Search Results (14)

Timely detection and prevention of citrus leaf diseases and pests are crucial for improving citrus yield. To address the issue of low efficiency in citrus disease and pest detection, this paper proposes a lightweight detection model named PEW-YOLO. First, the PP-LCNet backbone is optimized using a novel GSConv convolution, and a lightweight PGNet backbone is introduced to reduce model parameters while enhancing detection performance. Next, the C2f_EMA module, which integrates efficient multi-scale attention (EMA), replaces the original C2f module in the neck, thereby improving feature fusion capabilities. Finally, the Wise-IoU loss function is employed to address the challenge of identifying low-quality samples, further improving both convergence speed and detection accuracy. Experimental results demonstrate that PEW-YOLO achieves a 1.8% increase in mAP50, a 32.2% reduction in parameters, and a detection speed of 1.6 milliseconds per frame on the citrus disease and pest dataset, thereby meeting practical real-time detection requirements. Full article

In agriculture, machine learning (ML) and deep learning (DL) have increased significantly in the last few years. The use of ML and DL for image classification in plant disease has generated significant interest due to their cost, automatization, scalability, and early detection. However, high-quality image datasets are required to train robust classifier models for plant disease detection. In this work, we have created an image dataset of 649 orange leaves divided into two groups: control (n = 379) and huanglongbing (HLB) disease (n = 270). The images were acquired with several smartphone cameras of high resolution and processed to remove the background. The dataset enriches the information on characteristics and symptoms of citrus leaves with HLB and healthy leaves. This enhancement makes the dataset potentially valuable for disease identification through leaf segmentation and abnormality detection, particularly when applying ML and DL models. Full article

The citrus industry plays a pivotal role in modern agriculture. With the expansion of citrus plantations, the intelligent detection and prevention of diseases and pests have become essential for advancing smart agriculture. Traditional citrus leaf disease identification methods primarily rely on manual observation, which is often time-consuming, labor-intensive, and prone to inaccuracies due to inherent asymmetries in disease manifestations. This work introduces CBACA-YOLOv5, an enhanced YOLOv5s-based detection algorithm designed to effectively capture the symmetric and asymmetric features of common citrus leaf diseases. Specifically, the model integrates the convolutional block attention module (CBAM), which symmetrically enhances feature extraction across spatial and channel dimensions, significantly improving the detection of small and occluded targets. Additionally, we incorporate coordinate attention (CA) mechanisms into the YOLOv5s C3 module, explicitly addressing asymmetrical spatial distributions of disease features. The CARAFE upsampling module further optimizes feature fusion symmetry, enhancing the extraction efficiency and accelerating the network convergence. Experimental findings demonstrate that CBACA-YOLOv5 achieves an accuracy of 96.1% and a mean average precision (mAP) of 92.1%, and improvements of 0.6% and 2.3%, respectively, over the baseline model. The proposed CBACA-YOLOv5 model exhibits considerable robustness and reliability in detecting citrus leaf diseases under diverse and asymmetrical field conditions, thus holding substantial promise for practical integration into intelligent agricultural systems. Full article

Accurate citrus disease identification is essential for targeted orchard pesticide application. Current models struggle with accuracy and efficiency due to diverse leaf lesion patterns and complex orchard environments. This study presents YOLOv8n-DE, an improved lightweight YOLOv8-based model for enhanced citrus disease detection. It introduces the DR module structure for effective feature enhancement and the Detect_Shared architecture for parameter efficiency. Evaluated on public and orchard-collected datasets, YOLOv8n-DE achieves 97.6% classification accuracy, 91.8% recall, and 97.3% mAP, with a 90.4% mAP for challenging diseases. Compared to the original YOLOv8, it reduces parameters by 48.17%, computational load by 59.26%, and model size by 41.94%, while significantly decreasing classification and regression errors, and false positives/negatives. YOLOv8n-DE offers outstanding performance and lightweight advantages for citrus disease detection, supporting precision agriculture development in orchards. Full article

Background: Given the severe economic burden that citrus greening disease imposes on fruit farmers and related industries, rapid and accurate disease detection is particularly crucial. This not only effectively curbs the spread of the disease, but also significantly reduces reliance on manual detection within extensive citrus planting areas. Objective: In response to this challenge, and to address the issues posed by resource-constrained platforms and complex backgrounds, this paper designs and proposes a novel method for the recognition and localization of citrus greening disease, named the HHS-RT-DETR model. The goal of this model is to achieve precise detection and localization of the disease while maintaining efficiency. Methods: Based on the RT-DETR-r18 model, the following improvements are made: the HS-FPN (high-level screening-feature pyramid network) is used to improve the feature fusion and feature selection part of the RT-DETR model, and the filtered feature information is merged with the high-level features by filtering out the low-level features, so as to enhance the feature selection ability and multi-level feature fusion ability of the model. In the feature fusion and feature selection sections, the HWD (hybrid wavelet-directional filter banks) downsampling operator is introduced to prevent the loss of effective information in the channel and reduce the computational complexity of the model. Through using the ShapeIoU loss function to enable the model to focus on the shape and scale of the bounding box itself, the prediction of the bounding box of the model will be more accurate. Conclusions and Results: This study has successfully developed an improved HHS-RT-DETR model which exhibits efficiency and accuracy on resource-constrained platforms and offers significant advantages for the automatic detection of citrus greening disease. Experimental results show that the improved model, when compared to the RT-DETR-r18 baseline model, has achieved significant improvements in several key performance metrics: the precision increased by 7.9%, the frame rate increased by 4 frames per second (f/s), the recall rose by 9.9%, and the average accuracy also increased by 7.5%, while the number of model parameters reduced by ${0.137\times 10}^7$. Moreover, the improved model has demonstrated outstanding robustness in detecting occluded leaves within complex backgrounds. This provides strong technical support for the early detection and timely control of citrus greening disease. Additionally, the improved model has showcased advanced detection capabilities on the PASCAL VOC dataset. Discussions: Future research plans include expanding the dataset to encompass a broader range of citrus species and different stages of citrus greening disease. In addition, the plans involve incorporating leaf images under various lighting conditions and different weather scenarios to enhance the model’s generalization capabilities, ensuring the accurate localization and identification of citrus greening disease in diverse complex environments. Lastly, the integration of the improved model into an unmanned aerial vehicle (UAV) system is envisioned to enable the real-time, regional-level precise localization of citrus greening disease. Full article

Aphids affect citrus by causing leaf deformations and reducing fruit production. Additionally, aphids are a great concern due to their ability to transmit Citrus tristeza virus (CTV), the cause of tristeza, one of the main citrus diseases. In the last four years, citrus orchards in the south of Portugal (Algarve region) were sampled for aphid species identification and counting. Aphis spiraecola was the most abundant species, representing more than 80% of all identified aphids, and the damage (leaf deformation) it causes was directly proportional to its density. A. gossypii was the second most common species, followed by A. aurantii and Macrosiphum euphorbiae. The number of aphids in nymph stages was predominant over the adult stages (both wingless and winged) in all species. A. citricidus, the most efficient CTV vector, was not detected. The largest populations of A. spiraecola were observed in lemon and orange trees during spring (>100 individuals per shoot), with great damage observed in orange, lemon, and mandarin trees. A. gossypii was observed mainly in mandarin and tangor trees. There was a low activity of natural biological control agents, with the parasitism of A. spiraecola by Lysiphlebus spp. and Binodoxys spp. ranging from 0.3 to 1.5%. The numerical ratio ranged from 150 to 440 aphids per predator, and among these, syrphids were the most abundant, followed by lacewings and coccinellids (Scymnus). Full article

The lemon industry in the Mediterranean basin is strongly threatened by “mal secco” disease (MSD) caused by the fungus Plenodomus tracheiphlilus. Leaf pretreatments with Pseudomonas mediterranea 3C have been proposed as innovative tools for eco-sustainable interventions aimed at controlling the disease. In this study, by exploiting the results of previously performed RNAseq analysis, WCGNA was conducted among gene expression patterns in both inoculated (Pt) and pretreated and fungus-inoculated lemon plants (Citrus limon L.) (3CPt), and two indicators of fungal infection, i.e., the amount of fungus DNA measured in planta and the disease index (DI). The aims of this work were (a) to identify gene modules significantly associated with those traits, (b) to construct co-expression networks related to mal secco disease; (c) to define the effect and action mechanisms of P. mediterranea by comparing the networks. The results led to the identification of nine hub genes in the networks, with three of them belonging to receptor-like kinases (RLK), such as HERK1, CLAVATA1 and LRR, which play crucial roles in plant–pathogen interaction. Moreover, the comparison between networks indicated that the expression of those receptors is not induced in the presence of P. mediterranea, suggesting how powerful WCGNA is in discovering crucial genes that must undergo further investigation and be eventually knocked out. Full article

Citrus canker (CC), caused by one of the most destructive subfamilies of the bacterial phytopathogen Xanthomonas citri subsp. Citri (Xcc), poses a serious threat to the significantly important citrus fruit crop grown worldwide. This has been the subject of ongoing epidemiological and disease management research. Currently, five different forms have been identified of CC, in which Canker A (Xanthomonas citri subsp. citri) being the most harmful and infecting the majority of citrus cultivars. Severe infection symptoms include leaf loss, premature fruit drop, dieback, severe fruit blemishing or discoloration, and a decrease in fruit quality. The infection spreads rapidly through wind, rain splash, and warm and humid climates. The study of the chromosomal and plasmid DNA of bacterium has revealed the evolutionary pattern among the pathovars, and research on the Xcc genome has advanced our understanding of how the bacteria specifically recognize and infect plants, spread within the host, and propagates itself. Quarantine or exclusion programs, which prohibit the introduction of infected citrus plant material into existing stock, are still in use. Other measures include eliminating sources of inoculum, using resistant hosts, applying copper spray for protection, and implementing windbreak systems. The main focus of this study is to highlight the most recent developments in the fields of Xcc pathogenesis, epidemiology, symptoms, detection and identification, host range, spread, susceptibility, and management. Additionally, it presents an analysis of the economic impact of this disease on the citrus industry and suggests strategies to reduce its spread, including the need for international collaboration and research to reduce the impact of this disease on the global citrus industry. Full article

The rapid and accurate identification of citrus leaf diseases is crucial for the sustainable development of the citrus industry. Because citrus leaf disease samples are small, unevenly distributed, and difficult to collect, we redesigned the generator structure of FastGAN and added small batch standard deviations to the discriminator to produce an enhanced model called FastGAN2, which was used for generating citrus disease and nutritional deficiency (zinc and magnesium deficiency) images. The performance of the existing model degrades significantly when the training and test data exhibit large differences in appearance or originate from different regions. To solve this problem, we propose an EfficientNet-B5 network incorporating adaptive angular margin (Arcface) loss with the adversarial weight perturbation mechanism, and we call it EfficientNet-B5-pro. The FastGAN2 network can be trained using only 50 images. The Fréchet Inception Distance (FID) and Kernel Inception Distance (KID) are improved by 31.8% and 59.86%, respectively, compared to the original FastGAN network; 8000 images were generated using the FastGAN2 network (2000 black star disease, 2000 canker disease, 2000 healthy, 2000 deficiency). Only images generated by the FastGAN2 network were used as the training set to train the ten classification networks. Real images, which were not used to train the FastGAN2 network, were used as the test set. The average accuracy rates of the ten classification networks exceeded 93%. The accuracy, precision, recall, and F1 scores achieved by EfficientNet-B5-pro were 97.04%, 97.32%, 96.96%, and 97.09%, respectively, and they were 2.26%, 1.19%, 1.98%, and 1.86% higher than those of EfficientNet-B5, respectively. The classification network model can be successfully trained using only the images generated by FastGAN2, and EfficientNet-B5-pro has good generalization and robustness. The method used in this study can be an effective tool for citrus disease and nutritional deficiency image classification using a small number of samples. Full article

Citrus canker is a quarantined disease caused by the bacterial plant pathogen Xanthomonas citri subsp. citri (Xcc), which causes persistent surface damage, leaf and fruit drop, and tree decline in citrus plants. The citrus cultivar Citron C-05 (Citrus medica L.) is a disease-resistant genotype identified after years of screening at the National Center for Citrus Improvement (Changsha), which displays allergic, necrotic, and disease-resistant responses to Xcc. In this study, the BAK1 gene was identified in this cultivar to be a disease resistance gene involved in plant-microbe interaction between citrus and Xcc. Functional investigations of this gene revealed that both CsBAK1 (C. sinensis BAK1) or CmBAK1(C. medica BAK1) could inhibit the growth of Xcc to some extent when transiently expressed in the susceptible ‘Bingtang’ genotype of sweet orange. Critical regions of the CmBAK1 promoter sequence were identified by creating downstream deletions and exposing mutants to Xcc to determine effects on the resistance phenotype; a 426 bp region (−2000~–1574) was identified as a key functional region responsible for eliciting the hypersensitive response in plants. Through screening arrayed Citron C-05 cDNA libraries by yeast one-hybrid assays, a basic APETALA2/ETHYLENE RESPONSE FACTOR (AP2/ERF) transcription factor of CmRAP2-13 that binds directly to the 426 bp key sequence and activates expression of CmBAK1 was identified. Moreover, transcriptional analysis revealed an obvious increase in transcript levels of CsRAP2-13 in Citron C-05, American citron, and Finger citron. In this study, we present the identification of transcriptional activators that are found to interact with BAK1 proteins in response to Xcc. These results reveal a coordinated regulatory mechanism of RAP2-13, which may be involved in defence responses through the regulation of BAK1. Full article

The elaboration and definition of “metabolic fingerprints” can subsidize both the identification and determination of plant varieties, as well as the increase in knowledge about the responses and adaptations of plants to external and/or internal factors. The lemon tree (Citrus limon Burm.) is one of the most important crops in the Spanish southeast and is often consumed around the world. Although the study and characterization of its fruits are common due to its economic interest, its leaves are limited to specific functionalized studies related to the objective of the work (extraction of essential oils, stabilizing agent, aromatic extracts, etc.). So, this study aimed to identify the primary and secondary metabolites of Citrus limon Burm. (‘Verna’ variety) leaf samples cultivated under different conditions (three rootstocks and three culture media). In total, 19 metabolites were identified for all samples, of which 9 were amino acids, 5 organic acids, 3 sugars and 2 intermediate metabolites. The results pointed to a limited influence, both of the substrate and of the crop rootstock, on the metabolomic differentiation of lemon leaves. Knowledge and foliar metabolomic differentiation can offer important information that supports the application of crop foliar treatments but also helps in the management of diseases and pests. Full article

Aphids are small insects that feed on plant sap, and they belong to a superfamily called Aphoidea. They are among the major pests causing damage to citrus crops in most parts of the world. Precise and automatic identification of aphids is needed to understand citrus pest dynamics and management. This article presents a dataset that contains 665 healthy and unhealthy lemon leaf images. The latter are leaves with the presence of aphids, and visible white spots characterize them. Moreover, each image includes a set of annotations that identify the leaf, its health state, and the infestation severity according to the percentage of the affected area on it. Images were collected manually in real-world conditions in a lemon plant field in Junín, Manabí, Ecuador, during the winter, by using a smartphone camera. The dataset is called LeLePhid: lemon (Le) leaf (Le) image dataset for aphid (Phid) detection and infestation severity. The data can facilitate evaluating models for image segmentation, detection, and classification problems related to plant disease recognition. Full article

‘Candidatus Liberibacter asiaticus’ (CLas) is the pathogenic bacterium that causes the disease Huanglongbing (HLB) in citrus and some model plants, such as Nicotiana benthamiana. After infection, CLas releases a set of effectors to modulate host responses. One of these critical effectors is Sec-delivered effector 1 (SDE1), which induces chlorosis and cell death in N. benthamiana. In this study, we revealed the DEAD-box RNA helicase (DDX3) interacts with SDE1. Gene silencing study revealed that knockdown of the NbDDX3 gene triggers leaf chlorosis, mimicking the primary symptom of CLas infection in N. benthamiana. The interactions between SDE1 and NbDDX3 were localized in the cell membrane. Overexpression of SDE1 resulted in suppression of NbDDX3 gene expression in N. benthamiana, which suggests a critical role of SDE1 in modulating NbDDX3 expression. Furthermore, we verified the interaction of SDE1 with citrus DDX3 (CsDDX3), and demonstrated that the expression of the CsDDX3 gene was significantly reduced in HLB-affected yellowing and mottled leaves of citrus. Thus, we provide molecular evidence that the downregulation of the host DDX3 gene is a crucial mechanism of leaf chlorosis in HLB-affected plants. The identification of CsDDX3 as a critical target of SDE1 and its association with HLB symptom development indicates that the DDX3 gene is an important target for gene editing, to interrupt the interaction between DDX3 and SDE1, and therefore interfere host susceptibility. Full article

Alternaria alternata is a fungus that causes a serious disease in susceptible genotypes of citrus, particularly in mandarins. The Alternaria citri toxin (ACT) produced by the pathogen induces necrotic lesions on young leaves and fruits, defoliation and fruit drop. Here, we describe two methods of marker-assisted selection (MAS) that could be used for the early identification of Alternaria brown spot (ABS)-resistant mandarin hybrids. The first method is based on a nested PCR coupled to high resolution melting (HRM) analysis at the SNP08 locus, which is located at 0.4 cM from the ABS resistance locus, and was previously indicated as the most suitable for the selection of ABS-resistant hybrids. The method was validated on 41 mandarin hybrids of the CREA germplasm collection, and on 862 progenies generated from five crosses involving different susceptible parents. Four out of five populations showed Mendelian segregation at the analyzed locus, while a population involving Murcott tangor as male parent showed distorted segregation toward the susceptible hybrids. The second method is based on a cleaved amplified polymorphic sequences (CAPS) marker that was developed using the same primers as the nested PCR at the SNP08 locus, coupled with BccI restriction enzyme digestion. To verify the reliability of the two genotyping methods, in vitro leaf phenotyping was carried out by inoculating A. alternata spores onto young leaves of 101 hybrids, randomly chosen among the susceptible and resistant progenies. The phenotyping confirmed the SNP08 genotyping results, so the proposed method of selection based on HRM or CAPS genotyping could be routinely used as an alternative to KBioscience competitive allele specific polymerase chain reaction (KASPar) single nucleotide polymorphism (SNP) genotyping system to improve citrus breeding programs. While the study confirmed that the SNP08 marker is a reliable tool for MAS of new citrus hybrids with different genetic backgrounds, it also identified a small group of genotypes where the resistance mechanism requires further investigation. Full article