Related Experiment Video
Updated: Aug 16, 2026

Deployment and Retrieval of Mineral Samplers
Published on: January 20, 2026
Ancient Bengal numismatic dataset of Vanga Janapada, Gauda, Samatata, and Harikela coinage
Ishtiak Al Mamoon1, Ariful Islam1, Wafial Hasnat Laisa1
1Department of Computer Science and Engineering, International University of Business Agriculture and Technology, Dhaka, Bangladesh.
Abstract:
Ancient Bengal coinage represents one of the significant pieces of evidence for understanding the political, economic, and cultural development of Eastern South Asia. The study of coin analysis remains challenging because many types are rare, visually complex, dispersed across private collections, and difficult to identify without any specialized numismatics knowledge. Even though interest in applying machine learning and computer vision to numismatics research is growing, the absence of well-structured datasets has limited the computation research on the early coinage of Bengal. This study presents a novel image dataset along with metadata annotations of ancient Bengal coins, which cover a broad historical period from the 5th century BC to the 10th century AD. The dataset includes 2704 annotated coin images across 52 classes of four major numismatic traditions of Bengal: Vanga Janapada, Gauda, Harikela, and Samatata. Our collection captures the evolution of Bengal's monetary system from punch-marked coinage to the regional gold and silver coinage of the post-Gupta and early medieval periods. The images were collected from authenticated private collections and five permission granted auction houses. To ensure reliable class identification, each coin was carefully examined and annotated using historical classification, inscription, motifs, symbols, variety, and coin types. Additionally, all coins were checked and confirmed using numismatic sources, including SK Bose and Noman Nasir's numismatic books and auction house records. The dataset contains a wide range of rare and historically significant collections that provide a valuable resource for machine learning-based coin collection, feature extraction, image retrieval, and archaeological research. Moreover, the combination of modern computational methods with the ancient Bengal numismatics dataset offers a standardized benchmark for future interdisciplinary research in archaeology, numismatics, history, cultural heritage, and artificial intelligence.
More Related Videos
10:31Detection and Recovery of Palladium, Gold and Cobalt Metals from the Urban Mine Using Novel Sensors/Adsorbents Designated with Nanoscale Wagon-wheel-shaped Pores
Published on: December 6, 2015
07:05Use of Autometallography to Localize and Semi-Quantify Silver in Cetacean Tissues
Published on: October 4, 2018
Related Concept Videos
Base Quantities and Derived Quantities
The International Organization for...
Data Collection I
Data Collection II
Data Collection III
The principles to begin the physical assessment include conducting a comprehensive or problem-related history in a quiet, well-lit room, emphasizing privacy and comfort for the patient.
Sampling Distribution
Bulk Modulus