Pith. sign in

REVIEW 3 major objections 6 minor 51 references

A High-Accuracy SSIM-based Scoring System for Coin Die Link Identification

T0 review · 3 major / 6 minor · reviewed 2026-08-09 · deepseek-v4-flash

Pith's one-line read A SSIM-based image distance can identify ancient coins struck from the same die, and clustering on that distance recovers the true die-link groups nearly perfectly.

desk verdict A genuinely useful first dataset and a credible screening distance, but the 'outperforms current techniques' claim is not yet demonstrated because the baseline is only inspired by prior work. read the letter →

arxiv 2502.01186 v1 pith:ZJWU5HZQ submitted 2025-02-03 cs.CV

classification cs.CV
keywords coindielinkidentificationSSIMstructuralsimilarityancientnumismaticsimagedistanceclusteringbenchmarkdatasetJuillactreasure
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

This paper sets out to replace the slow, manual comparison of ancient coin photos with a fast automatic distance measure that tells when two coins were struck by the same engraved die. It introduces the first publicly labeled image dataset for die-link detection, containing 329 coin photos from the Juillac treasure, and proposes an SSIM-based distance computed after aligning coin images with keypoint matching. On this dataset, the distance separates true die links from non-links better than a Procrustes-based baseline inspired by prior work, and a simple agglomerative clustering with a max-threshold rule achieves perfect or near-perfect clustering (ARI 1.0 on six of eight datasets). If the result holds, numismatists can pre-screen huge hoards automatically and focus manual verification on the most likely links.

What carries the argument

The central object is the SSIM-based distance $M(A,B)$, a per-pixel average of the local metric $M_{ij}(A,B)=\sqrt{2-l_{ij}^{AB}-s_{ij}^{AB}c_{ij}^{AB}}$, which combines luminance, contrast, and structure terms from the structural similarity index. It is a true metric (satisfies the triangle inequality), unlike a simple $1-\text{SSIM}$ score, which matters for distance-based clustering. The pipeline uses preprocessing (grayscale, circular crop, CLAHE, non-local means denoising), then SIFT keypoints solely to estimate an affine overlay of the two images, after which the global SSIM distance is computed over the aligned surfaces. The baseline is a Procrustes-based distance that combines the log of the Procrustes distance between keypoint configurations with the inverse number of RANSAC inliers.

What would settle it

Re-run the same eight-dataset experiment with the original implementations of the cited methods (for example, the actual code from the papers cited as [21] and [30]) and with a second coin hoard photographed under varied lighting; if the SSIM-based distance then fails to separate true die links from non-links at the reported accuracy, the near-perfect clustering claim would not generalize.

Watch

Extended reading notes

Core claim

The paper claims that a global structural-similarity distance, computed after affine alignment, is a better discriminator of coin die links than the keypoint-and-Procrustes approaches previously used in the field. The distance is defined from a local SSIM metric, $M_{ij}(A,B)=\sqrt{2-l_{ij}^{AB}-s_{ij}^{AB}c_{ij}^{AB}}$, averaged over all pixels; the alignment is obtained by SIFT keypoints with a ratio test and an affine transform, with the transform discarded if the estimated scale is far from 1. Combined with agglomerative clustering using the maximum of leave-one-out optimal thresholds for the cutoff, the approach yields adjusted Rand index 1.0 on DS1, DS2, DS4, DS5, DS6, and DS7, and 0.947 and 0.909 on DS3 and DS8, with perfect precision on all datasets. The authors also release the dataset and position the method as a fast pre-screening tool for large coin hoards.

Load-bearing premise

The comparison baseline is an approximation of prior methods re-implemented by the authors, not the original published code, so the claim that the SSIM distance outperforms current techniques depends on this proxy being faithful.

Editorial extensions

If this is right

  • If the SSIM distance performs as reported, coin die-link analysis of large hoards becomes practical: the 1,350-hour manual comparison in the L'Isle-Jourdain example could be reduced to an automatic pre-screen that ranks pairs for human verification.
  • The released dataset gives other researchers a common benchmark for die-link detection, enabling direct comparison of future methods rather than evaluation on private collections.
  • The success of the max-threshold aggregation strategy suggests that the inter-cluster distance distribution has a stable lower bound across datasets, so thresholds learned on one set transfer to another of the same imaging protocol.
  • Because the SSIM-based distance runs in hours versus days for the Procrustes baseline, the method is plausible for online production use on newly excavated or photographed coin collections.
  • The method's global nature means it exploits full image detail rather than only matched keypoints, which may generalize to other artifact-matching tasks where local features fail to capture subtle die-specific engravings.

Reading between the lines

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

  • The core idea -- align globally, then compare with a full-image structural metric -- could be tested on other fine-grained object-identity problems, such as tool-mark matching in forensics or seal impressions in archaeology, where imaging conditions are similarly controlled.
  • The paper leaves implicit that its threshold-learning scheme assumes consistent lighting orientation (always lit from the left), so a natural extension is to test how the distance degrades when lighting direction varies or when coins are photographed under different protocols.
  • A straightforward improvement, not explored here, would be to replace the affine alignment with a full homography when scale is trusted, or to use multi-scale SSIM variants (MS-SSIM, CW-SSIM) that the authors list as future work; these could improve the two harder datasets DS3 and DS8.
  • The claimed superiority over state of the art rests on a re-implementation of prior methods rather than their original code; a stronger comparison would run the original published pipelines on the same dataset, which the authors note is currently impossible because those codes were not released.
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 / 6 minor

Summary. The paper introduces a labeled dataset of 329 ancient coin images for die link detection, proposes an SSIM-based image distance (Algorithm 1) that aligns images with SIFT and then computes a metric derived from the SSIM index, and evaluates this distance with agglomerative clustering using leave-one-out threshold estimation on eight datasets from the Juillac hoard. The authors claim that the SSIM-based distance achieves near-perfect die link identification when combined with a max-threshold aggregation rule (ACmax), and that it outperforms current techniques used in the field. The paper also reports comparisons against FSIM-, Procrustes-, and VGG-based distances.

Significance. If the central claims are substantiated, the released dataset and the fast, parameter-free SSIM distance would be valuable contributions to computational numismatics, providing a benchmark and a practical pre-screening tool. The evaluation is thorough in several respects: leave-one-out threshold estimation, multiple clustering and classification metrics, comparisons against FSIM and VGG baselines, and use of default library parameters for the proposed method. However, the comparative claim against the state of the art rests on a baseline that the authors themselves describe as 'inspired by' published methods rather than a faithful reproduction, so the empirical superiority over existing techniques is not yet established. The dataset itself is also filtered by lighting conditions without an analysis of the excluded images, which weakens the benchmark's representativeness.

major comments (3)
  1. [Section 4, Algorithm 2, and Section 5] The baseline used for the comparative claim is not a faithful implementation of any cited method. Section 3 describes [21] as using VLFeat descriptors with bounded-distortion matching, [30] as using SIFT with ratio test and bounded-distortion matching, and [46] as averaging ORB descriptor distances. Algorithm 2 instead combines Gaussian-process keypoints with ORB descriptors, cross-check matching, a RANSAC homography, and the distance dP(A,B)=log(P)+1/nin, which does not match any of the three cited pipelines. Since the paper's headline claim is that the SSIM-based method outperforms current techniques, the experiments need to compare against the actual published methods, or the authors must release the proxy code and demonstrate that it reproduces the original methods' behavior. In the absence of that, the claim should be restricted to the implemented baseline.
  2. [Section 2] The paper excludes 72 of 401 images because they are not lit from the left, but it provides no analysis of these excluded images or of how their removal affects the reported results. Because the dataset is presented as a benchmark and the exclusion criterion relies on the photographer's setup rather than an objective image property, the authors should report how many die-linked pairs were excluded and whether the remaining data are representative. At a minimum, the potential selection bias should be discussed as a limitation of the dataset.
  3. [Section 5, Table 4] The claim of 'near-perfect die link identification' depends on the ACmax aggregation rule, whose choice is justified only post hoc from the histograms in Fig. 7. The leave-one-out threshold estimation is performed on the other datasets from the same hoard with the same lighting and preprocessing, so the threshold may not generalize to other coin collections. Please provide an independent validation or a more principled justification for selecting the maximum of the optimal thresholds, and discuss the variability of results across aggregation rules as part of the central claim.
minor comments (6)
  1. [Figure 6 caption] The caption states 'Procruste-base distance in blue' but the SSIM-based distance is also described as blue in the same caption; the colors should be disambiguated.
  2. [Section 4, paragraph after Algorithm 2] There is a typo: 'inpired' should be 'inspired' in the sentence 'Now that the baseline distance inpired by the state of the art methods...'
  3. [Table 4] The column header 'Ajusted Rand Index' should be 'Adjusted Rand Index'.
  4. [Algorithm 1, line 9] The syntax 'if|s− 1|> 0.25; /* Wrong estimation */ then' contains an extra semicolon and an awkward comment placement; this should be rewritten for clarity.
  5. [Section 2] The sentence 'Out of the eight datasets we kept, i.e. 401 coins, the lighting was correct for only 329' is confusing because it suggests the datasets originally contained 401 coins and 329 were kept; it would be clearer to state that 72 images were removed due to lighting.
  6. [Section 5] The statement that k-means, k-medoids, and CoRe without repulsion 'resulted in very poor clustering predictions' is not accompanied by results; including these numbers in the supplementary material would strengthen the evaluation.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the SSIM distance is defined from published formulas with no fitted constants, and the clustering threshold is estimated via leave-one-out on other datasets; the main weakness is an unvalidated 'inspired by' baseline, which is a correctness concern, not a circular step.

full rationale

The SSIM-based distance is derived from published SSIM and Brunet et al.'s metric (Eqs. 1-3), with no fitted parameters; Algorithm 1 uses default library settings for preprocessing, SIFT, and alignment. The clustering thresholds are computed from other datasets in a leave-one-out fashion, not fitted to the target pairs, and the paper transparently reports all four aggregation variants (ACmax, ACmean, ACmed, ACmin). The choice of ACmax as the best aggregation rule is a model-selection step made after comparing the variants on the full dataset, which may overstate the headline clustering performance, but it does not reduce the result to its inputs by construction. The baseline Algorithm 2 is explicitly described as 'the best possible pipeline inspired by [21,30,46]' rather than a faithful reproduction of any cited method; this undermines the comparative claim 'outperforming current techniques' as an evidence-quality issue, not a circularity. No load-bearing self-citation chain appears: the author-linked dataset reference [24] is a data contribution, and the CoRe package [30] is used as an external baseline. The central derivation is therefore self-contained.

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

The evaluation depends on five kinds of external input: the correctness of expert die-link labels, the metric property of the SSIM distance from the cited literature, the adequacy of SIFT-based alignment for superimposing coin photos, the representativeness of left-lit images, and the faithfulness of the Procrustes baseline to prior methods. No new physical or mathematical entities are introduced.

free parameters (3)
  • Clustering decision threshold and aggregation rule (ACmax/mean/median/min) = Dataset-specific optimal thresholds; headline results use ACmax
    Thresholds are computed from ground-truth optimal values on the other datasets, and the aggregation rule is chosen after comparing all four rules on the target datasets. This data-dependent choice affects the near-perfect clustering numbers in Table 4.
  • SIFT alignment scale threshold = 0.25
    Algorithm 1 lines 9-11 discard estimated transformations when |s-1|>0.25. No sensitivity analysis is provided, and the value is chosen by hand.
  • Minimum number of SIFT matches for transformation estimation = 4
    Algorithm 1 line 6 only estimates a 2D transformation if more than 4 matched keypoints are found. This heuristic is not sensitivity-tested.
assumptions (4)
  • standard math The SSIM-based distance defined in Equations 2-3 is a valid metric, as proven in Brunet et al. [11].
    Section 4 relies on the metric property to justify the use of distance-based clustering algorithms.
  • domain assumption Ground-truth die-link labels assigned by expert numismatists are correct.
    Section 2 describes the labels as given by the archaeologists and numismatists studying the Juillac treasure; all evaluations depend on these labels.
  • domain assumption SIFT-based affine alignment with the scale check in Algorithm 1 is sufficient to superimpose same-die coin images for meaningful SSIM comparison.
    Algorithm 1 lines 5-12 estimate a 2D transformation from keypoint matches and use it to overlay images before computing SSIM. No validation of alignment quality on all pairs is reported.
  • ad hoc to paper Images lit from the left are representative of the die-link detection task.
    Section 2 excludes 72 of 401 coins because of inconsistent lighting and asserts that left-lighting is the conventional condition; no evaluation is provided for the excluded subset.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A High-Accuracy SSIM-based Scoring System for Coin Die Link Identification." pith.science (2026). https://pith.science/paper/ZJWU5HZQ

@misc{pith2026250201186,
  author       = {Pith},
  title        = {Pith review of: A High-Accuracy SSIM-based Scoring System for Coin Die Link Identification},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZJWU5HZQ}},
  note         = {Machine review of arXiv:2502.01186}
}
read the original abstract

The analyses of ancient coins, and especially the identification of those struck with the same die, provides invaluable information for archaeologists and historians. Nowadays, these die links are identified manually, which makes the process laborious, if not impossible when big treasures are discovered as the number of comparisons is too large. This study introduces advances that promise to streamline and enhance archaeological coin analysis. Our contributions include: 1) First publicly accessible labeled dataset of coin pictures (329 images) for die link detection, facilitating method benchmarking; 2) Novel SSIM-based scoring method for rapid and accurate discrimination of coin pairs, outperforming current techniques used in this research field; 3) Evaluation of clustering techniques using our score, demonstrating near-perfect die link identification. We provide datasets, to foster future research and the development of even more powerful tools for archaeology, and more particularly for numismatics.

Figures

Figures reproduced from arXiv: 2502.01186 by the authors.

Figure 1
Figure 1. Coins struck with the same die: example of ground truth [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. The Juillac treasure during the archaeological dig. [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Coin example for each dataset a numismatic point of view, good photos are taken with semi-glare lighting (to reproduce the slightest relief and leg￾ibility) and, above all, with a light source that is always to the left (i.e. aimed at the back of the emperor’s neck if the coin is correctly positioned under the lens). Out of the eight datasets we kept, i.e. 401 coins, the lighting was correct for only 329. This probl… view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Pre-processing steps from the original coin to the input [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 5
Figure 5. Figure 5: Steps of the computation of the SSIM-based distance, and comparison of two results (one linked, one unlinked). The greener the [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 6
Figure 6. Figure 6: ROC & PR curves: SSIM-based distance in blue, [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]
Figure 7
Figure 7. Figure 7: Histograms of the SSIM-based (left) and Procrustes-based (right) distance values. The bars in red with stripes correspond to the [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]
Figure 1
Figure 1. Figure 1: 1) Obverse legend = FL VAL CONSTANTINVS NOB C; 2) Ribbon code = 3 (i.e. 2 vertical ribbons); 3) Bust code = A*2 (i.e. bust laureate, draped, cuirassed, right, view from rear); 4) Obverse legend = GENIO POP ROM; 5) Reverse code = genio 6 (i.e. Genius, turreted, draped, …

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

51 extracted references · 46 canonical work pages

  1. [21]

    Unsupervised Statistical Learning for Die Analysis in Ancient Numismatics

    Andreas Heinecke, Emanuel Mayer, Abhinav Natarajan, and Yoonju Jung. Unsupervised statistical learning for die analysis in ancient numismatics. arXiv preprint arXiv:2112.00290 , 2021

  2. [30]

    Cohesion and repulsion in bayesian distance clustering

    Abhinav Natarajan, Maria De Iorio, Andreas Heinecke, Emanuel Mayer, and Simon Glenn. Cohesion and repulsion in bayesian distance clustering. Journal of the American Statistical Association , pages 1--11, 2023

  3. [46]

    The computer-aided die study (cads): A tool for conducting numismatic die studies with computer vision and hierarchical clustering

    Zachary McCord Taylor. The computer-aided die study (cads): A tool for conducting numismatic die studies with computer vision and hierarchical clustering. 2020

  4. [1]

    Deep ancient roman republican coin classification via feature fusion and attention

    Hafeez Anwar, Saeed Anwar, Sebastian Zambanini, and Fatih Porikli. Deep ancient roman republican coin classification via feature fusion and attention. Pattern Recognition , 114:107871, 2021

  5. [2]

    Reading ancient coins: automatically identifying denarii using obverse legend seeded retrieval

    Ognjen Arandjelovi \'c . Reading ancient coins: automatically identifying denarii using obverse legend seeded retrieval. In Computer Vision--ECCV 2012: 12th European Conference on Computer Vision, Florence, Italy, October 7-13, 2012, Proceedings, Part IV 12 , pages 317--330. Springer, 2012

  6. [3]

    Images of roman imperial denarii: A curated data set for the evaluation of computer vision algorithms applied to ancient numismatics, and an overview of challenges in the field

    Ognjen Arandjelovi \'c and Marios Zachariou. Images of roman imperial denarii: A curated data set for the evaluation of computer vision algorithms applied to ancient numismatics, and an overview of challenges in the field. Sci , 2(4):91, 2020

  7. [4]

    A review of artificial intelligence and remote sensing for archaeological research

    Argyro Argyrou and Athos Agapiou. A review of artificial intelligence and remote sensing for archaeological research. Remote Sensing , 14(23):6000, 2022

  8. [5]

    Two sides of the same coin: Improved ancient coin classification using graph transduction games

    Sinem Aslan, Sebastiano Vascon, and Marcello Pelillo. Two sides of the same coin: Improved ancient coin classification using graph transduction games. Pattern Recognition Letters , 131:158--165, 2020

Show all 51 references
  1. [6]

    The impact of triangular inequality violations on medoid-based clustering

    Saaid Baraty, Dan A Simovici, and Catalin Zara. The impact of triangular inequality violations on medoid-based clustering. In Foundations of Intelligent Systems: 19th International Symposium, ISMIS 2011, Warsaw, Poland, June 28-30, 2011. Proceedings 19 , pages 280--289. Springer, 2011

  2. [7]

    Visual analysis in archaeology

    Juan A Barcel \'o . Visual analysis in archaeology. an artificial intelligence approach. Morphometrics for Nonmorphometricians , pages 93--156, 2010

  3. [8]

    Flexible models for microclustering with application to entity resolution

    Brenda Betancourt, Giacomo Zanella, Jeffrey W Miller, Hanna Wallach, Abbas Zaidi, and Rebecca C Steorts. Flexible models for microclustering with application to entity resolution. Advances in neural information processing systems , 29, 2016

  4. [9]

    Random partition models for microclustering tasks

    Brenda Betancourt, Giacomo Zanella, and Rebecca C Steorts. Random partition models for microclustering tasks. Journal of the American Statistical Association , 117(539):1215--1227, 2022

  5. [10]

    G. Bradski. The OpenCV Library . Dr. Dobb's Journal of Software Tools , 2000

  6. [11]

    On the mathematical properties of the structural similarity index

    Dominique Brunet, Edward R Vrscay, and Zhou Wang. On the mathematical properties of the structural similarity index. IEEE Transactions on Image Processing , 21(4):1488--1499, 2011

  7. [12]

    A non-local algorithm for image denoising

    Antoni Buades, Bartomeu Coll, and J-M Morel. A non-local algorithm for image denoising. In 2005 IEEE computer society conference on computer vision and pattern recognition (CVPR'05) , volume 2, pages 60--65. Ieee, 2005

  8. [13]

    An algorithm for total variation minimization and applications

    Antonin Chambolle. An algorithm for total variation minimization and applications. Journal of Mathematical imaging and vision , 20:89--97, 2004

  9. [14]

    Learning to describe: a new approach to computer vision based ancient coin analysis

    Jessica Cooper and Ognjen Arandjelovi \'c . Learning to describe: a new approach to computer vision based ancient coin analysis. Sci , 2(2):27, 2020

  10. [15]

    Image quality assessment: Unifying structure and texture similarity

    Keyan Ding, Kede Ma, Shiqi Wang, and Eero P Simoncelli. Image quality assessment: Unifying structure and texture similarity. IEEE transactions on pattern analysis and machine intelligence , 44(5):2567--2581, 2020

  11. [16]

    Big is beautiful? faut-il vraiment \'e tudier les m \'e gad \'e p \^o ts mon \'e taires? EUT Edizioni Universit \`a di Trieste, 2019

    Jean-Marc Doyen. Big is beautiful? faut-il vraiment \'e tudier les m \'e gad \'e p \^o ts mon \'e taires? EUT Edizioni Universit \`a di Trieste, 2019

  12. [17]

    The theory of linkage

    WARREN W ESTY. The theory of linkage. The Numismatic Chronicle (1966-) , pages 205--221, 1990

  13. [18]

    Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography

    Martin A Fischler and Robert C Bolles. Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography. Communications of the ACM , 24(6):381--395, 1981

  14. [19]

    Gaussian process landmarking on manifolds

    Tingran Gao, Shahar Z Kovalsky, and Ingrid Daubechies. Gaussian process landmarking on manifolds. SIAM Journal on Mathematics of Data Science , 1(1):208--236, 2019

  15. [20]

    A siamese transformer network for zero-shot ancient coin classification

    Zhongliang Guo, Ognjen Arandjelovi \'c , David Reid, Yaxiong Lei, and Jochen B \"u ttner. A siamese transformer network for zero-shot ancient coin classification. Journal of Imaging , 9(6):107, 2023

  16. [22]

    Riedones3d: a celtic coin dataset for registration and fine-grained clustering

    Sofiane Horache, Jean-Emmanuel Deschaud, François Goulette, Katherine Gruel, Thierry Lejars, and Olivier Masson. Riedones3d: a celtic coin dataset for registration and fine-grained clustering. In Vedad Hulusic and Alan Chalmers, editors, Eurographics Workshop on Graphics and C...

  17. [23]

    Autarch: An ai-assisted workflow for object detection and automated recording in archaeological catalogues

    Kevin Klein, Alyssa Wohde, Alexander V Gorelik, Volker Heyd, Yoan Diekmann, and Maxime Brami. Autarch: An ai-assisted workflow for object detection and automated recording in archaeological catalogues. arXiv preprint arXiv:2311.17978 , 2023

  18. [24]

    Datasets for Accadil (V1) , 2024

    Patrice LABEDAN, Nicolas DROUGARD, and Francis DIEULAFAIT. Datasets for Accadil (V1) , 2024. Available at https://doi.org/10.34849/AFRCBK. https://doi.org/10.34849/AFRCBK doi:10.34849/AFRCBK

  19. [25]

    Feature matching with bounded distortion

    Yaron Lipman, Stav Yagev, Roi Poranne, David W Jacobs, and Ronen Basri. Feature matching with bounded distortion. ACM Transactions on Graphics (TOG) , 33(3):1--14, 2014

  20. [26]

    Object recognition from local scale-invariant features

    David G Lowe. Object recognition from local scale-invariant features. In Proceedings of the seventh IEEE international conference on computer vision , volume 2, pages 1150--1157. Ieee, 1999

  21. [27]

    Distinctive image features from scale-invariant keypoints

    David G Lowe. Distinctive image features from scale-invariant keypoints. International journal of computer vision , 60:91--110, 2004

  22. [28]

    Classification of ancient roman coins by denomination using colour, a forgotten feature in automatic ancient coin analysis

    Yuanyuan Ma and Ognjen Arandjelovi \'c . Classification of ancient roman coins by denomination using colour, a forgotten feature in automatic ancient coin analysis. Sci , 2(2):37, 2020

  23. [29]

    The computerization of archaeology: Survey on artificial intelligence techniques

    Lorenzo Mantovan and Loris Nanni. The computerization of archaeology: Survey on artificial intelligence techniques. SN Computer Science , 1:1--32, 2020

  24. [31]

    Understanding ssim

    Jim Nilsson and Tomas Akenine-M \"o ller. Understanding ssim. arXiv preprint arXiv:2006.13846 , 2020

  25. [32]

    Pytorch: An imperative style, high-performance deep learning library

    Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, et al. Pytorch: An imperative style, high-performance deep learning library. Advances in neural information processing systems , 32, 2019

  26. [33]

    And in Length of Days Understanding

    George Pavlidis. From digital recording to advanced ai applications in archaeology and cultural heritage. In “And in Length of Days Understanding”(Job 12: 12) Essays on Archaeology in the Eastern Mediterranean and Beyond in Honor of Thomas E. Levy , pages 1627--1656. Springer, 2023

  27. [34]

    Pedregosa, G

    F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V. Dubourg, J. Vanderplas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, and E. Duchesnay. Scikit-learn: Machine learning in P ython. Journal of Machine Learnin...

  28. [35]

    Adaptive histogram equalization and its variations

    Stephen M Pizer, E Philip Amburn, John D Austin, Robert Cromartie, Ari Geselowitz, Trey Greer, Bart ter Haar Romeny, John B Zimmerman, and Karel Zuiderveld. Adaptive histogram equalization and its variations. Computer vision, graphics, and image processing , 39(3):355--368, 1987

  29. [36]

    Realization of the contrast limited adaptive histogram equalization (clahe) for real-time image enhancement

    Ali M Reza. Realization of the contrast limited adaptive histogram equalization (clahe) for real-time image enhancement. Journal of VLSI signal processing systems for signal, image and video technology , 38:35--44, 2004

  30. [37]

    Orb: An efficient alternative to sift or surf

    Ethan Rublee, Vincent Rabaud, Kurt Konolige, and Gary Bradski. Orb: An efficient alternative to sift or surf. In 2011 International conference on computer vision , pages 2564--2571. Ieee, 2011

  31. [38]

    A study of triangle inequality violations in social network clustering

    Sanjit Kumar Saha and Tapashi Gosswami. A study of triangle inequality violations in social network clustering. Journal of Computer and Communications , 12(01):67--76, 2024

  32. [39]

    Non-ti clustering in the context of social networks

    Sanjit Kumar Saha and Ingo Schmitt. Non-ti clustering in the context of social networks. Procedia Computer Science , 170:1186--1191, 2020

  33. [40]

    Complex wavelet structural similarity: A new image similarity index

    Mehul P Sampat, Zhou Wang, Shalini Gupta, Alan Conrad Bovik, and Mia K Markey. Complex wavelet structural similarity: A new image similarity index. IEEE transactions on image processing , 18(11):2385--2401, 2009

  34. [41]

    Ancient roman coin recognition in the wild using deep learning based recognition of artistically depicted face profiles

    Imanol Schlag and Ognjen Arandjelovic. Ancient roman coin recognition in the wild using deep learning based recognition of artistically depicted face profiles. In Proceedings of the IEEE International Conference on Computer Vision Workshops , pages 2898--2906, 2017

  35. [42]

    A generalized solution of the orthogonal procrustes problem

    Peter H Sch \"o nemann. A generalized solution of the orthogonal procrustes problem. Psychometrika , 31(1):1--10, 1966

  36. [43]

    Metadata archaeology: Unearthing data subsets by leveraging training dynamics

    Shoaib Ahmed Siddiqui, Nitarshan Rajkumar, Tegan Maharaj, David Krueger, and Sara Hooker. Metadata archaeology: Unearthing data subsets by leveraging training dynamics. arXiv preprint arXiv:2209.10015 , 2022

  37. [44]

    Very deep convolutional networks for large-scale image recognition

    Karen Simonyan and Andrew Zisserman. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556 , 2014

  38. [45]

    History and definition of the sobel operator

    Irwin Sobel. History and definition of the sobel operator. Retrieved from the World Wide Web , 1505, 2014

  39. [47]

    S ch\"onberger, J uan Nunez-Iglesias , F ran c ois B oulogne, J oshua D

    S t\'efan van der Walt, J ohannes L . S ch\"onberger, J uan Nunez-Iglesias , F ran c ois B oulogne, J oshua D . W arner, N eil Y ager, E mmanuelle G ouillart, T ony Y u, and the scikit-image contributors. scikit-image: image processing in P ython. PeerJ , 2:e453, 6 2014. https...

  40. [48]

    Vlfeat: An open and portable library of computer vision algorithms

    Andrea Vedaldi and Brian Fulkerson. Vlfeat: An open and portable library of computer vision algorithms. In Proceedings of the 18th ACM international conference on Multimedia , pages 1469--1472, 2010

  41. [49]

    Image quality assessment: from error visibility to structural similarity

    Zhou Wang, Alan C Bovik, Hamid R Sheikh, and Eero P Simoncelli. Image quality assessment: from error visibility to structural similarity. IEEE transactions on image processing , 13(4):600--612, 2004

  42. [50]

    Multiscale structural similarity for image quality assessment

    Zhou Wang, Eero P Simoncelli, and Alan C Bovik. Multiscale structural similarity for image quality assessment. In The Thrity-Seventh Asilomar Conference on Signals, Systems & Computers, 2003 , volume 2, pages 1398--1402. Ieee, 2003

  43. [51]

    Fsim: A feature similarity index for image quality assessment

    Lin Zhang, Lei Zhang, Xuanqin Mou, and David Zhang. Fsim: A feature similarity index for image quality assessment. IEEE transactions on Image Processing , 20(8):2378--2386, 2011

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.