Automatic Coin Classification and Identification

Huber-Mörk, Reinhold; Nölle, Michael; Rubik, Michael; Hödlmoser, Michael; Kampel, Martin; Zambanini, Sebastian

doi:10.5772/35795

Cited by 10 publications

(10 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Previous to the above mentioned pre-sorting software CADS [7] and the supervised clustering in [8], computer vision techniques in numismatics have been used for tasks such as ancient coin classification by numismatic type or issue [16]- [21], coin identification [22], and high-volume type classification of contemporary coins [23]. While type classification is an interesting computer vision problem, it only is of limited value for historical research, because most available coin images already exist in the form of labelled data (Sec.…”

Section: Related Workmentioning

confidence: 99%

Unsupervised Statistical Learning for Die Analysis in Ancient Numismatics

Heinecke¹,

Mayer²,

Natarajan³

et al. 2021

Preprint

View full text Add to dashboard Cite

Die analysis is an essential numismatic method, and an important tool of ancient economic history. Yet, manual die studies are too labor-intensive to comprehensively study large coinages such as those of the Roman Empire. We address this problem by proposing a model for unsupervised computational die analysis, which can reduce the time investment necessary for large-scale die studies by several orders of magnitude, in many cases from years to weeks. From a computer vision viewpoint, die studies present a challenging unsupervised clustering problem, because they involve an unknown and large number of highly similar semantic classes of imbalanced sizes. We address these issues through determining dissimilarities between coin faces derived from specifically devised Gaussian process-based keypoint features in a Bayesian distance clustering framework. The efficacy of our method is demonstrated through an analysis of 1 135 Roman silver coins struck between 64-66 C.E..

show abstract

Section: Related Workmentioning

confidence: 99%

Unsupervised Statistical Learning for Die Analysis in Ancient Numismatics

Heinecke¹,

Mayer²,

Natarajan³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Zambanini and Kampel further proposed a classifier-free procedure for automated coin classification based on 2D image matching based on the SIFT operator (scale-invariant feature transform) in 2011, [34] and [35]. Huber-Moerk described the automated classification and identification of modern and ancient coins based on the analysis of 2D shapes and 3D shapes on a set of 12,000 coins [36]. Kavelar et al review existing methods for coin classification in detail, including OCR text recognition, and propose coin classification through extraction of legends of Roman Republican Coins by the SIFT descriptor in 2014 [37].…”

Section: Review Of Ongoing Research For Automated Coin Classificationmentioning

confidence: 99%

Application of multi-modal 2D and 3D imaging and analytical techniques to document and examine coins on the example of two Roman silver denarii

2018

View full text Add to dashboard Cite

This case study is applying imaging and analytical techniques from multiple scientific disciplines to digitise coins and evaluate 3D multi-modal visualisation. Two ancient Roman silver denarii were selected as test objects to establish whether the proposed digital recording methods can support professional numismatic comparison of features and properties. The coins raise questions concerning their provenance, authenticity, design, purpose of issue and historic usage, but they also pose considerable recording challenges due to their material and surface properties, which are the main focus in this paper. The coins have been examined by the following techniques: dome photography for image sets for PTM/RTI visualisation and photometric stereo; X-ray microtomography for detection of cracks or impurities; Scanning Electron Microscopy for detailed surface investigation; Energy-Dispersive X-ray Spectroscopy for elemental analysis; micro X-ray fluorescence spectrometry mapping; 3D laser and structured light scanning for 3D spatial capture; photogrammetry/structure from motion, focus-stacking. The results indicate the feasibility of such techniques for museum documentation and as contribution to scientific examination of coins in general.

show abstract

“…Shape information is also used in [15]: Generalized Hough Transform (GHT) and shape based segmentation were combined to detect the coins, with the assumption that only one coin is present in an image. In [16][17], the approach is based on edge detection. Hough Transform was used for modern coins; while a modified version of Hough Transform with a measure of compactness was used for ancient coins.…”

Section: Tail Of Coinmentioning

confidence: 99%

“…While approaches in [13][14][15][16][17] cannot handle multiple images per page, the way they characterize coin images thanks to texture and shape information is highly relevant for our scenario. However, a major challenge to address is the risk of damaging coin borders in the segmentation process, as such approaches do.…”

Section: Tail Of Coinmentioning

confidence: 99%

Segmentation system and its evaluation for gray scale coin documents

Say

Coustaty

Chazalon

et al. 2014

2014 4th International Conference on Image Processing Theory, Tools and Applications (IPTA)

View full text Add to dashboard Cite

The numismatics community undergoes an important change toward a digital archiving of coin trades. Historic transactions of coins are tracked; many catalogues and the digitization of such content is of great interest for augmenting experts work in estimating a coin value and preventing illegal sales. In order to recognize and index coin pictures in transactions archives, a segmentation step is required. We propose a system which enables the automatic segmentation of coins and labels in the documents. As opposed to state of the art approaches, our system can extract coin images of any shape and handle multiple coins per image, as well as text labels. We also present in detail the evaluation metric, pixel correspondence graph, and the adaptation of the metric that we used to evaluate the performance of the proposed system. The evaluation of the proposed system on seven books (230 pages) shows promising results: over 99% of coins and 90% of labels are correctly segmented.

show abstract

Automatic Coin Classification and Identification

Cited by 10 publications

References 44 publications

Unsupervised Statistical Learning for Die Analysis in Ancient Numismatics

Unsupervised Statistical Learning for Die Analysis in Ancient Numismatics

Application of multi-modal 2D and 3D imaging and analytical techniques to document and examine coins on the example of two Roman silver denarii

Segmentation system and its evaluation for gray scale coin documents

Contact Info

Product

Resources

About