Sugarcane disease recognition through visible and near-infrared spectroscopy using deep learning assisted continuous wavelet transform-based spectrogram.

Utilizing visible and near-infrared (Vis-NIR) spectroscopy in conjunction with chemometrics methods has been widespread for identifying plant diseases. However, a key obstacle involves the extraction of relevant spectral characteristics. This study aimed to enhance sugarcane disease recognition by combining convolutional neural network (CNN) with continuous wavelet transform (CWT) spectrograms for spectral features extraction within the Vis-NIR spectra (380-1400 nm) to improve the accuracy of sugarcane diseases recognition. Using 130 sugarcane leaf samples, the obtained one-dimensional CWT coefficients from Vis-NIR spectra were transformed into two-dimensional spectrograms. Employing CNN, spectrogram features were extracted and incorporated into decision tree, K-nearest neighbour, partial least squares discriminant analysis, and random forest (RF) calibration models. The RF model, integrating spectrogram-derived features, demonstrated the best performance with an average precision of 0.9111, sensitivity of 0.9733, specificity of 0.9791, and accuracy of 0.9487. This study may offer a non-destructive, rapid, and accurate means to detect sugarcane diseases, enabling farmers to receive timely and actionable insights on the crops' health, thus minimizing crop loss and optimizing yields.

Optimization of loop-mediated isothermal amplification assay for sunflower mildew disease detection.

Loop-Mediated Isothermal Amplification (LAMP) represents a valuable technique for DNA/RNA detection, known for its exceptional sensitivity, specificity, speed, accuracy, and affordability. This study focused on optimizing a LAMP-based method to detect early signs of Plasmopara halstedii, the casual pathogen of sunflower downy mildew, a severe threat to sunflower crops. Specifically, a set of six LAMP primers (two outer, two inner, and two loop) were designed from P. halstedii genomic DNA, targeting the ribosomal Large Subunit (LSU). These primers were verified by in silico analysis and experimental validation using both target and non-target species' DNAs. Optimizations encompassing reaction conditions (temperature, time) and component concentrations (magnesium, Bst DNA polymerase, primers, and dNTP) were determined. Validation of these optimizations was performed by agarose gel electrophoresis. Furthermore, various colorimetric chemicals (Neutral Red, Hydroxynaphthol Blue, SYBR Safe, Thiazole Green) were evaluated to facilitate method analysis, and the real-time analysis has been optimized, presenting multiple approaches for detecting sunflower downy mildew using the LAMP technique. The analytical sensitivity of the method was confirmed by detecting P. halstedii DNA concentrations as low as 0.5 pg/µl. This pioneering study, establishing P. halstedii detection through the LAMP method, stands as unique in its field. The precision, robustness, and practicality of the LAMP protocol make it an ideal choice for studies focusing on sunflower mildew, emphasizing its recommended use due to its operational ease and reliability.

Metabolome-driven microbiome assembly determining the health of ginger crop (Zingiber officinale L. Roscoe) against rhizome rot.

BACKGROUND: Plant-associated microorganisms can be found in various plant niches and collectively comprise the plant microbiome. The plant microbiome assemblages have been extensively studied, primarily in model species. However, a deep understanding of the microbiome assembly associated with plant health is still needed. Ginger rhizome rot has been variously attributed to multiple individual causal agents. Due to its global relevance, we used ginger and rhizome rot as a model to elucidate the metabolome-driven microbiome assembly associated with plant health. RESULTS: Our study thoroughly examined the biodiversity of soilborne and endophytic microbiota in healthy and diseased ginger plants, highlighting the impact of bacterial and fungal microbes on plant health and the specific metabolites contributing to a healthy microbial community. Metabarcoding allowed for an in-depth analysis of the associated microbial community. Dominant genera represented each microbial taxon at the niche level. According to linear discriminant analysis effect size, bacterial species belonging to Sphingomonas, Quadrisphaera, Methylobacterium-Methylorubrum, Bacillus, as well as the fungal genera Pseudaleuria, Lophotrichus, Pseudogymnoascus, Gymnoascus, Mortierella, and Eleutherascus were associated with plant health. Bacterial dysbiosis related to rhizome rot was due to the relative enrichment of Pectobacterium, Alcaligenes, Klebsiella, and Enterobacter. Similarly, an imbalance in the fungal community was caused by the enrichment of Gibellulopsis, Pyxidiophorales, and Plectosphaerella. Untargeted metabolomics analysis revealed several metabolites that drive microbiome assembly closely related to plant health in diverse microbial niches. At the same time, 6-({[3,4-dihydroxy-4-(hydroxymethyl)oxolan-2-yl]oxy}methyl)oxane-2,3,4,5-tetrol was present at the level of the entire healthy ginger plant. Lipids and lipid-like molecules were the most significant proportion of highly abundant metabolites associated with ginger plant health versus rhizome rot disease. CONCLUSIONS: Our research significantly improves our understanding of metabolome-driven microbiome structure to address crop protection impacts. The microbiome assembly rather than a particular microbe's occurrence drove ginger plant health. Most microbial species and metabolites have yet to be previously identified in ginger plants. The indigenous microbial communities and metabolites described can support future strategies to induce plant disease resistance. They provide a foundation for further exploring pathogens, biocontrol agents, and plant growth promoters associated with economically important crops. Video Abstract.

Extracellular proteases from microbial plant pathogens as virulence factors.

Plant pathogens deploy specialized proteins to aid disease progression, some of these proteins function in the apoplast and constitute proteases. Extracellular virulence proteases have been described to play roles in plant cell wall manipulation and immune evasion. In this review, we discuss the evidence for the hypothesized virulence functions of bacterial, fungal, and oomycete extracellular proteases. We highlight the contrast between the low number of elucidated functions for these proteins and the size of extracellular protease repertoires among pathogens. Finally, we suggest that the refinement of in planta 'omics' approaches, combined with recent advances in modeling and mechanism-based methods for trapping substrates finally make it possible to address this knowledge gap.

Monitoring of plant diseases caused by Fusarium commune and Rhizoctonia solani in bok choy using hyperspectral remote sensing and machine learning.

BACKGROUND: Local vegetable production is susceptible to various fungal pathogens, the most common and lethal of which are Fusarium commune and Rhizoctonia solani. Early detection of these pathogens is challenging, and by the time visual symptoms appear, the pathogens may have already spread extensively, causing massive damage to the production. In this study, we explored the use of hyperspectral data for early detection of diseases caused by F. commune or R. solani in bok choy Brassica rapa subsp. chinensis by collecting hyperspectral data from healthy plants and plants inoculated with either fungal pathogen in a controlled experimental set-up. RESULTS: Based on the collected data, we employed various tree-based, distribution-based, geometric, neural networks and ensemble learning algorithms to train detection models. Among the trained models, Multi-Layer Perceptron (MLP) models performed the best with overall accuracy reaching 95.9 ± 0.26%. MLP models could differentiate between healthy and infected plants with 99% precision after 1 day of infection, and distinguish between different fungal pathogens with 99% precision after 2 days. During this period, no visible symptoms of fungal infection could be observed. Further analysis into trained MLP models and general reflectance profiles of plants also revealed a high correlation of the spectral regions 445-460, 560-595, 606-620 and 719-728 nm with fungal infection in bok choy plants. CONCLUSION: Our findings highlight the potential of hyperspectral imaging as a highly precise early detection tool for fungal diseases in plants. © 2024 Society of Chemical Industry.

Insect hypovirulence-associated mycovirus confers entomopathogenic fungi with enhanced resistance against phytopathogens.

Mycoviruses can alter the biological characteristics of host fungi, including change virulence or pathogenicity of phytopathogens and entomopathogenic fungi (EPF). However, most studies on the mycoviruses found in EPF have focused on the effects of the viruses on the virulence of host fungi towards insect pests, with relatively few reports on the effects to the host fungi with regard to plant disease resistance in hosts. The present study investigated the effects of the mycovirus Beauveria bassiana chrysovirus 2 (BbCV2) virus infection on host biological characteristics, evaluated antagonistic activity of BbCV2 against two phytopathogenic fungi (Sclerotinia sclerotiorum and Botrytis cinerea), and transcriptome analysis was used to reveal the interactions between viruses and hosts. Our results showed that BbCV2 virus infection increased B. bassiana's growth rate, spore production, and biomass, it also enhanced the capacity of host fungi and their metabolic products to inhibit phytopathogenic fungi. BbCV2 virus infection reduced the contents of the two pathogens in tomato plants significantly, and transcriptome analysis revealed that the genes related to competition for ecological niches and nutrition, mycoparasitism and secondary metabolites in B. bassiana were significantly up-regulated after viral infection. These findings indicated that the mycovirus infection is an important factor to enhance the ability of B. bassiana against plant disease after endophytic colonization. We suggest that mycovirus infection causes a positive effect on B. bassiana against phytopathogens, which should be considered as a potential strategy to promote the plant disease resistance of EPF.

Deep learning for mango leaf disease identification: A vision transformer perspective.

Over the last decade, the use of machine learning in smart agriculture has surged in popularity. Deep learning, particularly Convolutional Neural Networks (CNNs), has been useful in identifying diseases in plants at an early stage. Recently, Vision Transformers (ViTs) have proven to be effective in image classification tasks. These architectures often outperform most state-of-the-art CNN models. However, the adoption of vision transformers in agriculture is still in its infancy. In this paper, we evaluated the performance of vision transformers in identification of mango leaf diseases and compare them with popular CNNs. We proposed an optimized model based on a pretrained Data-efficient Image Transformer (DeiT) architecture that achieves 99.75% accuracy, better than many popular CNNs including SqueezeNet, ShuffleNet, EfficientNet, DenseNet121, and MobileNet. We also demonstrated that vision transformers can have a shorter training time than CNNs, as they require fewer epochs to achieve optimal results. We also proposed a mobile app that uses the model as a backend to identify mango leaf diseases in real-time.

Coating seeds with biocontrol bacteria-loaded sodium alginate/pectin hydrogel enhances the survival of bacteria and control efficacy against soil-borne vegetable diseases.

Microbial seed coatings serve as effective, labor-saving, and ecofriendly means of controlling soil-borne plant diseases. However, the survival of microbial agents on seed surfaces and in the rhizosphere remains a crucial challenge. In this work, we embedded a biocontrol bacteria (Bacillus subtilis ZF71) in sodium alginate (SA)/pectin (PC) hydrogel as a seed coating agent to control Fusarium root rot in cucumber. The formula of SA/PC hydrogel was optimized with the highest coating uniformity of 90 % in cucumber seeds. SA/PC hydrogel was characterized using rheological, gel content, and water content tests, thermal gravimetric analysis, and Fourier transform infrared spectroscopy. Bacillus subtilis ZF71 within the SA/PC hydrogel network formed a biofilm-like structure with a high viable cell content (8.30 log CFU/seed). After 37 days of storage, there was still a high number of Bacillus subtilis ZF71 cells (7.23 log CFU/seed) surviving on the surface of cucumber seeds. Pot experiments revealed a higher control efficiency against Fusarium root rot in ZF71-SA/PC cucumber seeds (53.26 %) compared with roots irrigated with a ZF71 suspension. Overall, this study introduced a promising microbial seed coating strategy based on biofilm formation that improved performance against soil-borne plant diseases.

Phosphatases: Decoding the Role of Mycorrhizal Fungi in Plant Disease Resistance.

Mycorrhizal fungi, a category of fungi that form symbiotic relationships with plant roots, can participate in the induction of plant disease resistance by secreting phosphatase enzymes. While extensive research exists on the mechanisms by which mycorrhizal fungi induce resistance, the specific contributions of phosphatases to these processes require further elucidation. This article reviews the spectrum of mycorrhizal fungi-induced resistance mechanisms and synthesizes a current understanding of how phosphatases mediate these effects, such as the induction of defense structures in plants, the negative regulation of plant immune responses, and the limitation of pathogen invasion and spread. It explores the role of phosphatases in the resistance induced by mycorrhizal fungi and provides prospective future research directions in this field.

Noninvasive Raman spectroscopy for the detection of rice bacterial leaf blight and bacterial leaf streak.

Plant diseases pose significant threats to agricultural yields and are responsible for nearly 20 % of losses in total food production. Therefore, the rapid detection of plant pathogens is critically important for preventing the rapid development of plant diseases and minimizing crop damage. Raman spectroscopy (RS) has been shown to be effective for detecting living biological samples. Compared with traditional detection methods, RS is fast, sensitive, and non-destructive; it also does not require sample labeling. In this study, we used Laser tweezers Raman spectroscopy combined with convolutional neural networks to detect two closely related strains of bacteria, Xanthomonas oryzae pv. oryzae (Xoo) and Xanthomonas oryzae pv. oryzicola (Xoc), exuded from bacteria-infected rice leaves. The accuracy of this technique was 97.5 %. For the application of RS in the field, we used the portable Raman spectrometer to detect mock-inoculated as well as Xoo- and Xoc-infected rice leaves at different disease courses. The identification accuracy via this technique was 87.02 % in the early stage, in which no obvious symptoms were apparent. This method also revealed spectral differences in rice leaves caused by the two bacteria, which could be leveraged for subsequent analysis of the molecular mechanism of infection. Our results indicate that RS is a promising approach for the early detection of bacterial diseases in rice in the field, as well as for in-depth single-cell analysis in laboratory settings.

The Past, Present, and Future of Plant Activators Targeting the Salicylic Acid Signaling Pathway.

Plant activators have emerged as promising alternatives to conventional crop protection chemicals for managing crop diseases due to their unique mode of action. By priming the plant's innate immune system, these compounds can induce disease resistance against a broad spectrum of pathogens without directly inhibiting their proliferation. Key advantages of plant activators include prolonged defense activity, lower effective dosages, and negligible risk of pathogen resistance development. Among the various defensive pathways targeted, the salicylic acid (SA) signaling cascade has been extensively explored, leading to the successful development of commercial activators of systemic acquired resistance, such as benzothiadiazole, for widespread application in crop protection. While the action sites of many SA-targeting activators have been preliminarily mapped to different steps along the pathway, a comprehensive understanding of their precise mechanisms remains elusive. This review provides a historical perspective on plant activator development and outlines diverse screening strategies employed, from whole-plant bioassays to molecular and transgenic approaches. We elaborate on the various components, biological significance, and regulatory circuits governing the SA pathway while critically examining the structural features, bioactivities, and proposed modes of action of classical activators such as benzothiadiazole derivatives, salicylic acid analogs, and other small molecules. Insights from field trials assessing the practical applicability of such activators are also discussed. Furthermore, we highlight the current status, challenges, and future prospects in the realm of SA-targeting activator development globally, with a focus on recent endeavors in China. Collectively, this comprehensive review aims to describe existing knowledge and provide a roadmap for future research toward developing more potent plant activators that enhance crop health.

Two new species of Dendrostoma (Erythrogloeaceae, Diaporthales) associated with Castaneamollissima canker disease in China.

The genus Dendrostoma is known to inhabit tree barks associated with branch canker diseases in China and several countries of Europe. Previous studies indicated that species of Dendrostoma prefer inhabiting fagaceous hosts, especially species of Castanea. In the present study, we obtained four isolates from cankered branches of Chinese chestnut (C.mollissima) in Rizhao City, Shandong Province, China. Morphological comparisons and phylogenetical analyses of a combined ITS-tef1-rpb2 sequence matrix were conducted, which revealed two new species named Dendrostomarizhaoense sp. nov. and D.tianii sp. nov. The new taxa are compared with other Dendrostoma species and comprehensive descriptions and illustrations are provided herein.

Contemporary applications of vibrational spectroscopy in plant stresses and phenotyping.

Plant pathogens, including viruses, bacteria, and fungi, cause massive crop losses around the world. Abiotic stresses, such as drought, salinity and nutritional deficiencies are even more detrimental. Timely diagnostics of plant diseases and abiotic stresses can be used to provide site- and doze-specific treatment of plants. In addition to the direct economic impact, this "smart agriculture" can help minimizing the effect of farming on the environment. Mounting evidence demonstrates that vibrational spectroscopy, which includes Raman (RS) and infrared spectroscopies (IR), can be used to detect and identify biotic and abiotic stresses in plants. These findings indicate that RS and IR can be used for in-field surveillance of the plant health. Surface-enhanced RS (SERS) has also been used for direct detection of plant stressors, offering advantages over traditional spectroscopies. Finally, all three of these technologies have applications in phenotyping and studying composition of crops. Such non-invasive, non-destructive, and chemical-free diagnostics is set to revolutionize crop agriculture globally. This review critically discusses the most recent findings of RS-based sensing of biotic and abiotic stresses, as well as the use of RS for nutritional analysis of foods.

Order of arrival and nutrient supply alter outcomes of co-infection with two fungal pathogens.

A pathogen arriving on a host typically encounters a diverse community of microbes that can shape priority effects, other within-host interactions and infection outcomes. In plants, environmental nutrients can drive trade-offs between host growth and defence and can mediate interactions between co-infecting pathogens. Nutrients may thus alter the outcome of pathogen priority effects for the host, but this possibility has received little experimental investigation. To disentangle the relationship between nutrient availability and co-infection dynamics, we factorially manipulated the nutrient availability and order of arrival of two foliar fungal pathogens (Rhizoctonia solani and Colletotrichum cereale) on the grass tall fescue (Lolium arundinaceum) and tracked disease outcomes. Nutrient addition did not influence infection rates, infection severity or plant biomass. Colletotrichum cereale facilitated R. solani, increasing its infection rate regardless of their order of inoculation. Additionally, simultaneous and C. cereale-first inoculations decreased plant growth and-in plants that did not receive nutrient addition-increased leaf nitrogen concentrations compared to uninoculated plants. These effects were partially, but not completely, explained by the duration and severity of pathogen infections. This study highlights the importance of understanding the intricate associations between the order of pathogen arrival, host nutrient availability and host defence to better predict infection outcomes.

Soil Fungal Pathogens in Pinus pinaster Mature Reforestation: Silvicultural Treatments Effects.

Soil fungal communities play a key role in multiple functions and ecosystem services within forest ecosystems. Today, forest ecosystems are subject to multiple environmental and anthropogenic disturbances (e.g., fire or forest management) that mainly lead to changes in vegetation as well as in plant-soil interactions. Soil pathogens play an important role in controlling plant diversity, ecosystem functions, and human and animal health. In this work we analyzed the response of soil plant pathogenic fungi to forest management in a Pinus pinaster reforestation. We started from an experimental design, in which forest thinning and gap cutting treatments were applied at different intensities and sizes, respectively. The fungal communities of plant pathogens in spring were described, and the effect of the silvicultural treatments was evaluated 5 years after their application, as were the possible relationships between soil plant pathogenic fungal communities and other environmental factors. Only a strong low thinning treatment (35% basal area) was able to generate homogeneous changes in soil pathogenic diversity. In the gaps, only the central position showed significant changes in the soil plant pathogenic fungi community.

Optimizing Corn Tar Spot Measurement: A Deep Learning Approach Using Red-Green-Blue Imaging and Stromata Contour Detection Algorithm for Leaf-Level Disease Severity Analysis.

Visual detection of stromata (brown-black, elevated fungal fruiting bodies) is a primary method for quantifying tar spot early in the season, as these structures are definitive signs of the disease and essential for effective disease monitoring and management. Here, we present Stromata Contour Detection Algorithm version 2 (SCDA v2), which addresses the limitations of the previously developed SCDA version 1 (SCDA v1) without the need for empirical search of the optimal Decision Making Input Parameters (DMIPs), while achieving higher and consistent accuracy in tar spot stromata detection. SCDA v2 operates in two components: (i) SCDA v1 producing tar-spot-like region proposals for a given input corn leaf Red-Green-Blue (RGB) image, and (ii) a pre-trained Convolutional Neural Network (CNN) classifier identifying true tar spot stromata from the region proposals. To demonstrate the enhanced performance of the SCDA v2, we utilized datasets of RGB images of corn leaves from field (low, middle, and upper canopies) and glasshouse conditions under variable environments, exhibiting different tar spot severities at various corn developmental stages. Various accuracy analyses (F1-score, linear regression, and Lin's concordance correlation), showed that SCDA v2 had a greater agreement with the reference data (human visual annotation) than SCDA v1. SCDA v2 achievd 73.7% mean Dice values (overall accuracy), compared to 30.8% for SCDA v1. The enhanced F1-score primarily resulted from eliminating overestimation cases using the CNN classifier. Our findings indicate the promising potential of SCDA v2 for glasshouse and field-scale applications, including tar spot phenotyping and surveillance projects.

The impact of fine-tuning paradigms on unknown plant diseases recognition.

Plant diseases pose significant threats to agriculture, impacting both food safety and public health. Traditional plant disease detection systems are typically limited to recognizing disease categories included in the training dataset, rendering them ineffective against new disease types. Although out-of-distribution (OOD) detection methods have been proposed to address this issue, the impact of fine-tuning paradigms on these methods has been overlooked. This paper focuses on studying the impact of fine-tuning paradigms on the performance of detecting unknown plant diseases. Currently, fine-tuning on visual tasks is mainly divided into visual-based models and visual-language-based models. We first discuss the limitations of large-scale visual language models in this task: textual prompts are difficult to design. To avoid the side effects of textual prompts, we futher explore the effectiveness of purely visual pre-trained models for OOD detection in plant disease tasks. Specifically, we employed five publicly accessible datasets to establish benchmarks for open-set recognition, OOD detection, and few-shot learning in plant disease recognition. Additionally, we comprehensively compared various OOD detection methods, fine-tuning paradigms, and factors affecting OOD detection performance, such as sample quantity. The results show that visual prompt tuning outperforms fully fine-tuning and linear probe tuning in out-of-distribution detection performance, especially in the few-shot scenarios. Notably, the max-logit-based on visual prompt tuning achieves an AUROC score of 94.8 % in the 8-shot setting, which is nearly comparable to the method of fully fine-tuning on the full dataset (95.2 % ), which implies that an appropriate fine-tuning paradigm can directly improve OOD detection performance. Finally, we visualized the prediction distributions of different OOD detection methods and discussed the selection of thresholds. Overall, this work lays the foundation for unknown plant disease recognition, providing strong support for the security and reliability of plant disease recognition systems. We will release our code at https://github.com/JiuqingDong/PDOOD to further advance this field.

Suppressive Effect of Black Soldier Fly Larvae Frass on Fusarium Wilt Disease in Tomato Plants.

This study investigated the effect of black soldier fly larvae (BSFL) frass derived from BSFL reared on a diet composed of fruit/vegetable/bakery/brewery residues (FVBB diet) and on the Gainesville diet (GV diet) on the development of tomato (Solanum lycopersicum) Fusarium wilt caused by Fusarium oxysporum f. sp. lycopersici (FOL). Tomato plants were grown in a substrate inoculated with FOL that was amended (10%, v:v) or not (control) with either a commercial compost, pasteurized (70 °C for 1 h) frass from BSFL reared on a FVBB diet, non-pasteurized frass from BSFL reared on a FVBB diet, pasteurized frass from BSFL reared on the GV diet, or non-pasteurized frass from BSFL reared on the GV diet. The results show that frass from BSFL reared on the GV diet, irrespective of pasteurization, inhibited FOL root colonization and reduced the severity of tomato Fusarium wilt to a far greater extent than frass from BSFL reared on a FVBB diet and commercial compost made of peat, seaweed, and shrimps. This study suggests that BSFL frass, depending on the larval rearing diet, has the potential to serve as a pasteurized or non-pasteurized soil amendment with prophylactic properties against FOL in tomato plants, opening new avenues of research for the valorization of BSFL frass.

Diversity of Trichoderma species associated with the black rot disease of Gastrodia elata, including four new species.

Introduction: Trichoderma species establish symbiotic relationships with plants through both parasitic and mutualistic mechanisms. While some Trichoderma species act as plant pathogenic fungi, others utilize various strategies to protect and enhance plant growth. Methods: Phylogenetic positions of new species of Trichoderma were determined through multi-gene analysis relying on the internal transcribed spacer (ITS) regions of the ribosomal DNA, the translation elongation factor 1-α (tef1-α) gene, and the RNA polymerase II (rpb2) gene. Additionally, pathogenicity experiments were conducted, and the aggressiveness of each isolate was evaluated based on the area of the cross-section of the infected site. Results: In this study, 13 Trichoderma species, including 9 known species and 4 new species, namely, T. delicatum, T. robustum, T. perfasciculatum, and T. subulatum were isolated from the diseased tubers of Gastrodia elata in Yunnan, China. Among the known species, T. hamatum had the highest frequency. T. delicatum belonged to the Koningii clade. T. robustum and T. perfasciculatum were assigned to the Virens clade. T. subulatum emerged as a new member of the Spirale clade. Pathogenicity experiments were conducted on the new species T. robustum, T. delicatum, and T. perfasciculatum, as well as the known species T. hamatum, T. atroviride, and T. harzianum. The infective abilities of different Trichoderma species on G. elata varied, indicating that Trichoderma was a pathogenic fungus causing black rot disease in G. elata. Discussion: This study provided the morphological characteristics of new species and discussed the morphological differences with phylogenetically proximate species, laying the foundation for research aimed at preventing and managing diseases that affect G. elata.

PND-Net: plant nutrition deficiency and disease classification using graph convolutional network.

Crop yield production could be enhanced for agricultural growth if various plant nutrition deficiencies, and diseases are identified and detected at early stages. Hence, continuous health monitoring of plant is very crucial for handling plant stress. The deep learning methods have proven its superior performances in the automated detection of plant diseases and nutrition deficiencies from visual symptoms in leaves. This article proposes a new deep learning method for plant nutrition deficiencies and disease classification using a graph convolutional network (GNN), added upon a base convolutional neural network (CNN). Sometimes, a global feature descriptor might fail to capture the vital region of a diseased leaf, which causes inaccurate classification of disease. To address this issue, regional feature learning is crucial for a holistic feature aggregation. In this work, region-based feature summarization at multi-scales is explored using spatial pyramidal pooling for discriminative feature representation. Furthermore, a GCN is developed to capacitate learning of finer details for classifying plant diseases and insufficiency of nutrients. The proposed method, called Plant Nutrition Deficiency and Disease Network (PND-Net), has been evaluated on two public datasets for nutrition deficiency, and two for disease classification using four backbone CNNs. The best classification performances of the proposed PND-Net are as follows: (a) 90.00% Banana and 90.54% Coffee nutrition deficiency; and (b) 96.18% Potato diseases and 84.30% on PlantDoc datasets using Xception backbone. Furthermore, additional experiments have been carried out for generalization, and the proposed method has achieved state-of-the-art performances on two public datasets, namely the Breast Cancer Histopathology Image Classification (BreakHis 40 × : 95.50%, and BreakHis 100 × : 96.79% accuracy) and Single cells in Pap smear images for cervical cancer classification (SIPaKMeD: 99.18% accuracy). Also, the proposed method has been evaluated using five-fold cross validation and achieved improved performances on these datasets. Clearly, the proposed PND-Net effectively boosts the performances of automated health analysis of various plants in real and intricate field environments, implying PND-Net's aptness for agricultural growth as well as human cancer classification.