Plant Datasets Research Articles

Enhanced Deep Learning Architecture for Rapid and Accurate Tomato Plant Disease Diagnosis

Deep neural networks have demonstrated outstanding performances in agriculture production. Agriculture production is one of the most important sectors because it has a direct impact on the economy and social life of any society. Plant disease identification is a big challenge for agriculture production, for which we need a fast and accurate technique to identify plant disease. With the recent advancement in deep learning, we can develop a robust and accurate system. This research investigated the use of deep learning for accurate and fast tomato plant disease identification. In this research, we have used individual and merged datasets of tomato plants with 10 diseases (including healthy plants). The main aim of this work is to check the accuracy of the existing convolutional neural network models such as Visual Geometry Group, Residual Net, and DenseNet on tomato plant disease detection and then design a custom deep neural network model to give the best accuracy in case of the tomato plant. We have trained and tested our models with datasets containing over 18,000 and 25,000 images with 10 classes. We achieved over 99% accuracy with our custom model. This high accuracy was achieved with less training time and lower computational cost compared to other CNNs. This research demonstrates the potential of deep learning for efficient and accurate tomato plant disease detection, which can benefit farmers and contribute to improved agricultural production. The custom model’s efficient performance makes it promising for practical implementation in real-world agricultural settings.

Plant-LncPipe: a computational pipeline providing significant improvement in plant lncRNA identification.

Long non-coding RNAs (lncRNAs) play essential roles in various biological processes, such as chromatin remodeling, post-transcriptional regulation, and epigenetic modifications. Despite their critical functions in regulating plant growth, root development, and seed dormancy, the identification of plant lncRNAs remains a challenge due to the scarcity of specific and extensively tested identification methods. Most mainstream machine learning-based methods used for plant lncRNA identification were initially developed using human or other animal datasets, and their accuracy and effectiveness in predicting plant lncRNAs have not been fully evaluated or exploited. To overcome this limitation, we retrained several models, including CPAT, PLEK, and LncFinder, using plant datasets and compared their performance with mainstream lncRNA prediction tools such as CPC2, CNCI, RNAplonc, and LncADeep. Retraining these models significantly improved their performance, and two of the retrained models, LncFinder-plant and CPAT-plant, alongside their ensemble, emerged as the most suitable tools for plant lncRNA identification. This underscores the importance of model retraining in tackling the challenges associated with plant lncRNA identification. Finally, we developed a pipeline (Plant-LncPipe) that incorporates an ensemble of the two best-performing models and covers the entire data analysis process, including reads mapping, transcript assembly, lncRNA identification, classification, and origin, for the efficient identification of lncRNAs in plants. The pipeline, Plant-LncPipe, is available at: https://github.com/xuechantian/Plant-LncRNA-pipline.

The impact of imputation methods on the performance of Phase I Hotelling’s T 2 control chart

The objective of this study was to evaluate the impact of three different methods of handling missing data on the performance of Phase I Hotelling’s T 2 multivariate control chart. Using a Monte Carlo simulation, we studied the average, median, and standard deviation of the run length performance of multivariate data imputed using mean substitution, regression imputation, and predictive mean matching at three different levels of missingness ( 1 % , 10 % , and 25 % ) and three levels of variable correlation coefficients (0.2, 0.4, and 0.8). We found that predictive mean matching has average run length performance results comparable to that of the complete in-control data set at all levels of missingness and variable correlation, while the performance of mean substitution was adversely affected by high levels of missingness and by strong variable correlation. Based on the simulation (multivariate normal data), we concluded that predictive mean matching is superior to both regression imputation and mean substitution as a method for imputing missing values for the analysis of Phase I Hotelling’s T 2 control chart. Two applications were presented using the Altenrhein wastewater treatment plant and Olive oil datasets.

Instance transfer partial least squares for semi-supervised adaptive soft sensor

In order to deal with the unknown data drift problem caused by the change of working conditions in the process of industrial soft sensor modeling, this paper proposes a semi-supervised soft sensor modeling algorithm based on instance transfer learning, named instance transfer partial least squares. Under the framework of nonlinear partial least squares, the algorithm introduces the instance transfer inner model, and realizes the distribution adaptation of the modeling data and the data to be predicted through sample importance weighting. First, a nonlinear partial least squares model is constructed by using labeled samples in the source domain, which consists of a linear outer model and a nonlinear inner model. The linear outer model maps the original data to the hidden variable space, and the nonlinear inner model realizes the nonlinear modeling of hidden variables through kernel ridge regression. Second, a small number of labeled samples from the target domain are used to update the model weights of the nonlinear inner model, enabling adaptation of the latent variable distribution. The effectiveness of the algorithm is verified on a toy dataset, a simulation dataset and a real pulverized coal power plant dataset.

Monitoring global cement plants from space

As an energy-intensive industry, cement production can emit significant air pollutants that negatively impact human health and the environment. Geospatial data on cement plants are crucial for managing the cement industry and designing effective policies to mitigate ever-increasing air pollution. However, the available fundamental asset-level datasets of global cement plants are generally incomplete and may not adequately satisfy the above requirements. To address this gap, (i) we proposed an innovative semi-automatic framework, which is based on the high-temperature signal from active kilns in time-series Landsat-8 Operational Land Imager (OLI) and Sentinel-2 Multispectral Instrument (MSI) imagery and the iconic characteristics of cement processing equipment in high-resolution imagery, to identify global cement plants, assess their operational status, and evaluate relevant activity levels. (ii) We compiled a state-of-the-art inventory of global cement plants, enriched with detailed attributes, using multi-source spaceborne observations/products. We identified 3075 cement plants with rotary/shaft kilns and pinpointed 1960 locations for cement grinding and storage facilities. (iii) We investigated the status of global cement plants using time-series Sentinel-2 MSI imagery. Our results show that from 2016 to date, 69.3% of cement plants are continuously operating, 4.3% were newly constructed, and 26.4% were decommissioned. (iv) We proposed two normalized metrics, namely the occurrence rate and the mean detected area, to assess the activity level of cement plants. The results showed that the rotary kilns in wet-process cement plants had the highest activity level due to the higher energy consumption. The activity level of global cement plants increased by ∼9% during 2016–2021 and then decreased by ∼4% in 2022, possibly due to the influence of COVID-19 and weak market demand. The proposed framework provides valuable insights for investigating the locations and dynamics of other industrial hot sources, such as those in steel mills, coke ovens, refineries, and oil/gas development areas, to improve the estimates of CO2 emissions and air pollutants and then to promote climate action.

Modeling Large-Scale Manufacturing of Lithium-Ion Battery Cells: Impact of New Technologies on Production Economics

The global demand for electric vehicles is increasing exponentially, as is the demand for lithium-ion battery cells. This has led to a strong ongoing competition among companies to achieve the lowest battery cell production cost. Herein, to provide guidance on the identification of the best starting points to reduce production costs, a bottom-up cost calculation technique, process-based cost modeling (PBCM), for battery cell production is reproduced and validated by drawing on a consistent dataset of a real battery cell production plant. The model is based on teardowns of a real battery cell factory and will prove useful for planning activities of today's, so-called, “giga factories.” The PBCM performed in the present study involves discussions on, e.g., production balancing, relocation of factories to low-wage countries, usage of new production and cell technologies, etc. The use of novel approaches, such as tabless cell design, dry coating, and NMC811 chemistry, is discussed. Finally, the ways in which battery cell production costs can be reduced further in the forthcoming years are shown, and implications for researchers, practitioners, and policy makers are provided.

PlantBiCNet: A new paradigm in plant science with bi-directional cascade neural network for detection and counting

Deep learning is increasingly popular for precise plant detection and counting in computer vision applications. Despite the rising cross-disciplinary research in this field, an important issue has been neglected: due to the inherent constraints imposed by the natural growth patterns of plants, the costly and time-consuming process of collecting plant image data, resulting in a scarcity of real-world available data. Consequently, researchers are often constrained to exploring large-scale data representations or limited applications. Currently, still no generalized network architecture excels in demanding environments. To address this, we introduce two new bi-directional cascade neural network paradigms, PlantBiCNet and its lightweight variant, PlantBiCNet-Lite. Unlike previous research, we place greater emphasis on the network decoding process itself rather than the encoder. Our bi-directional cascade and weight fusion decoding approach maximizes high-level semantics and low-level spatial data, enhancing the localization signal in a way that respects gradient diversity. We evaluate PlantBiCNet and PlantBiCNet-Lite on five public plant detection and counting datasets and a novel plant dataset. The average precision AP50 on these six small-scale datasets reached 0.905, with a high coefficient of determination R2 of 0.935, while the reported frame rate was 196 FPS. These results significantly outperform state-of-the-art computer vision methods. Moreover, we found that the introduction of the Dropout mechanism markedly enhances the model's performance and generalization ability. Further analysis supports our experiments and provides valuable insights into network optimization. Overall, our organization's open datasets and network modeling details provide a solid foundation for future research in the field of plant science.

Measuring Carbon Dioxide Emissions From Liquefied Natural Gas (LNG) Terminals With Imaging Spectroscopy

AbstractThe rapid growth of liquefied natural gas (LNG) exports underscores the importance of CO2 monitoring for LNG export terminals. We present a method for measuring LNG terminal CO2 emissions using remote sensing imaging spectroscopy. The method is first validated using 47 power plant emission events with in situ measured data, then applied to 22 emission events in Sabine Pass and Cameron. The power plant data set shows a robust correlation between our estimates and in situ data, with R2 0.9146 and the average error −2%. At Sabine Pass, eight point sources are identified with emission rates from 219.69 ± 54.95 to 1,083.22 ± 308.06 t/hr. At Cameron, three point sources are identified with emission rates from 91.64 ± 25.81 to 265.61 ± 67.80 t/hr. The liquefaction carbon intensity estimates also align with past study ranges. This illustrates that remote sensing can validate environmental reporting and CO2 inventories for industrial facilities.

An Effective Ensemble Convolutional Learning Model with Fine-Tuning for Medicinal Plant Leaf Identification

Accurate and efficient medicinal plant image classification is of utmost importance as these plants produce a wide variety of bioactive compounds that offer therapeutic benefits. With a long history of medicinal plant usage, different parts of plants, such as flowers, leaves, and roots, have been recognized for their medicinal properties and are used for plant identification. However, leaf images are extensively used due to their convenient accessibility and are a major source of information. In recent years, transfer learning and fine-tuning, which use pre-trained deep convolutional networks to extract pertinent features, have emerged as an extremely effective approach for image-identification problems. This study leveraged the power by three-component deep convolutional neural networks, namely VGG16, VGG19, and DenseNet201, to derive features from the input images of the medicinal plant dataset, containing leaf images of 30 classes. The models were compared and ensembled to make four hybrid models to enhance the predictive performance by utilizing the averaging and weighted averaging strategies. Quantitative experiments were carried out to evaluate the models on the Mendeley Medicinal Leaf Dataset. The resultant ensemble of VGG19+DensNet201 with fine-tuning showcased an enhanced capability in identifying medicinal plant images with an improvement of 7.43% and 5.8% compared with VGG19 and VGG16. Furthermore, VGG19+DensNet201 can outperform its standalone counterparts by achieving an accuracy of 99.12% on the test set. A thorough assessment with metrics such as accuracy, recall, precision, and the F1-score firmly established the effectiveness of the ensemble strategy.

Leveraging three-tier deep learning model for environmental cleaner plants production

The world's population is expected to exceed 9 billion people by 2050, necessitating a 70% increase in agricultural output and food production to meet the demand. Due to resource shortages, climate change, the COVID-19 pandemic, and highly harsh socioeconomic predictions, such a demand is challenging to complete without using computation and forecasting methods. Machine learning has grown with big data and high-performance computers technologies to open up new data-intensive scientific opportunities in the multidisciplinary agri-technology area. Throughout the plant's developmental period, diseases and pests are natural disasters, from seed production to seedling growth. This paper introduces an early diagnosis framework for plant diseases based on fog computing and edge environment by IoT sensors measurements and communication technologies. The effectiveness of employing pre-trained CNN architectures as feature extractors in identifying plant illnesses has been studied. As feature extractors, standard pre-trained CNN models, AlexNet are employed. The obtained in-depth features are eliminated by proposing a revised version of the grey wolf optimization (GWO) algorithm that approved its efficiency through experiments. The features subset selected were used to train the SVM classifier. Ten datasets for different plants are utilized to assess the proposed model. According to the findings, the proposed model achieved better outcomes for all used datasets. As an average for all datasets, the accuracy of the proposed model is 93.84 compared to 85.49, 87.89, 87.04 for AlexNet, GoogleNet, and the SVM, respectively.

Intelligent fault diagnosis of hydroelectric units based on radar maps and improved GoogleNet by depthwise separate convolution

Abstract Fault diagnosis plays an essential role in maintaining the safe and stable operation of hydroelectric units. In this paper, an intelligent fault diagnosis method based on radar maps and improved GoogleNet by depthwise separate convolution (DSC) is proposed to address the problems of low recognition accuracy and weak computing speed of fault diagnosis models in the field of hydroelectric unit fault diagnosis at present. Firstly, a one-dimensional signal sequence is obtained and denoised. Secondly, five time-domain features are extracted and radar maps are plotted. Then, an improved GoogleNet intelligent fault diagnosis model based on DSC (DSC-GoogLeNet) is constructed for training and validation. To assess the effectiveness of the proposed model, two case studies are conducted using the simulated dataset of the rotor experimental bench and the actual measured dataset of a domestic hydroelectric power plant. The results demonstrate that the average recognition accuracy of the fault diagnosis method proposed in this paper is as high as 99.04% on the simulated dataset, and even though the recognition accuracy decreases on the actually measurement dataset, it still has a recognition rate of 98.79%. The fault diagnosis performance is better than the other types of comparison models. The results demonstrate that the proposed fault diagnosis method holds significant engineering applicability in the domain of safe operation of hydroelectric units. It effectively addresses the existing challenges in fault diagnosis within this field with accuracy, stability, and efficiency.

The structure of plant–pollinator networks is affected by crop type in a highly intensive agricultural landscape

In agricultural landscapes, bipartite networks formed by pollinators and the flowers they forage on, are characterised by the presence of honeybees (Apis mellifera) or crop plants. These managed species can affect the structural properties of these networks because of spatial and temporal variation in the availability of resources, and competition for these resources; for example, crop plants such as oilseed rape and sunflower produce a large number of flowers during a short blooming period. Here, we examined the structure of plant–pollinator networks in an intensive agricultural landscape, the Zone Atelier Plaine & Val de Sèvre (West of France). We compiled a six-year monitoring dataset of plant–pollinator interactions, sampling by sweep-nets along transects in the main six crop types (811 fields in total). To describe networks, we computed six metrics: connectance, nodes number, modularity, specialisation and a nestedness measure. We found that the degree distribution (number of partners per species in the network) fitted better to Fisher’s log-series rather than Power law. Strong differences between crops were observed in nestedness and specialisation, even when accounting for the effect of sample size: in oilseed rape, networks were much more nested and much less specialised than in sunflower. In addition, the link ‘honeybee–crop flower’ differed between the two mass-flowering crops. Honeybees and sunflowers appeared as specialist species in sunflower crops, interacting strongly and almost exclusively with each other, whereas they are usually considered highly generalist species. Indeed, sunflower pollination may be almost entirely driven by honeybees, conversely to oilseed rape crops, where the presence of wild bees and other insects tended to produce a more diversified network.

Forests over limestone in Guiuan, Eastern Samar, the Philippines: Species composition and influencing environmental factors

Abstract. Buot Jr IE, Villanueva ELC, Origenes MG, Obeña RDR. 2023. Forests over limestone in Guiuan, Eastern Samar, the Philippines: Species composition and influencing environmental factors. Biodiversitas 24: 5123-5136. Forests over limestone are critical ecosystems providing numerous ecosystem services. However, the ecology of these ecosystems is less studied like the Guiuan Marine Resource Protected Landscape and Seascape (GMRPLS) in Eastern Samar, Philippines which incidentally is affected by strong typhoons. Studies on forests over limestone are even more essential to understand the vegetation ecology providing insights in crafting appropriate management strategy to sustain ecosystem services. Hence, the study aims to identify the woody vegetation diversity and composition of forests over limestone in GMRPLS, and to analyze the environmental factors influencing the physiognomy and ecology of the forests. Standard vegetation techniques were used and a total of nine plots with size of 20 × 20m each were established to assess the tree species (?1m height). Plant abundance data were recorded and computed. Soil samples were collected while climate data were obtained from the Guiuan, Eastern Samar Weather Station. Cluster analysis using relative density of trees per plot was done in order to identify major plant communities. Subsequently, a canonical correspondence analysis was carried out to pinpoint which environmental variables influenced the plant data set. Results show a total of 37 species belonging to 17 families, with 18 species are listed under by IUCN and seven species are listed by DAO 2017-11. Cluster analysis indicated 3 vegetation groupings or clusters named after the dominant species per plot, namely Cluster I (Hancea-Calophyllum-Macaranga-Ficus-Gnetum), Cluster II (Monoon-Wallaceodendron-Artocarpus-Hancea-Bago-eho), and Cluster III (Bridelia-Monoon-Neonauclea-Gnetum-Artocarpus-Sterculia-Hancea). GUI 02 plot (Cluster 1) was observed as the most diverse with the highest Simpson’s index and Fisher’s alpha, as well as the highest in total number of tree species. Canonical correspondence analysis indicated elevation, temperature and anthropogenic disturbances as the major factors influencing the vegetation diversity and composition. This can be used to formulate forest management strategies for the sustainable conservation of forests over limestone in GMRPLS.

Prediction of plant LncRNA-protein interactions based on feature fusion and an improved residual network

LncRNA(long non-coding RNA)-protein interaction (LPI) has effects on chromosome structure and gene transcription, participating in key cellular processes such as signal transduction, chromosome replication, material transport, and mitosis. Accurate identification of LPIs will provide scientific basis for understanding the molecular mechanisms of LncRNA-related diseases, thereby promoting the progress of disease diagnosis technologies and the development of therapeutic procedures to a certain extent. How to comprehensively mine feature information reflecting functional attributes from LncRNAs and proteins, apply deep learning to extract advanced features from the original input features of proteins and LncRNAs, and effectively fuse them have been important research challenges in this field. Aiming at the limitations of existing methods, based on deep learning techniques such as bi-directional long short-term memory (BiLSTM), attention mechanism, and an improved residual network, an LPI prediction model called LPI-LSTM-ResNet is constructed in this paper. Firstly, the sequence and structural information of LncRNAs and proteins are extracted from different perspectives. Then, the deep interaction fused features between LncRNAs and proteins are obtained by BiLSTM and attention mechanism. Finally, the fused features are input into an improved residual network with LSTM as the residual element to preserve long-distance dependencies between sequences. The 5-fold cross-validation results indicate that the feature combination strategy, feature fusion strategy, and the improved residual network consistently improve the LPI prediction performance. Compared with existing methods on the same plant datasets, LPI-LSTM-ResNet exhibits superior performance.

Rice Leaf Disease Detection Using Convolutional Neural Network

In the agriculture sector, the rice crops getting diseased has become a significant concern recently, especially in India, where rice is one of the primary meals. Precise and early-stage detection of various diseases observed in the rice crops can help farmers to provide proper treatment to the crops. This paper presents a Convolutional Neural Network (CNN) based approach is used to detect rice plant leaf disease. CNN is one of the deep learning algorithms that help in image processing and classification with significant accuracy. The proposed algorithm is used for an image dataset of the diseased rice plant leaves, available on Kaggle. Two types of rice leaf diseases are considered for the analysis: brown spot and bacterial leaf blight. The images of these two diseases were pre-processed, segmented, and classified to identify the caused disease. The proposed model can also be used for the detection of the diseases present in other types of crops, faces recognition system, classifying animals, and car models. The overall accuracy of the developed model is nearly 67%.

Functional diversity metrics can perform well with highly incomplete data sets

Abstract Characterising changes in functional diversity at large spatial scales provides insight into the impact of human activity on ecosystem structure and function. However, the approach is often based on trait data sets that are incomplete and unrepresentative, with uncertain impacts on functional diversity estimates. To address this knowledge gap, we simulated random and biased removal of data from three empirical trait data sets: an avian data set (9579 species), a plant data set (2185 species) and a crocodilian data set (25 species). For these data sets, we assessed whether functional diversity metrics were robust to data incompleteness with and without using imputation to fill data gaps. We compared two metrics each calculated with two methods: functional richness (calculated with convex hulls and trait probabilities densities) and functional divergence (calculated with distance‐based Rao and trait probability densities). Without imputation, estimates of functional diversity (richness and divergence) for birds and plants were robust when 20%–70% of species had missing data for four out of 11 and two out of six continuous traits, respectively, depending on the severity of bias and method used. However, when missing traits were imputed, functional diversity metrics consistently remained representative of the true value when 70% of bird species were missing data for four out of 11 traits and when 50% of plant species were missing data for two out of six traits. Trait probability densities and distance‐based Rao were particularly robust to missingness and bias when combined with imputation. Convex hull‐based estimations of functional richness were less reliable. When applied to a smaller data set (crocodilians, 25 species), all functional diversity metrics were much more sensitive to missing data. Expanding global morphometric data sets to represent more taxa and traits, and to quantify intraspecific variation, remains a priority. In the meantime, our results show that widely used methods can successfully quantify large‐scale functional diversity even when data are missing for half of species, provided that missing traits are estimated using imputation. We recommend the use of trait probability densities or distance‐based Rao when working with large incomplete data sets and filling data gaps with imputation.

Analyzing the occurrence of foaming in batch fermentation processes using multiway partial least square approaches

AbstractThis article presents an application of multiway partial least squares (MPLS) methods to develop interpretative correlation models to monitor the foaming occurrence and improve batch fermentation. We choose the exhaust differential pressure as a quality variable to quantify the foaming occurrence and consider three‐dimensional datasets of different batches, process variables, and measurements. We integrate batch‐wise unfolding (BWU) and observation‐wise unfolding (OWU) of plant datasets with standard, dynamic, and kernel PLS methods. We find that dynamic PLS (DPLS) with OWU and time‐lagged quality variables to be the most efficient, accurate, and easy to implement. The BWU approach is useful for analyzing the differences between batches and identifying abnormalities and outliers, while the OWU quantifies the variation within a given batch. With OWU, the DPLS method with one unit of time lag in the quality variable is the most effective, accurate, and easy to implement. With both BWU and OWU, we identify the quantitative effects of process variables on the quality variable and providence guidance to improve fermentation performance.

Coarse-grained privileged learning for classification

Privileged information, a form of prior knowledge, can significantly enhance traditional machine learning performance through a novel paradigm known as learning using privileged information (LUPI). Although effective, current studies on LUPI require a distinct piece of privileged information per input, and these fine-grained priors are difficult to collect in practice. To this end, this paper proposes a brand new problem of learning with class-wise privileged information, where instances within the same class share identical privileged information. As far as we know, this problem has not yet been explored. We build a support vector machine with coarse-grained class-wise priors (CGSVM+) and put forward a novel and reliable augmenting strategy to solve it. In addition, two datasets are collected from nature reserves in Xinjiang, China, along with their class-wise privileged information annotated by professionals. Extensive experiments demonstrate the effectiveness of CGSVM+, with the best average accuracy of 80.16% (94.70%) and the best average F-score of 79.87% (94.57%) on the plant (animal) datasets.

Plant and Disease Recognition Based on PMF Pipeline Domain Adaptation Method: Using Bark Images as Meta-Dataset.

Efficient image recognition is important in crop and forest management. However, it faces many challenges, such as the large number of plant species and diseases, the variability of plant appearance, and the scarcity of labeled data for training. To address this issue, we modified a SOTA Cross-Domain Few-shot Learning (CDFSL) method based on prototypical networks and attention mechanisms. We employed attention mechanisms to perform feature extraction and prototype generation by focusing on the most relevant parts of the images, then used prototypical networks to learn the prototype of each category and classify new instances. Finally, we demonstrated the effectiveness of the modified CDFSL method on several plant and disease recognition datasets. The results showed that the modified pipeline was able to recognize several cross-domain datasets using generic representations, and achieved up to 96.95% and 94.07% classification accuracy on datasets with the same and different domains, respectively. In addition, we visualized the experimental results, demonstrating the model's stable transfer capability between datasets and the model's high visual correlation with plant and disease biological characteristics. Moreover, by extending the classes of different semantics within the training dataset, our model can be generalized to other domains, which implies broad applicability.

A dual track deep fusion network for citrus disease classification using group shuffle depthwise feature pyramid and Swin transformer

Citrus fruits are globally significant crops, valued for their unique taste, nutritional value, and versatility in culinary applications. However, these crops are vulnerable to diseases that can substantially reduce their yield and quality, leading to significant financial losses. It is crucial to efficiently identify and manage diseases affecting citrus fruits to ensure food security and support sustainable agriculture. Traditionally, identifying symptoms associated with these diseases requires expert scientific and observational skills. With the advent of deep learning methods, it has become possible to automatically identify disease patterns from images of affected plants. In this research, a novel dual branch deep-learning network was proposed for the classification of citrus plant diseases. The Group Shuffle Depthwise Feature Pyramid (GSDFP), which constitutes the first branch, utilizes convolution blocks to extract local features. The second branch leverages the Swin transformer to integrate global contextual awareness into the extracted features and facilitates the learning process through long-term dependencies. Subsequently, the features from the two branches are fused and passed through a shuffle attention module, effectively capturing contextual relationships between them. The effectiveness of the proposed approach was validated on two benchmark datasets, namely the Citrus Plant Dataset and the Citrus Disease Image Gallery Dataset. The proposed network obtained a classification accuracy of 98.19%.