Abstract The classification of banana leaf disease has a large impact on agricultural output and relies heavily on timely early detection, with reliability as a fundamental component of effective crop management. The framework proposed in this study comprises a Hamiltonian Full Node Coverage Graph Attention Network (HFNC-GAT) and Fuzzy C-Means super pixel graph learning to explain the classification of banana leaf diseases. The HFNC-GAT allows for the representation of segmented leaf areas as the nodes in a graph. This framework also makes optimal use of an attention learning model to represent the spatial dependence of diseased leaf regions, allowing it to leverage both local and global spatial dependencies. The HFNC-GAT was demonstrated through observed experiments to achieve high performance with 96.11 and 94.19 accuracy, 0.9111 Cohen's Kappa, 0.9111 MCC, 0.9344 F2-score, and 0.9939 ROC-AUC compared to the performance of conventional CNN, GCN, and baseline GAT models.
Why it matches plant phenotyping methodsバナナ葉の病斑領域を画像から抽出・分類するグラフ学習手法が研究の中心であり、植物の病害状態を直接推定するため、植物フェノタイピング手法として適格です。
abstractThe framework proposed in this study comprises a Hamiltonian Full Node Coverage Graph Attention Network (HFNC-GAT) and Fuzzy C-Means super pixel graph learning to explain the classification of banana leaf diseases.
Reproduction assets foundThe paper's Dataset Availability statement explicitly declares two public Kaggle banana leaf image datasets used for the phenotyping/classification experiments: Banana Leaf Disease Dataset V4 and BananaLSD. No author analysis code or trained model is reported as publicly available.Dataset · publicThe Banana Leaf Disease
Dataset V4 is available at https://www.kaggle.com/datasets/rayhanarlistya/banana-leaf-disease-dataset-v4.Open asset ↗Kaggle · banana-leaf-disease-dataset-v4pdf-page:19 lines:1-55Dataset · publicThe Banana Leaf Spot Diseases (BananaLSD) dataset is available at
https://www.kaggle.com/datasets/shifatearman/bananalsdOpen asset ↗Kaggle · bananalsdpdf-page:19 lines:1-55Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Abstract Abstract. Accurate and timely identi cation of plant diseases is essential for improving crop productivity and ensuring sustainable agricultural practices. This paper presents a comprehensive image dataset of fruit and leaf diseases covering six economically important horticultural crops: Apple, Banana, Citrus, Guava, Mango, and Papaya. The dataset comprises high-quality RGB images representing both healthy and diseased samples, with disease symptoms including spots, lesions, discoloration, blight, rot, and fungal and bacterial infections captured under diverse real-world conditions. Variations in illumination, background complexity, viewing angles, growth stages, and symptom severity are intentionally included to enhance the robustness and generalizability of learning models developed using this data. The dataset is structured in a class-wise manner and preprocessed to support direct integration with deep learning frameworks. It is extensively used to train, validate, and evaluate deep learning based plant disease classi cation models, enabling automatic feature learning from raw images without manual intervention. Experimental usage demonstrates that the dataset is well suited for convolutional neural networks and attentionbased architectures, facilitating e ective discrimination between multiple disease categories across di erent crops and plant organs. By providing a uni ed multi-crop, multi-disease benchmark, this dataset aims to accelerate research in automated crop disease diagnosis, precision agriculture, and intelligent decision-support systems for sustainable farming.
Why it matches plant phenotyping methods植物の葉・果実の病徴画像を収録したデータセット／ベンチマークであり、病害状態の画像ベース推定を中心的に扱うため。
abstractThis paper presents a comprehensive image dataset of fruit and leaf diseases covering six economically important horticultural crops
Reproduction assets foundThe paper's core asset is the ABCGMP fruit and leaf disease image dataset, publicly deposited on Mendeley Data, with author analysis code also stated to be available on GitHub. Both are paper-specific, public, and actionable.Dataset · publicData is available on Mendeley:1Open asset ↗pdf-page:33 lines:1-56Code · publicCode availability: Code is available on GitHub 2Open asset ↗GitHubpdf-page:33 lines:1-56Code / dataset availability confirmedOpenAlex · Crossref · checked 13 Sept 2026
Banana (Musa spp.) fruit morphology is a key determinant of yield and quality, yet modeling its 3D structural dynamics across genotypes remains difficult. To address this challenge, we developed a generic, biomass-driven 3D structural model for banana fruit fingers that quantitatively links growth and morphology. Field experiments were conducted over two growing seasons in Hainan, China, using three representative genotypes. Morphological traits, including outer and inner arc length, circumference, and pedicel length, along with dry (Wd) and fresh weight (Wf), were measured every 10 days after flowering until 110 days. Quantitative relationships between morphological traits and Wf, as well as between Wd and Wf, were fitted using linear or Gompertz functions with genotype-specific parameters. Based on these functions, a parameterized 3D reconstruction method was implemented in Python, combining biomass-driven growth equations, curvature geometry, and cross-sectional interpolation to simulate the fruit’s bending, tapering, and volumetric development. The resulting dynamic 3D models accurately reproduced genotype-specific differences in curvature, length, and shape with average fitting R2 > 0.95. The proposed biomass-driven 3D structural model provides a methodological framework for integrating banana fruit morphology into functional–structural plant models.
Why it matches plant phenotyping methodsバナナ果実の形態形質を推定・再現するバイオマス駆動型3D構造モデルを開発し、遺伝子型間の形状を検証しており、フェノタイピング手法が中心である。
abstractwe developed a generic, biomass-driven 3D structural model for banana fruit fingers that quantitatively links growth and morphology.
Reproduction assets foundThe paper explicitly states that the source code of the Banana Morphology Simulation System and the datasets are publicly available on GitHub at the authors' URL, which matches an allowed URL. This covers the paper's phenotyping datasets and analysis/3D modeling code.Code · publicData analysis was performed using a custom-developed software platform, the Banana
Morphology Simulation System. The source code and datasets are publicly available
on GitHub (https://github.com/Interstingsun/SimBanana, accessed on 4 January 2026).Open asset ↗Interstingsun/SimBananapdf-page:5 lines:1-24Dataset · publicThe source code and datasets are publicly available
on GitHub (https://github.com/Interstingsun/SimBanana, accessed on 4 January 2026).Open asset ↗Interstingsun/SimBananapdf-page:5 lines:1-24Code / dataset availability confirmedCrossref · Europe PMC · checked 15 Sept 2026
Abstract The growth and productivity of banana crops are critically affected by micronutrient deficiencies, which are often difficult to detect at early stages. Lightweight deep learning models, optimized through neural architecture search (NAS) and attention mechanisms, are hypothesized to provide accurate and efficient classification of such deficiencies for real-time agricultural applications. In this study, multiple convolutional neural networks (CNNs) and mobile-friendly architectures, including ResNet50, VGG16, NASNetMobile, and MobileNet variants (V1, V2, V3), were evaluated using transfer learning on a curated banana leaf deficiency dataset. To improve robustness and prediction accuracy, modified classification layers and ensemble strategies–initially average ensembling and later a NAS-guided dynamic attention weighting mechanism were employed. This optimization resulted in a novel lightweight model, NASMobV2 (NASNetMobile + MobileNetV2), capable of both classifying nutrient deficiencies and assessing their severity levels. The proposed model achieved a validation accuracy of 98.57%, outperforming baseline and state-of-the-art counterparts in precision, recall, and F1 score. To improve generalization, banana crop diseases along with an additional Coffee crop dataset were included for evaluation. Finally, the practical utility of the model was demonstrated by deploying the trained system in both mobile and web applications, enabling farmers and agronomists to perform fast and accurate diagnostics directly in the field.
Why it matches plant phenotyping methodsバナナ葉画像から栄養欠乏と重症度を推定する軽量深層学習モデルを開発・評価し、実運用アプリにも展開しており、植物状態の取得・推定手法が中心である。
titleA neural architecture search optimized lightweight attention ensemble model for nutrient deficiency and severity assessment in diverse crop leaves.
Reproduction assets foundThe paper's banana leaf nutrient-deficiency image dataset is publicly available on Mendeley Data and was directly used for the phenotyping/classification measurements. No author analysis code or trained model checkpoints are explicitly deposited; the GitHub/Streamlit links are deployment apps rather than deposited codeDataset · publiclidation and editing in addition to overall supervision.
Funding
Open access funding provided by Vellore Institute of Technology. We thank our Management “Vellore Institute of Technology, Vellore” for open access funding support.
Data availability
An openly available repository (Mendeley dataset) was used to perform this study;(https://data.mendeley.com/datasets/7vpdrbdkd4/1), Request for any data or materials shall be addressed to the author(sudhakar.m2020@vitstudent.ac.in).
Declarations
Competing interests
The authors declare that they have no competing interests.
References
1.
Sherefu A Zewide I
Review paper on effect of micronutrients for crop production
J. Nutr. Food Process. 2021
10.31Open asset ↗Mendeley · 7vpdrbdkd4lines:1245-1307Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Diagnosis of banana plant disease is a crucial aspect of sustaining the harvest of crops and their quality. Visual inspection of certain diseases like Black Sigatoka, Panama disease, and aphids is not easy and can lead to misjudgments. Generally, traditional deep learning approaches have been previously used but they have not performed well in addressing issues of class imbalance, sensitive disease differentiation and noisy images obtained in the field. Furthermore, most models are based on a collection of predetermined preprocessing methods and single-path networks that limit their ability to generalize to a wide variety of environments. Current methods of deep learning tend to achieve reasonable overall performance but fail to perform well on key performance indicators such as recall and F1-score when considering underrepresented and overlapping classes, such as Yellow and Black Sigatoka. Such constraints impede efficient field implementation, as diseases of minority classes are often falsely classified. To overcome these deficiencies, we develop a novel Duel-Path Attention Fusion Network (DPAFNet) that is trained utilizing adaptive quantum monarch butterfly optimization (AQMBO). The concept behind the proposed model is to feed MaxViT and HorNet-S two feature extractors to deliver global contextual details and minute-scale textural features. The traditional filters which do a reasonable job in handling dynamic noise and contrast are replaced by a learnable preprocessing unit. The cross-layer fusion attention encourages interclass discriminative learning of diseased plants. The suggested model has been trained and tested on an open-source dataset of Mendeley banana disease, which includes 5,170 images in 7 disease categories and 1 control condition. The accuracy, F1-score and MCC of 98.6% and 0.93 and 0.87 respectively (achieved experimentally) demonstrate the superiority of DPAFNet over baseline models such as EfficientNetB0 (accuracy 95.0%), DenseNet121 and ResNet50 (accuracy 93.50% and 92.0% respectively). As can be seen, the model had a 0.26-0.48 increase in F1-score in the challenging Panama disease category. These results prove that the proposed architecture can be successfully used to achieve high-accuracy disease classification in smart agriculture that is robust and prepared for field implementation.
Why it matches plant phenotyping methodsバナナ植物の病徴画像から病害状態を推定する深層学習手法を開発し、データセットとベースラインで性能検証しており、植物フェノタイピング手法が中心である。
abstractTo overcome these deficiencies, we develop a novel Duel-Path Attention Fusion Network (DPAFNet) that is trained utilizing adaptive quantum monarch butterfly optimization (AQMBO).
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Dataset · publicThe suggested model has been trained and tested
on an open-source dataset of Mendeley banana disease, which includes 5,170 images in 7
disease categories and 1 control condition.Open asset ↗pdf-page:1 lines:1-55Code / dataset availability confirmedEurope PMC · checked 6 Sept 2026
Abstract Plant diseases initiate major agricultural challenges because they lead to 16\% of worldwide crop loss in farms. The vulnerability of bananas to diseases including Xanthomonas Wilt and Sigatoka leaf spot puts severe threats to food security because of their extensive damage potential. These diseases possess the risk of damage to the complete harvest so their impact can reach 100%. While deep learning models, particularly YOLO-based architecture, have demonstrated success in plant disease identification, key research gaps remain. One major challenge is the lack of large-scale, annotated datasets for banana leaf diseases, limiting the development and evaluation of robust AI models. Addressing these challenges is crucial, and this study aims to do so by creating a large dataset and developing a robust disease detection model. The dataset comprises more than 5000 samples categorized into three classes: Healthy, Xanthomonas Wilt infected, and Sigatoka leaf spot infected. This study examines a novel framework by employing advanced optimization techniques such as Krill Herd Optimization (KHO) for YOLOv8 and its variants. Our research findings highlight the exceptional performance of the KHO-YOLOv8 model, achieving an impressive accuracy of 96.47%.
Why it matches plant phenotyping methodsバナナ葉の病害状態を画像から分類するデータセット構築とYOLOv8検出モデル開発・評価が研究の中心であり、植物の病徴を直接推定するため対象範囲に該当する。
abstractthis study aims to do so by creating a large dataset and developing a robust disease detection model.
Reproduction assets foundThe paper states its banana leaf disease dataset (5,000 annotated images) and code are publicly available on GitHub, but no concrete repository URL or identifier is provided in the supplied text, and the only allowed URL is an unrelated cited reference. The claim of public availability is explicit, but the asset is notDataset · publicMore than 5000 banana leaf images were acquired from different areas of Arba Minch
Zuria. Five plant pathologists meticulously verified the image classifications twice to
maintain accuracy. Code and dataset is publicly available on GitHub.Open asset ↗GitHubpdf-page:3 lines:1-44Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Banana is an important cash and food crop worldwide. Recent outbreaks of banana diseases are threatening the global banana industry and smallholder livelihoods. Remote sensing data offer the potential to detect the presence of disease, but there is a need for formal analysis to compare inferred with observed disease data. Here we use Landsat-8 data to investigate the detection of two banana diseases: banana bunchy top disease (BBTD) and Fusarium wilt Tropical Race 4 (TR4). We use satellite imagery to develop meteorology-driven predictive models for vegetation phenology, specifically based on healthy crops. Machine learning is then applied to identify anomalies associated with diseased plants by comparing the predicted vegetation indices of healthy crops with the observed indices from published data when disease is present. Our results show a correlation between changes in vegetation indices and the number of infected cases, highlighting the potential of this approach for large-scale disease surveillance.
Why it matches plant phenotyping methods衛星画像と機械学習により、バナナの病害に伴う植生指数の異常を推定する手法を開発・適用しており、感染植物の状態を大規模に評価することが中心です。
abstractWe use satellite imagery to develop meteorology-driven predictive models for vegetation phenology, specifically based on healthy crops.
Reproduction assets foundThe paper's Data Availability Statement explicitly deposits analysis code and data in a public GitHub repository by the authors, covering the remote-sensing/ML phenotyping analysis.Code · publicn, data collection and analysis, decision to publish, or preparation of the
manuscript. We thank Australian Banana Growers’ Council (ABGC) and banana inspectors for the
original data collecting.
Data Availability Statement: Analysis code and data used in this study can be accessed at the
following URL: (accessed on 3 July 2025) https://github.com/rretkute/BananaDiseasesRS.
Conflicts of Interest: The authors declare no conflicts of interest.
References
1. Voora, V.; Larrea, C.; Bermudez, S. Global Market Report: Bananas; International Institute for Sustainable Development:
Winnipeg, MB, Canada, 2020.
2. Ploetz, R.C. Management of Fusarium wilt of banana: A review with special reference to troOpen asset ↗rretkute/BananaDiseasesRS · rretkute/BananaDiseasesRSpdf-layout-page:15 lines:1-58Code / dataset availability confirmedEurope PMC · checked 6 Sept 2026
Chlorophyll levels are a key indicator of plant nitrogen status, which plays a critical role in optimizing agricultural yields. This study evaluated the performance of three low-cost multi-spectral sensors, AS7262, AS7263, and AS7265x, for non-destructive chlorophyll measurement. Measurements were taken from a diverse set of five leaf types, including smooth, uniform leaves (banana and mango), textured leaves (jasmine and sugarcane), and narrow leaves (rice). Partial least squares regression models were used to fit sensor spectra to chlorophyll levels, using nested cross-validation to ensure robust model evaluation. Sensor performance was assessed using R2 and mean absolute error (MAE) scores. The AS7265x demonstrated the best performance on smooth, uniform leaves with validation R2 scores of 0.96-0.95. Its performance decreased for the other leaves, with R2 scores of 0.75-0.85. The AS7262 and AS7263 sensors, while slightly less accurate, achieved reasonable R2 scores ranging from 0.93 to 0.86 for smooth leaves, and from 0.85 to 0.73 for the other leaves. All sensors, particularly the AS7265x, show potential for non-destructive chlorophyll measurement in agricultural applications. Their low cost and reasonable accuracy make them suitable for agricultural applications such as monitoring plant nitrogen levels.
Why it matches plant phenotyping methods低コストマルチスペクトルセンサーによる葉のクロロフィル測定法を評価・比較し、交差検証で性能を検証しているため、植物フェノタイピング手法が中心です。
abstractThis study evaluated the performance of three low-cost multi-spectral sensors, AS7262, AS7263, and AS7265x, for non-destructive chlorophyll measurement.
Reproduction assets foundThe authors publicly release raw sensor data, analysis scripts, firmware, and GUI in the GitHub repository KyleLopin/asm_chloro_test, plus supplementary information including extracted chlorophyll reference measurements (S2) at the MDPI supplement URL.Code · publicRaw data, scripts to generate the data and figures used in the manuscript, programs to run the sensors, and GUI used to collect the data are available at https://github.com/KyleLopin/asm_chloro_test (accessed on 25 March 2025).Open asset ↗KyleLopin/asm_chloro_test · KyleLopin/asm_chloro_testlines:187-200Code · publicThe microcontroller code to operate the sensor and a GUI for data collection are available at https://github.com/KyleLopin/asm_chloro_test/tree/master/source (accessed on 25 March 2025).Open asset ↗KyleLopin/asm_chloro_test · KyleLopin/asm_chloro_testlines:155-167Supplement · publicThe following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/s25072198/s1 . Supplementary Information S1: Device Electrical Characterization. Supplementary Information S2: Extracted Chlorophyll Reference Measurements.Open asset ↗lines:176-186Code / dataset availability confirmedEurope PMC · checked 6 Sept 2026
Radopholus similis severely damages banana roots causing significant yield losses. Field screening for resistance is labor intensive and inconsistent due to environmental variation and mixed nematode populations. The screenhouse offers a controlled environment but is limited by the time needed for root development and variation in plant growth. We developed and validated a high-throughput in vitro method for phenotyping banana resistance to R. similis using sand-Murashige and Skoog (MS) media. Tissue culture plantlets grown in sterilized sand-MS were inoculated with 50 female R. similis after root development and nematodes extracted eight weeks after inoculation to calculate the reproduction factor (RF). Although RF values were higher for in vitro than in the screenhouse, accession responses showed similar trends under both conditions. The in vitro method was rapid, cost-effective with higher throughput, accelerating phenotyping and enabling rapid assessment of banana accessions for breeding programs. Some accessions responded differently to the two methods indicating that additional methods, such as root necrosis scores are important to confirm resistance. This study is the first in vitro-based demonstration of phenotyping for nematode resistance using modified sand-MS media with improved root development and pathogen interactions.
Why it matches plant phenotyping methodsバナナの線虫抵抗性という植物状態を評価する高スループットin vitroフェノタイピング法を開発・検証し、既存のスクリーンハウス法と比較しているため、方法が研究の中心である。
abstractWe developed and validated a high-throughput in vitro method for phenotyping banana resistance to R. similis using sand-Murashige and Skoog (MS) media.
Reproduction assets foundThe paper's Data Availability statement explicitly deposits the plant entry list, raw datasets, and generated/analyzed datasets (Extended data) for this banana R. similis resistance phenotyping study on Figshare under CC-BY 4.0, matching an allowed URL. No author analysis code was deposited.Dataset · publicThe list of all plant entries, raw datasets, and datasets generated during and/or analyzed during the current study (Extended data) referred to in the manuscript text as supplementary materials are publicly available in Figshare: High-throughput resistance phenotyping of banana ( Musa spp.) against Radopholus similis . https://doi.org/10.6084/m9.figshare.28787480.v3 . The dataset has a CC-BY 4.0 license applied.Open asset ↗Figshare · 10.6084/m9.figshare.28787480.v3lines:139-144Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Plant leaf diseases are a significant concern in agriculture due to their detrimental impact on crop productivity and food security. Effective disease management depends on the early and accurate detection and diagnosis of these conditions, facilitating timely intervention and mitigation strategies. In this study, we address the pressing need for accurate and efficient methods for detecting leaf diseases by introducing a new architecture called DenseNet201Plus. DenseNet201 was modified by including superior data augmentation and pre-processing techniques, an attention-based transition mechanism, multiple attention modules, and dense blocks. These modifications enhance the robustness and accuracy of the proposed DenseNet201Plus model in diagnosing diseases related to plant leaves. The proposed architecture was trained using two distinct datasets: Banana Leaf Disease and Black Gram Leaf Disease. Through extensive experimentation, we evaluated the performance of DenseNet201Plus in terms of various classification metrics and achieved values of 0.9012, 0.9012, 0.9012, and 0.9716 for accuracy, precision, recall, and AUC for the banana leaf disease dataset, respectively. Similarly, the black gram leaf disease dataset model provides values of 0.9950, 0.9950, 0.9950, and 1.0 for accuracy, precision, recall, and AUC. Compared to other well-known pre-trained convolutional neural network (CNN) architectures, our proposed model demonstrates superior performance in both utilized datasets. Last but not least, we combined the strength of Grad-CAM++ with our proposed model to enhance the interpretability and localization of disease areas, providing valuable insights for agricultural practitioners and researchers to make informed decisions and optimize disease management strategies.
Why it matches plant phenotyping methods植物葉の病害領域を画像から分類・局在化する深層学習手法を開発し、複数データセットで性能評価しているため、病害状態のフェノタイピング手法が中心です。
abstractwe address the pressing need for accurate and efficient methods for detecting leaf diseases by introducing a new architecture called DenseNet201Plus.
Reproduction assets foundThe paper used two publicly available Mendeley leaf image datasets (banana leaf disease and black gram leaf disease) as its phenotyping inputs; both have explicit public URLs in the Data availability statement. No author analysis code or trained model was deposited.Dataset · public+: generalized gradient-based visual explanations for deep convolutional networks
2018 IEEE Winter Conference on Applications of Computer Vision (WACV) 2018
IEEE
839 847
Data availability
This study utilized publicly accessible datasets for analysis, which can be accessed at the following links: the Banana leaf disease dataset: https://data.mendeley.com/datasets/rjykr62kdh/1
[30] and the Black Gram leaf disease dataset: https://data.mendeley.com/datasets/zfcv9fmrgv/3
[31] .Open asset ↗lines:560-673Dataset · publicnter Conference on Applications of Computer Vision (WACV) 2018
IEEE
839 847
Data availability
This study utilized publicly accessible datasets for analysis, which can be accessed at the following links: the Banana leaf disease dataset: https://data.mendeley.com/datasets/rjykr62kdh/1
[30] and the Black Gram leaf disease dataset: https://data.mendeley.com/datasets/zfcv9fmrgv/3
[31] .Open asset ↗lines:560-673Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
During the 1950s, the Gros Michel species of bananas were nearly wiped out by the incurable Fusarium Wilt, also known as Panama Disease. Originating in Southeast Asia, Fusarium Wilt is a banana pandemic that has been threatening the multi-billion-dollar banana industry worldwide. The disease is caused by a fungus that spreads rapidly throughout the soil and into the roots of banana plants. Currently, the only way to stop the spread of this disease is for farmers to manually inspect and remove infected plants as quickly as possible, whereas it is a time-consuming process. The main purpose of this study is to build a deep Convolutional Neural Network (CNN) using a transfer learning approach to rapidly identify fusarium wilt infections on banana crop leaves. We chose to use the ResNet50 architecture as the base CNN model for our transfer learning approach owing to its remarkable performance in image classification, which was demonstrated through its victory in the ImageNet competition. After its initial training and fine-tuning on a data set consisting of 300 healthy and diseased images, the CNN model achieved near-perfect accuracy of 0.99 and was fine-tuned to adapt the ResNet base model. ResNet50’s distinctive residual block structure could be the reason behind these results. To evaluate this CNN model, 500 test images, consisting of 250 diseased and healthy banana leaf images, were classified by the model. The deep CNN model was able to achieve an accuracy of 0.98 and an F-1 score of 0.98 by correctly identifying the class of 492 of the 500 images. These results show that this DCNN model outperforms existing models such as Sangeetha et al., 2023’s deep CNN model by at least 0.07 in accuracy and is a viable option for identifying Fusarium Wilt in banana crops.
Why it matches plant phenotyping methodsバナナ葉の画像からFusarium萎凋病の感染状態をCNNで推定する手法を開発・評価しており、植物の病徴状態の取得が研究の中心である。
abstractThe main purpose of this study is to build a deep Convolutional Neural Network (CNN) using a transfer learning approach to rapidly identify fusarium wilt infections on banana crop leaves.
Reproduction assets foundThe paper publicly shares the banana leaf image dataset used to evaluate its Fusarium wilt detection CNN via a Google Drive link in the Data Availability Statement. No author analysis code or trained model checkpoints are explicitly deposited; other URLs are generic libraries or cited prior work.Dataset · publicve Program, Agriculture Economics and Rural Communities, under grant no. 2023-69006-40213.
Data Availability Statement: This study utilizes a modified data set from a previous work (Medhi and Deb, 2022)
for training neural networks. The data set used in evaluating the trained neural network is accessible via the
following link: https://drive.google.com/drive/folders/17wONO9e_goGjfTl-wSJp1SgIEthxzB1l?usp=share_link.
Supplementary information and any additional data not included in the main manuscript can be obtained from
the corresponding author upon reasonable request.
Conflicts of Interest: The authors have no known competing financial or non-financial interests that are directly
or indirecOpen asset ↗pdf-layout-page:11 lines:1-61Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Banana is among major crops cultivated by most smallholder farmers in Tanzania and other parts of Africa. This crop is very important in the household economy as well as food security since it serves as both food and cash crops. Despite these benefits, the majority of smallholder farmers are experiencing low yields which are attributed to diseases. The most problematic diseases are Black Sigatoka and Fusarium Wilt Race 1. Black Sigatoka is a disease that produces spots on the leaves of bananas and is caused by an air-borne fungus called Pseudocercospora fijiensis , formerly known as Mycosphaerella fijiensis . Fusarium Wilt Race 1 disease is one of the most destructive banana diseases that is caused by a soil-borne fungus called Fusarium oxysporum f.sp. Cubense (Foc). The dataset of curated banana crop image is presented in this article. Images of both healthy and diseased banana leaves and stems were taken in Tanzania and are included in the dataset. Smartphone cameras were used to take pictures of the banana leaves and stems. The dataset is the largest publicly accessible dataset for banana leaves and stems and includes 16,092 images. The dataset is significant and can be used to develop machine learning models for early detection of diseases affecting bananas. This dataset can be used for a number of computer vision applications, including object detection, classification, and image segmentation. The motivation for generating this dataset is to contribute to developing machine learning tools and spur innovations that will help to address the issue of crop diseases and help to eradicate the problem of food security in Africa.
Why it matches plant phenotyping methodsバナナの健全・罹病状態を画像で記録した公開データセットが論文の中心であり、植物病害状態の画像ベース表現型解析に利用できる。
abstractThe dataset of curated banana crop image is presented in this article.
Reproduction assets foundThe paper's banana leaf/stem image dataset (16,092 images) is publicly deposited on Harvard Dataverse (doi:10.7910/DVN/LQUWXW), and annotation was done with the Makerere AI Lab public web annotation tool on GitHub. Both are paper-specific, public, and actionable.Dataset · publiccation
• Institution: The Nelson Mandela African Institution of Science and Technology (NM-AIST), The International Institute of Tropical Agriculture (IITA)
• City/Town/Region: Arusha
• Country: Tanzania
Data accessibility
Repository name: Harvard Dataverse
Data identification number: doi: 10.7910/DVN/LQUWXW
Direct URL to data: https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/LQUWXW
Value of the Data
•
Machine learning models for early detection of Black Sigatoka and Fusarium Wilt Race 1 diseases that affect productivity can be trained using this dataset.
•
Researchers in the field of machine learning can use the collected imagery dataset of bananas to develop the endOpen asset ↗Harvard Dataverse · doi:10.7910/DVN/LQUWXWlines:1-53Code · publict and Remove Duplicate Pictures
2023 https://visipics.en.softonic.com (Accessed 10 February 2023)
3
Chen Q. Zobel J. Zhang X. Verspoor K.
Supervised learning for detection of duplicates in genomic sequence databases
PLoS ONE 11 2016 1 15
10.1371/journal.pone.0159644
PMC4973881
27489953
4
Makerere AI Lab
Web Annotation Tool
2023 https://github.com/AI-Lab-Makerere/web-annotation-tool (Accessed 5 March 2023)
5
Mduma N. Leo J. Loyani L. Jomanga K. Kamara A. Msaki I. Sanga S.
Banana Dataset Tanzania, Havard Dataverse
2022
10.7910/DVN/LQUWXW
Data Availability
Bananas Dataset Tanzania (Original data) (Dataverse).
Acknowledgments
The authors would like to extend their gratitude to Rockefeller FoundaOpen asset ↗github.com/AI-Lab-Makerere/web-annotation-toollines:89-126Code / dataset availability confirmedEurope PMC · checked 8 Sept 2026
Current chemical methods used to control plant diseases cause a negative impact on the environment and increase production costs. Accurate and early detection is vital for designing effective protection strategies for crops. We evaluate advanced distributed edge intelligence techniques with distinct learning principles for early black sigatoka disease detection using hyperspectral imaging. We discuss the learning features of the techniques used, which will help researchers improve their understanding of the required data conditions and identify a method suitable for their research needs. A set of hyperspectral images of banana leaves inoculated with a conidial suspension of black sigatoka fungus ( Pseudocercospora fijiensis ) was used to train and validate machine learning models. Support vector machine (SVM), multilayer perceptron (MLP), neural networks, N-way partial least square-discriminant analysis (NPLS-DA), and partial least square-penalized logistic regression (PLS-PLR) were selected due to their high predictive power. The metrics of AUC, precision, sensitivity, prediction, and F1 were used for the models' evaluation. The experimental results show that the PLS-PLR, SVM, and MLP models allow for the successful detection of black sigatoka disease with high accuracy, which positions them as robust and highly reliable HSI classification methods for the early detection of plant disease and can be used to assess chemical and biological control of phytopathogens.
Why it matches plant phenotyping methodsバナナ葉の病徴をハイパースペクトル画像と複数の機械学習モデルで早期検出し、モデル性能を評価する手法研究であり、植物病害状態の取得・推定が中心です。
abstractWe evaluate advanced distributed edge intelligence techniques with distinct learning principles for early black sigatoka disease detection using hyperspectral imaging.
Reproduction assets foundThe paper's hyperspectral banana-leaf training/validation datasets and the authors' analysis source code (PLS-PLR, NPLS-DA, SVM, MLP) are explicitly stated as publicly available on the authors' GitHub repository.Code · publicThe source programs are available at the following link: https://github.com/JUG2019/Sigatoka-detect (accessed on 21 August 2022).Open asset ↗JUG2019/Sigatoka-detectlines:34-65Dataset · publicThe two datasets used in this study (i.e., the training dataset and validation dataset are available at: https://github.com/JUG2019/Sigatoka-detect (accessed on 21 August 2022).Open asset ↗JUG2019/Sigatoka-detectlines:234-247Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
In recent times, the classification and identification of different fruits and food crops have become a necessity in the field of agricultural science; for sustainable growth. Probable processes have been developed worldwide to improve the production of food crops. Problem-specific, clean and crisp datasets are also lagging in the sector. This article introduces an image dataset of varieties of banana plants and the diseases related to them. The varieties of Banana plants that we have considered in the dataset are the Malbhog ( Musa assamica ), Jahaji ( Musa chinensis ), Kachkol ( Musa paradisiaca L. ), Bhimkol ( M. Balbisiana Colla ). And the diseases and pathogens that we have considered here are the Bacterial Soft Rot, Banana Fruit Scarring Beetle, Black Sigatoka, Yellow Sigatoka, Panama disease, Banana Aphids, and Pseudo-Stem Weevil. A dataset of Potassium deficiency has been also considered in this article. A total of 8000+ processed images are present in the dataset. The purpose of this article is to provide the Researchers and Students in getting access to our dataset that would help them in their research and in developing some machine learning models.
Why it matches plant phenotyping methodsバナナ植物の器官・品種・病害・カリウム欠乏を画像化した大規模データセットを提供しており、植物の状態推定に再利用できるデータ基盤が中心である。
abstractThis article introduces an image dataset of varieties of banana plants and the diseases related to them.
Reproduction assets foundThe paper is a data descriptor for the PSFD-Musa banana image dataset, publicly deposited on Mendeley Data with an explicit URL and DOI.Dataset · publics of different backgrounds to train, test, and validate classification models.
Data source location
•
BORTARI VILLAGE,
Chaygaon, Kukurmara, District – Kamrup (Rural), Assam, India.
•
HAJO VILLAGE,
District – Kamrup (Rural), Assam, India.
Data accessibility
Data is available at Mendeley Data, under the DOI: 10.17632/4wyymrcpyz.1 https://data.mendeley.com/datasets/4wyymrcpyz/1
Value of the Data
•
The dataset provided here is the collection of different varieties of banana plants, some common diseases that affect them, and their deficiency. These varieties of banana plants are indigenously found in Assam. The data can be useful in the way to classifying the different diseases and pathogens whicOpen asset ↗Mendeley Data · 10.17632/4wyymrcpyz.1lines:1-54Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
The existing methods of callose quantification include epifluorescence microscopy and fluorescence spectrophotometry of aniline blue-stained callose particles, immuno-fluorescence microscopy and indirect assessment of both callose synthase and β-(1,3)-glucanase enzyme activities. Some of these methods are laborious, time consuming, not callose-specific, biased and require high technical skills. Here, we describe a method of callose quantification based on Sandwich Enzyme-Linked Immunosorbent Assay (S-ELISA). Tissue culture-derived banana plantlets were inoculated with Xanthomonas campestris pv. musacearum ( Xcm ) bacteria as a biotic stress factor inducing callose production. Banana leaf, pseudostem and corm tissue samples were collected at 14 days post-inoculation (dpi) for callose quantification. Callose levels were significantly different in banana tissues of Xcm -inoculated and control groups except in the pseudostems of both banana genotypes. The method described here could be applied for the quantification of callose in different plant species with satisfactory level of specificity to callose, and reproducibility. Additionally, the use of 96-well plate makes this method suitable for high throughput callose quantification studies with minimal sampling and analysis biases. We provide step-by-step detailed descriptions of the method.
Why it matches plant phenotyping methods植物組織中のカロース量を定量するELISA法を開発・再現性評価し、高スループット測定への適用性を示した研究であり、植物状態の取得方法が中心です。
abstractHere, we describe a method of callose quantification based on Sandwich Enzyme-Linked Immunosorbent Assay (S-ELISA).
Reproduction assets foundThe paper's supplementary material (Table S1) publicly hosts the callose quantification measurements (concentrations in leaves, pseudostems, corms of Xcm-inoculated vs. control banana plantlets) underlying this study's analysis. No author analysis code, images, or trained models are deposited; the R statistical packageDataset · publicor up to 12 months).
Dissolve para-nitrophenyl phosphate (pNPP) in substrate buffer to a working concentration of 1 mg/mL.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Materials
The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/mps5040054/s1 , Table S1: Analysis of callose concentration in the leaves, pseudostems and corms of banana plants inoculated and non-inoculated (control) with Xcm (Independent sample t-test, α ≤ 0.05).
Click here for additional data file.
Author Contributions
Conceptualization, A.K.T.; methodology, A.S.M., A.K.T. and P.S.; validatiOpen asset ↗lines:167-297Code / dataset availability confirmedCrossref · checked 9 Sept 2026
Solving many phenotyping problems involves not only automatic detection of objects in an image, but also counting the number of parts per object. We propose a solution in the form of a single deep network, tested for three agricultural datasets pertaining to bananas-per-bunch, spikelets-per-wheat-spike, and berries-per-grape-cluster. The suggested network incorporates object detection, object resizing, and part counting as modules in a single deep network, with several variants tested. The detection module is based on a Retina-Net architecture, whereas for the counting modules, two different architectures are examined: the first based on direct regression of the predicted count, and the other on explicit parts detection and counting. The results are promising, with the mean relative deviation between estimated and visible part count in the range of 9.2% to 11.5%. Further inference of count-based yield related statistics is considered. For banana bunches, the actual banana count (including occluded bananas) is inferred from the count of visible bananas. For spikelets-per-wheat-spike, robust estimation methods are employed to get the average spikelet count across the field, which is an effective yield estimator.
Why it matches plant phenotyping methods植物器官の可視パーツ数を画像から検出・計数する深層学習手法を開発し、複数作物データセットで評価しているため、表現型取得・推定法が中心です。
abstractWe propose a solution in the form of a single deep network, tested for three agricultural datasets pertaining to bananas-per-bunch, spikelets-per-wheat-spike, and berries-per-grape-cluster.
Reproduction assets foundThe paper's grape experiments use the public Embrapa WGISD dataset, extended by the authors with berry dot annotations that they state were made publicly available as part of that dataset extension. The banana and wheat datasets (Israel Phenomics Consortium) and the authors' code/models have no stated public release,; Dataset · publicThe dot annotations were made
publicly available as part of Embrapa WGISD dataset extension.Open asset ↗Embrapa WGISDpdf-page:5 lines:1-59Code / dataset availability confirmedOpenAlex · Europe PMC · checked 8 Sept 2026
Leaf counting in potted plants is an important building block for estimating their health status and growth rate and has obtained increasing attention from the visual phenotyping community in recent years. Two novel deep learning approaches for visual leaf counting tasks are proposed, evaluated, and compared in this study. The first method performs counting via direct regression but using multiple image representation resolutions to attend leaves of multiple scales. The leaf count from multiple resolutions is fused using a novel technique to get the final count. The second method is detection with a regression model that counts the leaves after locating leaf center points and aggregating them. The algorithms are evaluated on the Leaf Counting Challenge (LCC) dataset of the Computer Vision Problems in Plant Phenotyping (CVPPP) conference 2017, and a new larger dataset of banana leaves. Experimental results show that both methods outperform previous CVPPP LCC challenge winners, based on the challenge evaluation metrics, and place this study as the state of the art in leaf counting. The detection with regression method is found to be preferable for larger datasets when the center-dot annotation is available, and it also enables leaf center localization with a 0.94 average precision. When such annotations are not available, the multiple scale regression model is a good option.
Why it matches plant phenotyping methods植物画像から葉数を推定する2つの深層学習手法を開発・比較し、既存および新規データセットで評価しているため、表現型取得手法が中心である。
abstractTwo novel deep learning approaches for visual leaf counting tasks are proposed, evaluated, and compared in this study.
Reproduction assets foundThe paper's authors explicitly state their full leaf-counting code (MSR and DRN models) is freely available on GitHub. The banana leaf dataset is proprietary and not public; the LCC datasets are external community benchmarks, not paper-specific assets.Code · publicThe entire code is freely accessible at https://github.com/farjon/Leaf-Counting .Open asset ↗farjon/Leaf-Countinglines:300-310Code / dataset availability confirmedOpenAlex · Europe PMC · checked 8 Sept 2026
3D reconstruction of fruit is important as a key component of fruit grading and an important part of many size estimation pipelines.Like many computer vision challenges, the 3D reconstruction task suffers from a lack of readily available training data in most domains, with methods typically depending on large datasets of high-quality image-model pairs.In this paper, we propose an unsupervised domain-adaptation approach to 3D reconstruction where labelled images only exist in our source synthetic domain, and training is supplemented with different unlabelled datasets from the target real domain.We approach the problem of 3D reconstruction using volumetric regression and produce a training set of 25,000 pairs of images and volumes using hand-crafted 3D models of bananas rendered in a 3D modelling environment (Blender).Each image is then enhanced by a GAN to more closely match the domain of photographs of real images by introducing a volumetric consistency loss, improving performance of 3D reconstruction on real images.Our solution harnesses the cost benefits of synthetic data while still maintaining good performance on real world images.We focus this work on the task of 3D banana reconstruction from a single image, representing a common task in plant phenotyping, but this approach is general and may be adapted to any 3D reconstruction task including other plant species and organs.
Why it matches plant phenotyping methods果実の3D再構成と体積回帰を対象とする教師なしドメイン適応手法を開発しており、植物器官の形態・サイズ推定に用いるフェノタイピング手法が研究の中心である。
abstractIn this paper, we propose an unsupervised domain-adaptation approach to 3D reconstruction
Reproduction assets foundThe paper's synthetic banana image-volume dataset (25,000 image-volume pairs) is publicly deposited at the authors' project site, and the pipeline/training code is deposited on the authors' GitHub. Both are paper-specific, public, and actionable.Code · publicThe code used to create the dataset for this study has been deposited on github at https://github.com/zanehartley . The code for the neural networks used for this study has been deposited on github at https://github.com/zanehartley .Open asset ↗github.com/zanehartleylines:158-160Code / dataset availability confirmedCrossref · Europe PMC · checked 15 Sept 2026
The production of banana-one of the highly consumed fruits-is highly affected due to loss of certain number of banana plants in an early phase of vegetation. This affects the ability of farmers to forecast and estimate the production of banana. In this paper, we propose a deep learning (DL) based method to precisely detect and count banana plants on a farm exclusive of other plants, using high resolution RGB aerial images collected from Unmanned Aerial Vehicle (UAV). An attempt to detect the plants on the normal RGB images resulted less than 78.8% recall for our sample images of a commercial banana farm in Thailand. To improve this result, we use three image processing methods-Linear Contrast Stretch, Synthetic Color Transform and Triangular Greenness Index-to enhance the vegetative properties of orthomosaic, generating multiple variants of orthomosaic. Then we separately train a parameter-optimized Convolutional Neural Network (CNN) on manually interpreted banana plant samples seen on each image variants, to produce multiple results of detection on our region of interest. 96.4%, 85.1% and 75.8% of plants were correctly detected on three of our dataset collected from multiple altitude of 40, 50 and 60 meters, of same farm. Further discussion on results obtained from combination of multiple altitude variants are also discussed later in the research, in an attempt to find better altitude combination for data collection from UAV for the detection of banana plants. The results showed that merging the detection results of 40 and 50 meter dataset could detect the plants missed by each other, increasing recall upto 99%.
Why it matches plant phenotyping methodsUAV RGB画像からバナナ個体を検出・計数する画像解析手法の開発と性能評価が研究の中心であり、植物個体数という観測可能な植物形質を抽出している。
abstractwe propose a deep learning (DL) based method to precisely detect and count banana plants on a farm exclusive of other plants, using high resolution RGB aerial images collected from Unmanned Aerial Vehicle (UAV).
Reproduction assets foundThe paper's Data Availability statement deposits the paper-specific UAV-collected banana plant image datasets (the phenotyping inputs used for detection/counting) in a public Figshare repository with an authors' URL, making it directly actionable.Dataset · publichas-pdf yes pmc-prop-has-supplement no pmc-prop-pdf-only no pmc-prop-suppress-copyright no pmc-prop-is-real-version no pmc-prop-is-scanned-article no pmc-prop-preprint no pmc-prop-in-epmc yes pmc-license-ref CC BY Data Availability Data available from the Figshare Repository (DOIs: 10.6084/m9.figshare.7981547 ), and the URL is: https://figshare.com/s/62e391492b1be99515b4 .
Data Availability
Data available from the Figshare Repository (DOIs: 10.6084/m9.figshare.7981547 ), and the URL is: https://figshare.com/s/62e391492b1be99515b4 .
Introduction
Significance of counting banana plantsOpen asset ↗Figshare · 10.6084/m9.figshare.7981547lines:1-36Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Background The development of leaf area is one of the fundamental variables to quantify plant growth and physiological function and is therefore widely used to characterize genotypes and their interaction with the environment. To date, analysis of leaf area often requires elaborate and destructive measurements or imaging-based methods accompanied by automation that may result in costly solutions. Consequently in recent years there is an increasing trend towards simple and affordable sensor solutions and methodologies. A major focus is currently on harnessing the potential of applications developed for smartphones that provide access to analysis tools to a wide user basis. However, most existing applications entail significant manual effort during data acquisition and analysis. Results With the development of Plant Screen Mobile we provide a suitable smartphone solution for estimating digital proxies of leaf area and biomass in various imaging scenarios in the lab, greenhouse and in the field. To distinguish between plant tissue and background the core of the application comprises different classification approaches that can be parametrized by users delivering results on-the-fly. We demonstrate the practical applications of computing projected leaf area based on two case studies with Eragrostis and Musa plants. These studies showed highly significant correlations with destructive measurements of leaf area and biomass from both ground truth measurements and estimations from well-established screening systems. Conclusions We show that a smartphone together with our analysis tool Plant Screen Mobile is a suitable platform for rapid quantification of leaf and shoot development of various plant architectures. Beyond the estimation of projected leaf area the app can also be used to quantify color and shape parameters of other plant material including seeds and flowers.
Why it matches plant phenotyping methodsスマートフォン画像と分類アルゴリズムによる葉面積・バイオマス推定アプリを開発し、破壊測定および既存システムと検証しており、植物表現型取得が中心である。
abstractWith the development of Plant Screen Mobile we provide a suitable smartphone solution for estimating digital proxies of leaf area and biomass in various imaging scenarios in the lab, greenhouse and in the field.
Reproduction assets foundThe authors deposited the plant image data and corresponding ground truth measurements from the banana and Eragrostis case studies in the e!DAL research data publication system (DOI 10.25622/FZJ/2018/1), a paper-specific public asset. The project homepage (fz-juelich.de/ibg/ibg-2/psm) hosts the app and manual but is a Dataset · publichave no competing interests.
Availability of data and materials
The app is accompanied by a detailed manual and checkerboard images for calibration, which can also be downloaded from the project homepage. The datasets generated and/or analyzed during the current study are available in the e!DAL research data publication system, http://dx.doi.org/10.25622/FZJ/2018/1 [ 30 ].
Availability and requirements
Project name: Plant Screen Mobile. Project home page: https://fz-juelich.de/ibg/ibg-2/psm . Operating system(s): Android OS 4.0 (Ice Cream Sandwich) or higher. Programming language: Java. Other requirements: OpenCV manager (will be installed during Plant Screen Mobile setup). License: GNU GOpen asset ↗e!DAL research data publication system · 10.25622/FZJ/2018/1lines:127-198