Pith. sign in

REVIEW 3 major objections 4 minor 50 references

DExNet: Combining Observations of Domain Adapted Critics for Leaf Disease Classification with Limited Data

T0 review · 3 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read DExNet claims nine domain-adapted CNN critics, a fused feature vector, and a Bi-LSTM classifier push few-shot tomato leaf disease classification to 89–94% accuracy at 5–15 shots, and 98.09% at 80 shots, rivaling fully supervised training…

desk verdict Clean tomato split, broken no-overlap promise in the non-tomato benchmarks; the abstract overclaims. read the letter →

arxiv 2506.18173 v2 pith:33YQQTBO submitted 2025-06-22 cs.CV

Sabbir Ahmed , Md. Bakhtiar Hasan , Tasnim Ahmed , Md. Hasanul Kabir This is my paper
classification cs.CV
keywords few-shotlearningplantdiseaseclassificationdomainadaptationfeaturefusionBi-LSTMensembleofCNNsVillageleaf
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

DExNet is a few-shot learning pipeline for classifying plant leaf diseases when only a handful of labeled images are available per disease. The paper's central claim is that fine-tuning nine standard pre-trained CNN feature extractors on a large source set of leaf images from other crops, then concatenating their feature vectors and passing them through a Bi-LSTM classifier, yields high accuracy on entirely unseen target classes with very few training samples. On the 10 tomato classes of PlantVillage, it reports 89.06%, 92.46%, and 94.07% accuracy for 5-, 10-, and 15-shot classification, and 98.09% for 80-shot classification, within about one point of fully supervised methods while using roughly 20 times less data. The authors also report gains over existing few-shot baselines in mixed-domain, cross-domain, and field-image settings. If the claim holds, it would mean that a pretrained ensemble plus domain adaptation can substitute for large labeled disease datasets in practical agricultural diagnosis.

What carries the argument

The pipeline has three blocks. The Domain Adaptation Block fine-tunes nine pre-trained CNNs—ResNet-18/34/50/101/152 and DenseNet-121/161/169/201—on the 28 non-tomato classes of PlantVillage, none of which overlap with the target classes. The Feature Fusion Block concatenates the nine models' embeddings for each image into a single 13,985-dimensional observation vector. The Classifier Block passes that vector through a 1024-unit Bi-LSTM followed by a fully connected layer; the bidirectional LSTM is the component the paper credits with capturing cross-feature dependencies within the concatenated representation.

What would settle it

Train DExNet with a source set that contains a disease shared with the target (e.g., early blight from potato) and then evaluate on the same disease in the tomato target, while also evaluating a version where that shared disease is removed from the source; if removing the shared class collapses accuracy, the reported few-shot gains come from label overlap, not from generic expertise. A second decisive test would apply the pipeline to a target crop whose diseases are absent from the PlantVillage classes entirely and measure accuracy against chance.

Watch

Extended reading notes

Core claim

The central discovery is that combining several 'domain-adapted critics' works better than any single feature extractor for few-shot leaf disease classification. Each critic is a pre-trained CNN (five ResNet variants and four DenseNet variants) fine-tuned on the 28 non-tomato PlantVillage classes at the meta-training stage. The features the critics emit for a query image—called observations—are concatenated into a single high-dimensional vector and classified by a bidirectional LSTM. In the paper's framing, the domain adaptation supplies general knowledge about what healthy and diseased leaves look like, the ensemble supplies diverse feature perspectives, and the Bi-LSTM models the interactions among those perspectives, so that a few support samples suffice. Across the paper's experiments, the full 9-critic configuration with concatenated features and Bi-LSTM classifier outperforms each individual backbone and the no-domain-adaptation versions, with the gap narrowing as the number of shots grows.

Load-bearing premise

The pipeline assumes that fine-tuning on the 28 non-tomato PlantVillage classes teaches genuinely transferable leaf-disease knowledge rather than exploiting shared class labels or identical imaging conditions with the tomato target classes.

Editorial extensions

If this is right

  • If replicated, the pipeline could bring plant disease diagnosis to regions where labeled disease images are scarce, because the source set used for domain adaptation is already public and the target requires only a few images per disease.
  • The reported 80-shot result, 98.09%, about 1.2% below fully supervised benchmarks with a 94.5% reduction in training data, suggests that near-full performance can be reached with 20 times fewer samples.
  • The consistent gains from ensembling over single backbones imply that feature diversity, not any one architecture, is the main driver of few-shot accuracy in this setting.
  • Strong cross-domain results (leaf classes tested after training only on pest classes) imply the learned representations are not tied to one image style, making the method a candidate for field deployment.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • An open question the paper does not directly settle is whether the source and target sets overlap in disease labels: PlantVillage contains the same disease name (e.g., early blight) across several crops, so a cross-check that removes shared-disease classes would clarify whether the adaptation gain is from generic leaf-disease features or from category leakage.
  • Because the pipeline's gains over the no-adaptation baseline shrink as k grows, a reasonable extrapolation is that the domain adaptation is most valuable in the extreme few-shot regime (k <= 10), and that future work could focus the adaptation stage specifically on that regime.
  • The ensembling design treats each CNN as an independent observer; a lighter variant that distills the nine critics into a single network at deployment, or adapts a smaller subset of critics per target crop, could preserve much of the accuracy at lower computational cost, although the paper does not test this.
  • A natural testable extension would apply DExNet to genuinely novel disease categories absent from PlantVillage (e.g., wheat rust or cassava diseases) to see whether the domain-adaptation step transfers across crop families and image acquisition conditions.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper proposes DExNet, a few-shot learning pipeline for leaf disease classification that (i) fine-tunes nine pre-trained CNNs ('critics') on a source domain of 28 non-tomato PlantVillage classes, (ii) extracts and concatenates their feature embeddings ('observations') for target images, and (iii) classifies the fused representation with a Bi-LSTM. The main experiment is 10-class tomato leaf disease classification, reporting accuracies of 89.06%, 92.46%, 94.07% for 5-, 10-, and 15-shot, and 98.09 ± 0.7% for 80-shot. Additional experiments claim state-of-the-art performance on the Argueso benchmark (Table 5), on the 'Plants & Pest' single- and cross-domain tasks (Table 6), and on potato and cotton field datasets (Table 7).

Significance. If the results held under a clean evaluation protocol, DExNet would be a useful contribution to few-shot plant disease classification: the main tomato experiment is reported with error bars over 100 tasks, systematic ablations of feature extractors and classifiers are provided, and the idea of domain-adapting multiple critics has practical appeal. However, the significance strongly depends on the validity of the no-overlap assumption between the domain-adaptation set and the downstream target sets. The evidence for the paper's broad comparative claims is compromised by class leakage, so the significance is currently not established beyond the tomato-only setting.

major comments (3)
  1. [Sections 3 and 4.2; Tables 5 and 7] The no-overlap premise is violated for the non-tomato benchmarks. Section 3 states that meta-training and meta-testing classes are non-overlapping, and Section 4.2 defines the domain-adaptation set as the 28 non-tomato PlantVillage classes. However, Table 5 evaluates on the Argueso benchmark whose 6 target classes (apple scab, apple black rot, apple cedar rust, blueberry healthy, cherry powdery mildew, cherry healthy) are exactly among those 28 classes, and Table 7 evaluates on the Potato Leaf Disease dataset whose classes (early blight, late blight, healthy) are also included. The critics were therefore fine-tuned on images of the very diseases being classified, so the few-shot accuracies in Tables 5 and 7 do not test generalization to unseen classes. The abstract's claim of outperforming existing works in 'laboratory and real-life conditions' is unsupported as written.
  2. [Section 5.3; Table 4] The final reported accuracies are not independent of model selection. Section 5.3 compares Dense, LSTM, GRU, Bi-GRU, and Bi-LSTM classifiers using the full tomato test set (Q = Full), and Section 5.2 similarly uses the same test-set tasks to choose the concatenated-observation strategy. The chosen design (concatenated features + Bi-LSTM) is then re-evaluated on the same test set in Table 4. Because the architecture choices were tuned on the evaluation data, the headline numbers in Table 4 (e.g., 98.09% at 80-shot) are optimistically biased. A held-out validation split should be used for model selection, with the test set used only once for the final reported accuracies.
  3. [Table 5 and Section 5.5] The comparison with prior work is internally inconsistent and unfair in its current form. The table is captioned as a comparison on the PlantVillage dataset, yet the 'Ours' results (99.73–99.9%) are obtained on the 6-class apple/blueberry/cherry subset whose classes overlap with the domain-adaptation set, while the tomato-only result in Table 4 is 98.09% at 80-shot. The near-perfect accuracy in Table 5 is plausibly a direct consequence of the leakage, not of the method's few-shot capability. Moreover, Section 5.5 says the pipeline 'outperformed the works of [27,46]' when discussing Table 6, but Wang et al. [46] does not appear in Table 6; the comparison is with Li et al. and Nuthalapati et al. The claims of broad superiority over prior work therefore need to be re-established under a non-overlapping source/target split.
minor comments (4)
  1. [Table 2] The column header 'Domain Adopt' is likely 'Domain Adaptation'; the check/cross symbols would be clearer as 'Yes/No'.
  2. [Table 5] The '×' in the k=10 column for Wang et al. is unexplained; please state that the result was not reported for that setting.
  3. [Section 5.5] The text refers to 'the works of [27,46]' but Table 6 lists [27] and [33]; the citation should be adjusted to match the table.
  4. [General notation] Use consistent casing for 'k-shot' (e.g., 'k = 1' vs. 'K = 1') across tables and text.

Circularity Check

2 steps flagged · score 6.0 of 10

Non-tomato benchmarks reuse target classes in the domain-adaptation set, so Tables 5 and 7 (potato) are not unseen-class predictions.

  1. fitted input called prediction [Abstract; Section 4.2 (Domain Adaptation); Section 5.5 (Table 5)]
    "These critics are 'domain adapted' using a publicly available leaf disease dataset having no overlapping classes with the specific downstream task of interest. ... For this, we used the 28 classes of the PlantVillage dataset except for the samples of the tomato leaves. ... In this task, the target classes contained 6 leaf diseases of apple, blueberry, and cherry plants."

    The domain-adaptation set is fixed as all non-tomato PlantVillage classes, which include the apple, blueberry, and cherry disease classes used as the Argüeso benchmark target classes. The feature extractors were therefore fine-tuned on labeled images of the exact classes that Table 5 then reports as few-shot predictions. The stated no-overlap condition is violated, so the near-perfect accuracies are not evidence of generalization to unseen classes; they are evaluations on classes already present in the adaptation set.

  2. fitted input called prediction [Section 3 (Task Formulation); Section 4.2 (Domain Adaptation); Section 5.5 (Table 7)]
    "For this, we used the 28 classes of the PlantVillage dataset except for the samples of the tomato leaves. ... we conducted experiments on two publicly available datasets, 'Potato Leaf Disease Classification' [37] and 'Cotton Leaf Disease' [32], consisting of diversified samples collected from the field."

    The 28-class PlantVillage adaptation set includes the potato early blight, late blight, and healthy classes, and the field benchmark is a potato-leaf-disease dataset. The paper does not describe any change to the adaptation set for this experiment, so the potato results in Table 7 measure classes whose labeled PlantVillage images were used to fine-tune the feature extractors. The cotton half of the field evaluation is not affected, since cotton is not among the 28 PlantVillage classes used for adaptation.

full rationale

The main tomato experiments are not circular: the 10 tomato classes are withheld from the 28-class PlantVillage adaptation set, so the reported 5-shot, 10-shot, 15-shot, and 80-shot accuracies are genuine held-out evaluations of the proposed pipeline. However, the paper's broader claim of outperforming prior work in single-domain and field scenarios is partially undermined by class overlap in two external benchmarks. Section 4.2 defines the adaptation set once as all non-tomato PlantVillage classes, and Section 5.5 then evaluates the pipeline on an apple/blueberry/cherry disease benchmark and on a potato field dataset; both reuse disease classes that are included in that adaptation set. Because the feature extractors were fine-tuned on those exact classes, the few-shot accuracies in Tables 5 and the potato column of Table 7 reduce to recognition of previously seen classes rather than prediction of unseen classes. The Plants & Pest and cotton experiments do not share this particular overlap, and the tomato pipeline itself remains a substantial independent contribution, so the overall circularity is partial rather than total. Score 6 reflects that one or more predictions are forced by the construction of the training set, while the central tomato result retains independent content.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central empirical results rest mostly on standard transfer learning; no invented entities are introduced, but several design choices are fitted to the evaluation benchmark.

free parameters (3)
  • Bi-LSTM hidden units = 1024
    Selected by empirically comparing 256, 512, 1024, and 2048 units on the same evaluation tasks (Section 4.4), so the final accuracy is partially tuned to the benchmark.
  • Ensemble composition of nine critics = ResNet18/34/50/101/152 and DenseNet121/161/169/201
    The set of nine architectures was chosen because the all-nine ensemble outperformed individual models on the target tasks (Table 1); the choice is empirical, not principled.
  • Feature vector dimension after concatenation = 13,984
    Determined by the chosen critics and pooling layers; not independently selected.
assumptions (4)
  • domain assumption Features learned on 28 non-target PlantVillage classes transfer to the tomato classes.
    Section 4.2 states this as the motivation for domain adaptation; no independent evidence is given beyond the reported accuracy.
  • domain assumption Concatenation of nine frozen feature vectors is a sufficient fusion rule.
    Section 4.3 assumes complementary information is preserved by concatenation; no comparison to learned fusion is provided.
  • ad hoc to paper Bi-LSTM can model useful dependencies among arbitrarily ordered concatenated features.
    Section 5.3 argues recurrent networks capture spatial relationships, yet the feature vector has no intrinsic sequence ordering.
  • standard math Repeating random tasks 100 times makes average accuracy representative.
    Sections 3 and 5 use 100 tasks to estimate accuracy; the number of tasks is conventional but no confidence interval beyond the standard deviation is described.

how reviews work

0 comments
Cite this review

Pith. "Pith review of DExNet: Combining Observations of Domain Adapted Critics for Leaf Disease Classification with Limited Data." pith.science (2026). https://pith.science/paper/33YQQTBO

@misc{pith2026250618173,
  author       = {Pith},
  title        = {Pith review of: DExNet: Combining Observations of Domain Adapted Critics for Leaf Disease Classification with Limited Data},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/33YQQTBO}},
  note         = {Machine review of arXiv:2506.18173}
}
read the original abstract

While deep learning-based architectures have been widely used for correctly detecting and classifying plant diseases, they require large-scale datasets to learn generalized features and achieve state-of-the-art performance. This poses a challenge for such models to obtain satisfactory performance in classifying leaf diseases with limited samples. This work proposes a few-shot learning framework, Domain-adapted Expert Network (DExNet), for plant disease classification that compensates for the lack of sufficient training data by combining observations of a number of expert critics. It starts with extracting the feature embeddings as 'observations' from nine 'critics' that are state-of-the-art pre-trained CNN-based architectures. These critics are 'domain adapted' using a publicly available leaf disease dataset having no overlapping classes with the specific downstream task of interest. The observations are then passed to the 'Feature Fusion Block' and finally to a classifier network consisting of Bi-LSTM layers. The proposed pipeline is evaluated on the 10 classes of tomato leaf images from the PlantVillage dataset, achieving promising accuracies of 89.06%, 92.46%, and 94.07%, respectively, for 5-shot, 10-shot, and 15-shot classification. Furthermore, an accuracy of 98.09+-0.7% has been achieved in 80-shot classification, which is only 1.2% less than state-of-the-art, allowing a 94.5% reduction in the training data requirement. The proposed pipeline also outperforms existing works on leaf disease classification with limited data in both laboratory and real-life conditions in single-domain, mixed-domain, and cross-domain scenarios.

Figures

Figures reproduced from arXiv: 2506.18173 by the authors.

Figure 1
Figure 1. Overview of the Proposed Pipeline DExNet leaves. Although they are not trained on the target classes, their learned representations remain descriptive and informative, allowing the model to perform effectively even with very few target samples. DExNet comprises three main components: the Domain Adaptation Block (DAB), the Feature Fusion Block (FFB), and the Classifier Block (CB). In the DAB, we fine￾tune nine pre-tr… view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

50 extracted references · 46 canonical work pages

  1. [46]

    In: IEEE 21st International Conference on Bioinformatics and Bioengineering (BIBE)

    Wang, Y ., Wang, S.: Imal: An improved meta-learning approach for few-shot classification of plant diseases. In: IEEE 21st International Conference on Bioinformatics and Bioengineering (BIBE). pp. 1–7 (2021)

    work page 2021

  2. [1]

    Computers and Electronics in Agriculture 187, 106279 (2021)

    Abbas, A., Jain, S., Gour, M., Vankudothu, S.: Tomato plant disease detection using trans- fer learning with C-GAN synthetic images. Computers and Electronics in Agriculture 187, 106279 (2021)

    work page 2021

  3. [2]

    In: 2024 6th International Conference on Sustainable Technologies for Industry 5.0 (STI)

    Ahamed, M., Kabir, R.B., Dipto, T.T., Al Mushabbir, M., Ahmed, S., Kabir, M.H.: Perfor- mance analysis of few-shot learning approaches for bangla handwritten character and digit recognition. In: 2024 6th International Conference on Sustainable Technologies for Industry 5.0 (STI). pp. 1–6 (2024)

    work page 2024

  4. [3]

    Msc thesis, Department of Computer Science and Engineering (CSE), Islamic University of Technology (2022)

    Ahmed, S.: Classification of Plant Disease from Leaf Images Using Few-Shot Learning. Msc thesis, Department of Computer Science and Engineering (CSE), Islamic University of Technology (2022)

    work page 2022

  5. [4]

    IEEE Access 10, 68868–68884 (2022)

    Ahmed, S., Hasan, M.B., Ahmed, T., Sony, M.R.K., Kabir, M.H.: Less is more: Lighter and faster deep neural architecture for tomato leaf disease classification. IEEE Access 10, 68868–68884 (2022)

    work page 2022

  6. [5]

    In: 2024 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE)

    Ahmed, T., Hasan, M.B., Ahmed, S., Kabir, M.H.: Exe-net: Explainable ensemble network for potato leaf disease classification. In: 2024 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE). pp. 335–339 (2024)

    work page 2024

  7. [6]

    IEEE Access 10, 45001–45028 (2022)

    Alanezi, M.A., Shahriar, M.S., Hasan, M.B., Ahmed, S., Sha’aban, Y .A., Bouchekara, H.R.E.H.: Livestock management with unmanned aerial vehicles: A review. IEEE Access 10, 45001–45028 (2022)

    work page 2022

  8. [7]

    Computers and Electronics in Agriculture 175, 105542 (2020)

    Argüeso, D., Picon, A., Irusta, U., Medela, A., San-Emeterio, M.G., Bereciartua, A., Alvarez- Gila, A.: Few-shot learning approach for plant disease classification using images taken in the field. Computers and Electronics in Agriculture 175, 105542 (2020)

    work page 2020

Show all 50 references
  1. [8]

    IEEE Access 10, 92597–92632 (2022)

    Ashikur Rahman, A.B.M., Hasan, M.B., Ahmed, S., Ahmed, T., Ashmafee, M.H., Kabir, M.R., Kabir, M.H.: Two decades of bengali handwritten digit recognition: A survey. IEEE Access 10, 92597–92632 (2022)

    2022

  2. [9]

    In: In- ternational Conference on Electrical, Computer and Communication Engineering (ECCE)

    Ashmafee, M.H., Ahmed, T., Ahmed, S., Hasan, M.B., Jahan, M.N., Ashikur Rahman, A.: An efficient transfer learning-based approach for apple leaf disease classification. In: In- ternational Conference on Electrical, Computer and Communication Engineering (ECCE). pp. 1–6 (2023)

    2023

  3. [10]

    Sensors 21(6) (2021)

    Buja, I., Sabella, E., Monteduro, A.G., Chiriacò, M.S., De Bellis, L., Luvisi, A., Maruccio, G.: Advances in plant disease detection and monitoring: From traditional assays to in-field diagnostics. Sensors 21(6) (2021)

    2021

  4. [11]

    CoRR abs/1904.04232 (2019)

    Chen, W., Liu, Y ., Kira, Z., Wang, Y .F., Huang, J.: A closer look at few-shot classification. CoRR abs/1904.04232 (2019)

    2019 arXiv

  5. [12]

    In: 2024 27th Interna- tional Conference on Computer and Information Technology (ICCIT)

    Chowdhury, R.H., Ahmed, S.: Mangoleafvit: Leveraging lightweight vision transformer with runtime augmentation for efficient mango leaf disease classification. In: 2024 27th Interna- tional Conference on Computer and Information Technology (ICCIT). pp. 699–704 (2024). https://d...

    2024

  6. [13]

    In: Computer Vision – ECCV 2020

    Dvornik, N., Schmid, C., Mairal, J.: Selecting relevant features from a multi-domain rep- resentation for few-shot classification. In: Computer Vision – ECCV 2020. pp. 769–786. Springer International Publishing (2020)

    2020

  7. [14]

    Frontiers in Plant Science 13 (2022)

    Egusquiza, I., Picon, A., Irusta, U., Bereciartua-Perez, A., Eggers, T., Klukas, C., Navarra- Mestre, R.: Analysis of few-shot techniques for fungal plant disease classification and eval- uation of clustering capabilities over real datasets. Frontiers in Plant Science 13 (2022)

    2022

  8. [15]

    In: Proceedings of the 34th International Conference on Machine Learning

    Finn, C., Abbeel, P., Levine, S.: Model-agnostic meta-learning for fast adaptation of deep net- works. In: Proceedings of the 34th International Conference on Machine Learning. vol. 70, pp. 1126–1135. PMLR (06–11 Aug 2017) DExNet: Domain Adapted Critics for Leaf Disease Classi...

    2017

  9. [16]

    Neural networks 18(5-6), 602–610 (2005)

    Graves, A., Schmidhuber, J.: Framewise phoneme classification with bidirectional lstm and other neural network architectures. Neural networks 18(5-6), 602–610 (2005)

    2005

  10. [17]

    In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

    He, K., Zhang, X., Ren, S., Sun, J.: Deep Residual Learning for Image Recognition. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR). pp. 770–778 (2016)

    2016

  11. [18]

    In: 2023 International Conference on Information and Communication Technology for Sustainable Development (ICICT4SD)

    Herok, A., Ahmed, S.: Cotton leaf disease identification using transfer learning. In: 2023 International Conference on Information and Communication Technology for Sustainable Development (ICICT4SD). pp. 158–162 (2023)

    2023

  12. [19]

    Computers and Electronics in Agriculture 163, 104852 (2019)

    Hu, G., Wu, H., Zhang, Y ., Wan, M.: A low shot learning method for tea leaf’s disease identification. Computers and Electronics in Agriculture 163, 104852 (2019)

    2019

  13. [20]

    In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

    Huang, G., Liu, Z., Van Der Maaten, L., Weinberger, K.Q.: Densely Connected Convolu- tional Networks. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR). pp. 2261–2269 (2017)

    2017

  14. [21]

    arXiv - Computing Research Repository (2015)

    Hughes, D.P., Salathé, M.: An open access repository of images on plant health to enable the development of mobile disease diagnostics through machine learning and crowdsourcing. arXiv - Computing Research Repository (2015)

    2015

  15. [22]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2019)

    Jamal, M.A., Qi, G.J.: Task agnostic meta-learning for few-shot learning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2019)

    2019

  16. [23]

    In: 25th International Conference on Computer and Infor- mation Technology (ICCIT)

    Khan, A.M., Ashrafee, A., Sayera, R., Ivan, S., Ahmed, S.: Rethinking cooking state recog- nition with vision transformers. In: 25th International Conference on Computer and Infor- mation Technology (ICCIT). pp. 170–175 (2022)

    2022

  17. [24]

    In: Proceedings of the 32nd Int

    Koch, G., Zemel, R., Salakhutdinov, R.: Siamese neural networks for one-shot image recog- nition. In: Proceedings of the 32nd Int. Conference on Machine Learning. JMLR (2015)

    2015

  18. [25]

    In: ECCV 2020

    Kolesnikov, A., Beyer, L., Zhai, X., Puigcerver, J., Yung, J., Gelly, S., Houlsby, N.: Big transfer (bit): General visual representation learning. In: ECCV 2020. pp. 491–507 (2020)

    2020

  19. [26]

    IEEE Access 9, 56683–56698 (2021)

    Li, L., Zhang, S., Wang, B.: Plant Disease Detection and Classification by Deep Learning—A Review. IEEE Access 9, 56683–56698 (2021)

    2021

  20. [27]

    Computers and Electronics in Agriculture 182, 106055 (2021)

    Li, Y ., Yang, J.: Meta-learning baselines and database for few-shot classification in agricul- ture. Computers and Electronics in Agriculture 182, 106055 (2021)

    2021

  21. [28]

    Applied Sciences 10(4) (2020)

    Maeda-Gutierrez, V ., Galvan-Tejada, C.E., Zanella-Calzada, L.A., Celaya-Padilla, J.M., Galván-Tejada, J.I., Gamboa-Rosales, H., Luna-Garcia, H., Magallanes-Quintanar, R., Guer- rero Mendez, C.A., Olvera-Olvera, C.A.: Comparison of Convolutional Neural Network Ar- chitectures ...

    2020

  22. [29]

    In: 2022 12th International Conference on Electrical and Computer Engineering (ICECE)

    Morshed, M.S., Ahmed, S., Ahmed, T., Islam, M.U., Ashikur Rahman, A.: Fruit quality assessment with densely connected convolutional neural network. In: 2022 12th International Conference on Electrical and Computer Engineering (ICECE). pp. 1–4 (2022)

    2022

  23. [30]

    CoRR abs/2102.12295 (2021)

    Nesteruk, S., Shadrin, D., Pukalchik, M.: Image augmentation for multitask few-shot learn- ing: Agricultural domain use-case. CoRR abs/2102.12295 (2021)

    2021 arXiv

  24. [31]

    CoRR abs/1803.02999 (2018)

    Nichol, A., Achiam, J., Schulman, J.: On first-order meta-learning algorithms. CoRR abs/1803.02999 (2018)

    2018 arXiv

  25. [32]

    Journal of Intelligent & Fuzzy Systems 40, 12383–12398 (2021), 6

    Noon, S.K., Amjad, M., Ali Qureshi, M., Mannan, A.: Computationally light deep learning framework to recognize cotton leaf diseases. Journal of Intelligent & Fuzzy Systems 40, 12383–12398 (2021), 6

    2021

  26. [33]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops

    Nuthalapati, S.V ., Tunga, A.: Multi-domain few-shot learning and dataset for agricultural applications. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops. vol. abs/2109.09952, pp. 1399–1408 (October 2021)

    2021 arXiv

  27. [34]

    Agronomy (2021)

    Panno, S., Davino, S., Caruso, A.G., Bertacca, S., Crnogorac, A., Mandi ´c, A., Noris, E., Mati´c, S.: A Review of the Most Common and Economically Important Diseases That Un- dermine the Cultivation of Tomato Crop in the Mediterranean Basin. Agronomy (2021)

    2021

  28. [35]

    Ahmed et al

    Rafi, M.H., Ratul Mahjabin, M., Rahman, M.S., Hasanul Kabir, M., Ahmed, S.: A critical analysis of deep learning applications in crop pest classification: Promising pathways and 16 S. Ahmed et al. limitations. In: 2023 26th International Conference on Computer and Information ...

    2023

  29. [36]

    In: ICCV 2025 Workshop on Cultural Continuity of Artists (2025)

    Raiyan, S.R., Amio, Z.Z., Ahmed, S.: Hasper: An image repository for hand shadow puppet recognition. In: ICCV 2025 Workshop on Cultural Continuity of Artists (2025)

    2025

  30. [37]

    In: International Electronics Symposium (IES)

    Sholihati, R.A., Sulistijono, I.A., Risnumawan, A., Kusumawati, E.: Potato leaf disease clas- sification using deep learning approach. In: International Electronics Symposium (IES). pp. 392–397 (2020)

    2020

  31. [38]

    arXiv preprint arXiv:1409.1556 (2014)

    Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recog- nition. arXiv preprint arXiv:1409.1556 (2014)

    2014 arXiv

  32. [39]

    In: Advances in Neural Information Processing Systems

    Snell, J., Swersky, K., Zemel, R.: Prototypical networks for few-shot learning. In: Advances in Neural Information Processing Systems. vol. 30. Curran Associates, Inc. (2017)

    2017

  33. [40]

    Agronomy Journal 116(3), 1204–1216 (2024)

    Sun, J., Cao, W., Fu, X., Ochi, S., Yamanaka, T.: Few-shot learning for plant disease recog- nition: A review. Agronomy Journal 116(3), 1204–1216 (2024)

    2024

  34. [41]

    In: IEEE Conference on Computer Vision and Pattern Recogni- tion (CVPR)

    Szegedy, C., Vanhoucke, V ., Ioffe, S., Shlens, J., Wojna, Z.: Rethinking the inception archi- tecture for computer vision. In: IEEE Conference on Computer Vision and Pattern Recogni- tion (CVPR). pp. 2818–2826 (2016)

    2016

  35. [42]

    Tian, Y ., Wang, Y ., Krishnan, D., Tenenbaum, J.B., Isola, P.: Rethinking few-shot image classification: A good embedding is all you need? In: Computer Vision – ECCV 2020. pp. 266–282. Springer International Publishing (2020)

    2020

  36. [43]

    Multimedia Tools and Applications 83(20), 58293–58308 (Jun 2024)

    Uskaner Hepsa ˘g, P.: Efficient plant disease identification using few-shot learning: a transfer learning approach. Multimedia Tools and Applications 83(20), 58293–58308 (Jun 2024)

    2024

  37. [44]

    In: Advances in Neural Information Processing Systems

    Vinyals, O., Blundell, C., Lillicrap, T., kavukcuoglu, k., Wierstra, D.: Matching networks for one shot learning. In: Advances in Neural Information Processing Systems. vol. 29. Curran Associates, Inc. (2016)

    2016

  38. [45]

    IEEE Access 7, 151754–151763 (2019)

    Wang, B., Wang, D.: Plant leaves classification: A few-shot learning method based on siamese network. IEEE Access 7, 151754–151763 (2019)

    2019

  39. [47]

    Frontiers in Plant Science Volume 14 - 2023 (2023)

    Xu, M., Kim, H., Yang, J., Fuentes, A., Meng, Y ., Yoon, S., Kim, T., Park, D.S.: Embracing limited and imperfect training datasets: opportunities and challenges in plant disease recog- nition using deep learning. Frontiers in Plant Science Volume 14 - 2023 (2023)

    2023

  40. [48]

    In: 2021 Joint 10th International Confer- ence on Informatics, Electronics & Vision (ICIEV) and 2021 5th International Conference on Imaging, Vision & Pattern Recognition (icIVPR)

    Yasmeen, A., Rahman, F.I., Ahmed, S., Kabir, M.H.: Csvc-net: Code-switched voice com- mand classification using deep cnn-lstm network. In: 2021 Joint 10th International Confer- ence on Informatics, Electronics & Vision (ICIEV) and 2021 5th International Conference on Imaging, ...

    2021

  41. [49]

    Multimedia Tools and Applications 83(18), 54147–54164 (2024)

    Zabihzadeh, D., Masoudifar, M.: Zs-dml: Zero-shot deep metric learning approach for plant leaf disease classification. Multimedia Tools and Applications 83(18), 54147–54164 (2024)

    2024

  42. [50]

    Big Data Mining and Analytics 8(1), 145–188 (2025)

    Zhao, J., Kong, L., Lv, J.: An overview of deep neural networks for few-shot learning. Big Data Mining and Analytics 8(1), 145–188 (2025)

    2025

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.