Abstract
Original language | English |
---|---|
Title of host publication | Proceedings of CAA-2006 |
Place of Publication | Fargo, ND, USA |
Publisher | Unknown Publisher |
Pages | online |
Publication status | Published - 2006 |
Fingerprint
Cite this
}
Computer vision and machine learning for archaeology. / van der Maaten, L.J.P.; Boon, P.; Lange, G.; Paijmans, J.J.; Postma, E.
Proceedings of CAA-2006. Fargo, ND, USA : Unknown Publisher, 2006. p. online.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review
TY - GEN
T1 - Computer vision and machine learning for archaeology
AU - van der Maaten, L.J.P.
AU - Boon, P.
AU - Lange, G.
AU - Paijmans, J.J.
AU - Postma, E.
PY - 2006
Y1 - 2006
N2 - Until now, computer vision and machine learning techniques barely contributed to the archaeological domain. The use of these techniques can support archaeologists in their assessment and classification of archaeological finds. The paper illustrates the use of computer vision techniques for archaeology with two examples: (1) a content-based image retrieval system for historical glass and (2) an automatic system for medieval coin classification. The content-based image retrieval system automatically finds artifact drawings in a reference collection that are most similar to a photograph or drawing of an excavated historical glass. The similarity measurements are based on the outer shape contours of the artifacts. The system can speed up the process of classifying historical glass, and make it more objective and controllable. The coin classification system will be trained on a collection of Dutch early-medieval coins. For this system, we present preliminary results on modern coin data.
AB - Until now, computer vision and machine learning techniques barely contributed to the archaeological domain. The use of these techniques can support archaeologists in their assessment and classification of archaeological finds. The paper illustrates the use of computer vision techniques for archaeology with two examples: (1) a content-based image retrieval system for historical glass and (2) an automatic system for medieval coin classification. The content-based image retrieval system automatically finds artifact drawings in a reference collection that are most similar to a photograph or drawing of an excavated historical glass. The similarity measurements are based on the outer shape contours of the artifacts. The system can speed up the process of classifying historical glass, and make it more objective and controllable. The coin classification system will be trained on a collection of Dutch early-medieval coins. For this system, we present preliminary results on modern coin data.
M3 - Conference contribution
SP - online
BT - Proceedings of CAA-2006
PB - Unknown Publisher
CY - Fargo, ND, USA
ER -