then fine-tuning the upper layers to better fit the PlantVillage dataset, transfer learning was used. Resizing, standardization and data augmentation are a few of the image preprocessing approaches that were used to increase the variety of the dataset and boost model performance. The efficacy of the model was assessed through the use of metrics like F1-score, recall, accuracy and precision. 
 
Result: The constructed model performed well in the classification of plant diseases, with over 95% accuracy on the test set. The success of the model in generalizing across different plant disease categories was largely attributed to the application of transfer learning and data augmentation. The findings show that deep learning techniques-in particular, the use of VGG19-can significantly enhance agricultural practices and decision-making processes by helping to quickly and accurately identify plant illnesses. VGG9-based image processing offers an accurate and automated solution for plant disease identification, achieving 98% accuracy in classifying healthy and diseased leaves. By integrating mobile applications, drones and smart farming cameras, farmers can detect diseases early and take timely action, improving crop health and yield. Future advancements in dataset expansion and real-time processing will further enhance its effectiveness in precision agriculture. 
&nbsp;

Deep Learning VGG19 Model for Precise Plant Disease Detection

Background: Legume crops are an essential component of global agriculture and are widely supplied for human consumption, livestock feed and soil improvement due to their vital nutritional nature. The economic and nutritional significance of legumes is threatened by a multitude of diseases that can cause substantial yield losses. Traditional methods for disease detection, relying on visual inspection, are often subjective and inefficient, leading to delayed intervention. 
 
Methods: This study investigates the utilization of machine learning algorithms for the early identification of diseases affecting legume crops. A comprehensive evaluation is conducted on machine learning algorithms, namely Support Vector Machines (SVM) and Convolutional Neural Networks (CNN) with respect to the domain of disease detection. Through a comparative analysis of their performance across different environmental conditions and phases of crop development, this study also explores their strengths and weaknesses. 
 
Result: The findings and comparative examination offered significant perspectives on the potential of machine learning algorithms in the realm of early legume crop disease detection. In addition to enhancing crop health and disease management, the research provides support for sustainable agricultural practices and possesses the capacity to augment environmental sustainability and food security through the application of machine learning techniques.&nbsp;

Machine Learning Algorithms for Early Detection of Legume Crop Disease

Background: Plant disease detection remains a major challenge in agriculture, with direct implications for improving crop productivity and ensuring food security. Seasonal variation significantly influences plant characteristics, making the classification of plant leaves by season-specifically summer and winter-important for optimizing disease detection and management strategies. 
 
Methods: In this study, a plant leaf disease detection dataset was developed and categorized based on seasonal conditions. The dataset includes 47 classes representing summer crops and 16 classes for winter crops. To classify plant leaf diseases effectively, we propose a novel dual-encoder Variational Autoencoder (VAE) model that integrates ResNet and VGGNet as parallel encoders. These encoders extract complementary feature maps from the seasonal datasets, which are then concatenated to improve classification accuracy. 
 
Result: Experimental evaluation demonstrates the robustness and accuracy of the proposed approach. The dual-encoder VAE achieved a classification accuracy of 98.86% on the summer dataset and 97.53% on the winter dataset, highlighting the model&rsquo;s ability to generalize effectively across seasonal variations in plant leaf disease detection. 
&nbsp;

Dual-encoder Variational Autoencoder for Detection and Classification of Plant Leaf Diseases

<div>Background: To investigate seroprevalence, clinical manifestations, molecular diagnosis of Lumpy Skin Disease in cattle in an around Parbhani during outbreak of LSD. 
 
Methods: The cattle irrespective of their age, breed and gender were screened for detection of LSDV. Duration of study period was from December, 2020 to August, 2021. The cattle showed high fever, presence of nodules on body, nasal discharge, ocular discharge, brisket oedema and lameness were screened for Lumpy Skin Disease and were further confirmed using molecular tests. Total 478 cattle were screened for the seroprevalence of Lumpy Skin Disease by ELISA and PCR tests. The cattle were categorized into different age groups as calves (less than 0-2 years), young (between 2-4 years) and adults (4 years and above). The various breeds such as Khillar, Red Kandhari, Gir, Holstein-Friesen, cross-bred and non-descript cattle were screened. The seroprevalence of Lumpy Skin Disease in female and male was also recorded. 
 
Result: The molecular assay of LSD revealed that, out of 478 cattle screened, 267 (55.85%) serum samples were found positive by ELISA test. Whereas, among 267 blood samples and 8 scab samples tested by PCR, 82 blood samples (30.7%) and 5 scab samples (62.5%) were found positive for LSDV in cattle. LSD is an emerging, transboundary and economically important viral disease of cattle. The overall seroprevalence of LSD in cattle was 55.85%. Prominent clinical manifestations in LSD affected cattle was Prominent clinical manifestations in LSD affected cattle were fever followed by generalized appearance of skin nodules on body. In addition, LSD affected cattle showed oculo-nasal discharge, enlarged lymph nodes, brisket oedema, lameness, corneal opacity and anorexia. The scab samples were more reliable material for LSDV confirmation by PCR.</div>

Molecular Detection and Seroprevalence of Lumpy Skin Disease in Cattle

Background: Automatic feature extraction using convolutional neural networks has proven to be useful for a variety of computer vision tasks. Disease detection in plants is one such task that can be performed using convolutional neural network. Precise and timely disease detection in plants is crucial for better crop yield. So, state of the art technologies like convolutional neural network can help in developing efficient applications for this purpose. 
 
Methods: Here, we have performed an empirical study of five convolutional neural network architectures namely, VGG19, ResNet152V2, InceptionV3, MobileNetV2 and DenseNet201 for detecting diseases in tea leaves. Tea leaves affected with gray blight, red spot, brown blight, algal spot and helopeltis disease were used for the study. We have employed transfer learning models to address the issue of requiring a large number of data samples for training a convolutional neural network. The models were ranked based on their performances. We also proposed an enhanced DenseNet201 (E-DenseNet201) model by integrating channel attention module with DenseNet201 and compared its performance with the convolutional neural network architectures used here. 
 
Result: DenseNet201 demonstrated the highest performance among the five models with precision and recall value of 95.97% and 95.49% respectively. Further improvements were observed in the performance of E-DenseNet201 with precision and recall value of 98.15% and 97.5%.

E-DenseNet201: An Enhanced Method for Detecting Diseases in Tea Leaves

<div style="text-align: justify;">Concanavalin A (Con A), a lectin interacts with carbohydrate moieties of viruses and provide stable and sensitive detection when used as a capture agent. Indirect ELISA methods need purified Newcastle disease virus (NDV) or recombinant antigens for adsorption, whereas use of Con A as capture agent will enable the use of non-purified and non-concentrated virus as antigen replacing costly and time-consuming virus purification step. Con A based sandwich ELISA with non-purified NDV whole virus antigen with single serum dilution format has been developed in this study. The optimum concentrations of the capture agent, Con A and non-purified antigen preparations were determined by checker-board titration. Briefly, microplates were coated with predetermined optimum concentration of ConA (0.5 mg/ml; 50&micro;g per well) and incubated for 18h at 4&deg;C. After washing, allantoic fluid with Newcastle disease virus (NDV) LaSota (HA titre, 210) at a constant predetermined dilution (1:1; 50&micro;l) was coated and incubated for 45 min at 37&deg;C, followed by blocking with 2 % bovine serum albumin for 45 min at 37&deg;C. The antigen coated plates were used in the detection of antibody titre against NDV in serum samples at single serum dilution of 1: 500. Then, wells were added with goat anti-chicken IgG horseradish peroxidase conjugate and incubated for 1h at 37&deg;C, followed by addition of TMB substrate and the plates were read spectrophotometrically at 450 nm. ELISA antibody titres were determined by standard serial dilution of positive sera and endpoints were calculated by a subtraction method. By using positive negative threshold curve (PNT), intercept and slope of the standard curve were calculated. Total of 271 random chicken serum samples were analyzed for antibodies against NDV by Haemagglutination inhibition assay (HI), indirect ELISA and compared with the Con A- S- ELISA developed in this study. The Con A-S-ELISA showed a high coefficient of correlation (r=0.85, n=271, P&lt;0.01) and an agreement of &ecirc;=0.99 with the commercially available Indirect-ELISA. The relative sensitivity and specificity were 98% and 85% respectively in comparison to HI test. Hence, the Con A-S-ELISA is a simple, easy and effective for monitoring serum antibody levels against NDV.</div>

Concanavalin A- Sandwich-ELISA for the detection of antibodies against Newcastle disease virus (NDV) in chicken

<div style="text-align: justify;">Remote sensing is a rapid, non-invasive and efficient technique which can acquire and analyze spectral properties of earth surfaces from various distances, ranging from satellites to ground-based platforms. This modern technology holds promise in agricultural crop production including crop protection. Variability in the reflectance spectra of plants resulting from occurrence of disease and pests, allows their identification using remote sensing data. Various spectroscopic and imaging techniques like visible, infrared, multiband and fluorescence spectroscopy, fluorescence imaging, multispectral and hyperspectral imaging, thermography, nuclear magnetic resonance spectroscopy etc. have been studied for the detection of plant diseases. Several of these techniques have great potential in phytopathometry. Remote sensing technologies will be extremely helpful to greatly spatialize diagnostic results and thereby rendering agriculture more sustainable and safe, avoiding expensive use of pesticides in crop protection.&nbsp;</div>

Remote sensing and its use in detection and monitoring plant diseases: A review

Background: Livestock farming is experiencing a digital transformation and is becoming more information-driven. However, this type of data is often kept in separate storage towers, making it incapable of practicing its potential to boost animal welfare. Lumpy Skin Disease (LSD) is a serious threat to the health of cattle worldwide and has caused financial problems for many cattle farming enterprises. It has been shown that combining machine learning (ML) and artificial intelligence (AI) with biosensor data and conventional visual inspections can improve the identification and diagnosis of this serious condition. 
&nbsp; 
Methods: This study&nbsp;presents an extremely precise livestock farming framework that combines data streams from a wide array of disciplines of dairy cattle to see if ever wider/vast data sources enhance the overall projections for diseases and if using the more complicated prediction models can reimburse for less diverse data to some extent. Using images from the farming landscape, this study highlights the utility of convolutional neural networks (CNNs) in the identification of the Lumpy Skin Disease Virus (LSDV) in animals. 
&nbsp; 
Result: An analysis of the relative weight given to individual factors in predicting accuracy shows that disease in dairy cattle results from the intricate interactions between many life domains/parameters, such as housing, nutrition and climate; that prediction performance is enhanced by incorporating a wider range of data sources; and that current information can be repurposed to produce useful information for vaccine development. The study highlights the potential of data-driven dairy interventions, focusing on artificial intelligence for disease prediction in cattle, to improve animal welfare and reduce the risk of disease. A Convolutional Neural Network (CNN) based model effectively classified skin conditions with an overall accuracy of 73.89% after 27 training epochs. This study demonstrates CNN&rsquo;s useful applications in the field of veterinary medicine by highlighting its potential for early detection of Lumpy Skin Disease (LSD).

Detection of Disease in Calves using Artificial Intelligence

Background: Tomatoes play a pivotal role in global agriculture fostering economic growth while also significantly enhancing food security. This paper introduces a data-based approach to improve the early detection of fungal infections in tomatoes. 
 
Methods: The study utilises a dataset comprised of 4,362 high-resolution images obtained from Mendeley, showing healthy tomato plants alongside those damaged by leaf mold, early blight and late blight diseases. Deep learning techniques have been applied to develop a Convolutional Neural Network (CNN) for disease classification based on these images. 
 
Result: The CNN model exhibits an overall accuracy rate of 90.83%, underscoring its efficacy in identifying fungal growth in tomato plants. The study proposes the use of machine learning in the detection and treatment of fungal diseases affecting tomatoes, which consequently leads to increased crop yield and quality preservation. It recommends the development of automated tools for farmers to detect and respond to disease outbreaks, thus improving their agricultural practices. 
&nbsp; 
&nbsp;