Using the Optical Mouse Sensor as a Two-Euro Counterfeit Coin Detector 英文参考文献.docVIP

Sensors 2009, 9, 7083-7096; doi:10.3390/OPEN ACCESS sensors ISSN 1424-8220 /journal/sensors Article Using the Optical Mouse Sensor as a Two-Euro Counterfeit Coin Detector Marcel Tresanchez, Tomàs Pallejà, Mercè Teixidó and Jordi Palacín* Department of Computer Science and Industrial Engineering, University of Lleida, Jaume II, 69, 25001 Lleida, Spain; E-Mails: mtresanchez@diei.udl.cat (M.T.); tpalleja@diei.udl.cat (T.P.); mteixido@diei.udl.cat (M.T.) * Author to whom correspondence should be addressed; E-Mail: palacin@diei.udl.cat; Tel.: +34-973-702724; Fax: +34-973-702702. Received: 7 August 2009; in revised form: 20 August 2009 / Accepted: 21 August 2009/ Published: 4 September 2009 Abstract: In this paper, the sensor of an optical mouse is presented as a counterfeit coin detector applied to the two-Euro case. The detection process is based on the short distance image acquisition capabilities of the optical mouse sensor where partial images of the coin under analysis are compared with some partial reference coin images for matching. Results show that, using only the vision sense, the counterfeit acceptance and rejection rates are very similar to those of a trained user and better than those of an untrained user. Keywords: optical sensor; optical mouse sensor; vision based counterfeit 1. Introduction A counterfeit coin is an imitation of a genuine coin made with the intent to defraud. Currently, automatic counterfeit coin detection in cash or vending machines is based on the use of low cost sensors to measure some physical coin properties [1] such as weight, size, thickness, conductivity [2], magnetic [3] or acoustic properties [4,5]. Alternatively, when the physical properties of the coins (weight, size) are very similar, image based coin recognition can be used. For example, image based recognition was applied to ancient coin classification in [6], and in [7] coin image recognition