CoChat
Sign Up
Cited by 2

The DOI matches its Crossref record; 3 of the 3 references it cites were found in a registry or an index.

Learning from the Rare: Overcoming Class Imbalance in Archaeological Object Detection with Boosting Methods

Argyrou, Argyro, Fasson, Federico, Farinetti, Emeri, Papakonstantinou, Apostolos, Alexakis, Dimitrios D., Agapiou, Athos·2026·Heritage
DOI 10.3390/heritage9030099
Open this paper in CoChat — free account

Abstract

Detecting surface potsherds using low-altitude remote sensing is challenging due to severe class imbalance and limited training data. This study develops and validates a semi-automatic detection methodology that adapts threshold-optimized boosting classifiers (AdaBoost, XGBoost) to maximize ceramic detection recall under extreme class imbalance in the Western Megaris archeological landscape, Greece. Models were trained on only 15% of the available data to simulate realistic field conditions. Evaluation emphasized recall-oriented metrics (precision, recall, F1-score, AUC) for the minority class, addressing the accuracy paradox where high overall accuracy masks poor rare-class performance. Threshold optimization enabled AdaBoost and XGBoost to achieve substantially improved recall compared to baseline methods, with detection-to-ground-truth ratios of 2.5 and 3.2, respectively, reflecting deliberate prioritization of recall over precision for exploratory survey purposes. The results demonstrate that this methodological framework provides archeologically interpretable screening tools for identifying high-probability ceramic locations, supporting more efficient field survey design and heritage documentation workflows in Mediterranean landscapes.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.3390/heritage9030099

Cites

3 of 3 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Flagged first, then by citations

Show all 3 references

Research works printed but not linked: 54

  1. Attema, 2020, A guide to good practice in Mediterranean surface survey projects, J. Greek Archaeol., 5, 1, 10.32028/9781789697926-2 · DOI
  2. Argyrou, A., and Agapiou, A. (2022). A Review of Artificial Intelligence and Remote Sensing for Archaeological Research. Remote Sens., 14. · DOI
  3. Orengo, 2019, A brave new world for archaeological survey: Automated machine learning-based pot-sherd detection using high-resolution drone imagery, J. Archaeol. Sci., 112, 105013, 10.1016/j.jas.2019.105013 · DOI
  4. Traviglia, A., and Torsello, A. (2017). Landscape pattern detection in archaeological remote sensing. Geosciences, 7. · DOI
  5. Tapete, D. (2018). Remote Sensing and Geosciences for Archaeology. Geosciences, 8. · DOI
  6. Lasaponara, 2011, Satellite remote sensing in archaeology: Past, present and future perspectives, J. Archaeol. Sci., 38, 1995, 10.1016/j.jas.2011.02.002 · DOI
  7. Materazzi, 2022, Archaeological crop marks detection through drone multispectral remote sensing and vegetation indices: A new approach tested on the Italian pre-Roman city of Veii, J. Archaeol. Sci. Rep., 41, 103235
  8. Trier, 2019, Using deep neural networks on airborne laser scanning data: Results from a case study of semi-automatic mapping of archaeological topography on Arran, Scotland, Archaeol. Prospect., 26, 165, 10.1002/arp.1731 · DOI
  9. Agapiou, A., Vionis, A., and Papantoniou, G. (2021). Detection of Archaeological Surface Ceramics Using Deep Learning Image-Based Methods and Very High-Resolution UAV Imageries. Land, 10. · DOI
  10. Altaweel, M., Khelifi, A., Li, Z., Squitieri, A., Basmaji, T., and Ghazal, M. (2022). Automated Archaeological Feature Detection Using Deep Learning on Optical UAV Imagery: Preliminary Results. Remote Sens., 14. · DOI
  11. Bickler, 2018, Machine Learning Identification and Classification of Historic Ceramics, Archaeol. N. Zeal., 61, 20
  12. Bickler, 2018, Prospects for Machine Learning for Shell Midden Analysis, Archaeol. N. Zeal., 61, 48
  13. Lambers, 2019, Learning to Look at LiDAR: The Use of R-CNN in the Automated Detection of Archaeological Objects in LiDAR Data from the Netherlands, J. Comput. Appl. Archaeol., 2, 31
  14. Reese, 2021, Deep learning artificial neural networks for non-destructive archaeological site dating, J. Archaeol. Sci., 132, 105413, 10.1016/j.jas.2021.105413 · DOI
  15. Bonhage, 2021, A modified Mask region-based convolutional neural network approach for the automated detection of archaeological sites on high-resolution light detection and ranging-derived digital elevation models in the North German Lowland, Archaeol. Prospect., 28, 177, 10.1002/arp.1806 · DOI · Paper page
  16. Pawlowicz, 2021, Applications of deep learning to decorated ceramic typology and classification: A case study using Tusayan White Ware from Northeast Arizona, J. Archaeol. Sci., 130, 105375, 10.1016/j.jas.2021.105375 · DOI
  17. Davis, 2020, Defining what we study: The contribution of machine automation in archaeological research, Digit. Appl. Archaeol. Cult. Herit., 18, e00152
  18. Olivier, 2021, Implementing State of-the-Art Deep Learning Approaches for Archaeological Object Detection in Remotely-Sensed Data: The Results of Cross-Domain Collaboration, J. Comput. Appl. Archaeol., 4, 274
  19. Richards-Rissetto, F., Newton, D., and Al-Zadjali, A. (September, January 28). A 3D point cloud Deep Learning approach using Lidar to identify ancient Maya archaeological sites. Proceedings of the 28th CIPA Symposium “Great Learning & Digital Emotion”, Beijing, China.
  20. Berganzo-Besga, I., Orengo, H.A., Lumbreras, F., Carrero-Pazos, M., Fonte, J., and Vilas-Estévez, B. (2021). Hybrid MSRM-Based Deep Learning and Multitemporal Sentinel-2-Based Machine Learning Algorithm Detects Near 10k Archaeological Tumuli in North-Western Iberia. Remote Sens., 13. · DOI
  21. Verschoof-van der Vaart, W.B., Lambers, K., Kowalczyk, W., and Bourgeois, Q.P. (2020). Combining Deep Learning and Location-Based Ranking for Large-Scale Archaeological Prospection of LiDAR Data from The Netherlands. ISPRS Int. J. Geo-Inf., 9. · DOI · Paper page
  22. Trier, 2021, Automated mapping of cultural heritage in Norway from airborne lidar data using faster R-CNN, Int. J. Appl. Earth Obs. Geoinf., 95, 102241
  23. Somrak, M., Sašo Džeroski, S., and Kokalj, Z. (2020). Learning to Classify Structures in ALS-Derived Visualizations of Ancient Maya Settlements with CNN. Remote Sens., 12. · DOI
  24. Maxwell, A.E., Pourmohammadi, P., and Poyner, J.D. (2020). Mapping the Topographic Features of Mining-Related Valley Fills Using Mask R-CNN Deep Learning and Digital Elevation Data. Remote Sens., 12. · DOI
  25. Martin-Abadal, M., Piñar-Molina, M., Martorell-Torres, A., Oliver-Codina, G., and Gonzalez-Cid, Y. (2021). Underwater Pipe and Valve 3D Recognition Using Deep Learning Segmentation. J. Mar. Sci. Eng., 9. · DOI
  26. Ball, 2017, Comprehensive survey of deep learning in remote sensing: Theories, tools, and challenges for the community, J. Appl. Remote Sens., 11, 042609, 10.1117/1.JRS.11.042609 · DOI · Paper page
  27. Fu, 2018, Using convolutional neural network to identify irregular segmentation objects from very high-resolution remote sensing imagery, J. Appl. Remote Sens., 12, 025010, 10.1117/1.JRS.12.025010 · DOI
  28. Guyot, A., Hubert-Moy, L., and Lorho, T. (2018). Detecting neolithic burial mounds from lidar derived elevation data using a multi-scale approach and machine learning techniques. Remote Sens., 10. · DOI
  29. Zhang, 2016, Deep learning for remote sensing data: A technical tutorial on the state of the art, IEEE Geosci. Remote Sens. Mag., 4, 22, 10.1109/MGRS.2016.2540798 · DOI
  30. Organista, 2020, Artificial intelligence provides greater accuracy in the classification of modern and ancient bone surface modifications, Sci. Rep., 10, 18862, 10.1038/s41598-020-75994-7 · DOI
  31. Argyrou, A., and Agapiou, A. (2024). Archaeological Surface Ceramics Detection Using Low Altitude Images Based on Weak Learners. Proceedings of the 2024 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Athens, Greece, 7–12 July 2024, IEEE. · DOI
  32. Sun, 2009, Classification of Imbalanced Data: A Review, Int. J. Pattern Recognit. Artif. Intell., 23, 687, 10.1142/S0218001409007326 · DOI
  33. Valverde-Albacete, F.J., and Peláez-Moreno, C. (2014). 100% Classification Accuracy Considered Harmful: The Normalized Information Transfer Factor Explains the Accuracy Paradox. PLoS ONE, 9. · DOI · Paper page
  34. Davis, 2021, The Importance of Understanding False Discoveries and the Accuracy Paradox When Evaluating Quantitative Studies, Stud. Soc. Sci. Res., 2, 1, 10.22158/sssr.v2n2p1 · DOI
  35. Congalton, 1983, Assessing Landsat classification accuracy using discrete multivariate analysis statistical techniques, Photogramm. Eng. Remote Sens., 49, 1671
  36. Fitzgerald, 1994, Assessing the classification accuracy of multisource remote sensing data, Remote Sens. Environ., 47, 362, 10.1016/0034-4257(94)90103-1 · DOI
  37. Foody, 2020, Explaining the unsuitability of the kappa coefficient in the assessment and comparison of the accuracy of thematic maps obtained by image classification, Remote Sens. Environ., 239, 111630, 10.1016/j.rse.2019.111630 · DOI
  38. Joshi, M.V., Kumar, V., and Agarwal, R.C. (December, January 29). Evaluating boosting algorithms to classify rare classes: Comparison and improvements. Proceedings of the 2001 IEEE International Conference on Data Mining (ICDM’01), San Jose, CA, USA.
  39. Weiss, 2004, Mining with rarity: A unifying framework, SIGKDD Explor. Newsl., 6, 7, 10.1145/1007730.1007734 · DOI · Paper page
  40. Van Leusen, M., Mayoral Herrera, V., Farinetti, E., de Haas, T., and Vandam, R. (2025, April 10). FIDO Guidelines for Fieldwalking Documentation 1.1. Available online: https://www.academia.edu/109603068/FIDO_Guidelines_for_Fieldwalking_Documentation_1_1.
  41. Argyrou, A., Agapiou, A., Papakonstantinou, A., and Alexakis, D.D. (2023). Comparison of Machine Learning Pixel-Based Classifiers for Detecting Archaeological Ceramics. Drones, 7. · DOI · Paper page
  42. Argyrou, A., Vionis, A., Papakonstantinou, A., Christoforou, M., Alexakis, D., and Agapiou, A. (2025, January 5–9). Utilizing Low-Altitude Imagery and Weak Learner Algorithms for the Detection of Archaeological Surface Ceramics: The Kophinou case study. Proceedings of the 52nd CAA International Conference, Athens, Greece.
  43. Breiman, 2001, Random forests, Mach. Learn., 45, 5, 10.1023/A:1010933404324 · DOI · Paper page
  44. John, G.H., and Langley, P. (1995). Estimating continuous distributions in Bayesian classifiers. Proceedings of the Eleventh Conference on Uncertainty in Artificial Intelligence, Montreal, QC, Canada, 18–20 August 1995, Morgan Kaufmann Publishers.
  45. Schapire, 1990, The Strength of Weak Learnability, Mach. Learn., 5, 197, 10.1023/A:1022648800760 · DOI · Paper page
  46. Fulk, M.A., and Case, J. (1990). Boosting a Weak Learning Algorithm by Majority. Proceedings of the Third Annual Workshop on Computational Learning Theory (COLT), University of Rochester, Rochester, NY, USA, 6–8 August 1990, Morgan Kaufmann Publishers Inc.
  47. Freund, 1997, A decision-theoretic generalization of on-line learning and an application to boosting, J. Comput. Syst. Sci., 55, 119, 10.1006/jcss.1997.1504 · DOI · Paper page
  48. Chen, T., and Guestrin, C. (2016). XGBoost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA, 13–17 August 2016, Association for Computing Machinery. · DOI · Paper page
  49. Prati, R.C., and Batista, G.E.A.P.A. (2004). Class imbalances versus class overlapping: An analysis of a learning system behavior. Proceedings of the Mexican International Conference on Artificial Intelligence (MICAI), Mexico City, Mexico, 26–30 April 2004, Springer. · DOI
  50. Brodersen, K.H., Ong, C.S., Stephan, K.E., and Buhmann, J.M. (2010). The balanced accuracy and its posterior distribution. Proceedings of the 2010 20th International Conference on Pattern Recognition, Istanbul, Turkey, 23–26 August 2010, IEEE. · DOI · Paper page
  51. Sokolova, 2009, A systematic analysis of performance measures for classification tasks, Inf. Process. Manag., 45, 427, 10.1016/j.ipm.2009.03.002 · DOI · Paper page
  52. Hanley, 1982, The meaning and use of the area under a receiver operating characteristic (ROC) curve, Radiology, 143, 29, 10.1148/radiology.143.1.7063747 · DOI · Paper page
  53. Sun, 2021, Research Progress on Few-Shot Learning for Remote Sensing Image Interpretation, IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., 14, 2387, 10.1109/JSTARS.2021.3052869 · DOI
  54. Zhou, 2024, ISPRS Journal of Photogrammetry and Remote Sensing Gaussian meta-feature balanced aggregation for few-shot synthetic aperture radar target detection, ISPRS J. Photogramm. Remote Sens., 208, 89, 10.1016/j.isprsjprs.2024.01.003 · DOI

Cited by

Sorted by citations

See all 2

Metadata

Authors
Argyrou, A., Fasson, F., Farinetti, E., Papakonstantinou, A., Alexakis, D. D., & Agapiou, A.
Venue
Heritage
Publisher
MDPI AG
Published
2026-03-02
Volume
9
Issue
3
Pages
99
ISSN
2571-9408
DOI
10.3390/heritage9030099
Cited by
2
References
3
Language
en
Licence
creativecommons.org/licenses/by/4.0/ (vor)
Funders
  • Cyprus University of Technology internal funding
  • European Regional Development Fund and the Republic of Cyprus (SMALL SCALE INFRASTRUCTURES/1222/0062)

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review