應用渦電流感測及統計圖樣分析法於硬幣識別技術

Application of eddy-current sensing and statistical pattern analysis to coin identification

張文柳(Wen-Liu Chang)

指導教授 ： 呂志誠 ； 鄭振宗

國立臺北科技大學/機電學院/機電整合研究所/碩士(2011年)

https://doi.org/10.6841/NTUT.2011.00685

摘要

在本篇的研究中，我們提出了一種使用渦電流偵測器及統計圖樣分析來辨識硬幣真偽的方法，實驗的模組包含一組模擬投幣裝置，兩組渦電流感測器，兩組溫度補償電路以及硬幣學習及辨識的介面程式。此程式是利用NI LabVIEWR軟體開發，透過資料擷取卡讀取渦電流偵測器的輸出訊號，據以建立資料及辦識硬幣。影響辨識的主要因素為硬幣的直徑、厚度以及材質。實驗結果顯示，民國九十五年代的真幣標準對各年代的真幣分類及偽幣辨識有較好的結果。為解決因溫度變動所造成的電壓偏移，我們設計了兩組溫度補償電路，分別量測環境溫度和電路盒內的溫度，經過了溫度補償，偽幣及真幣的辨識率可以有效的提升。

關鍵字

渦電流 ； 硬幣辨識 ； 溫度補償

並列摘要

An effective method based on eddy-current sensing and statistical pattern analysis is proposed to discriminate the authenticity of different coin classes. By choosing diameter, thickness, and materials of coins as the major factors of recognition, an experimental module featuring two eddy-current sensors and an interface program for rapid coin identification was built. The variation of identification standard caused from different defaced coins was also discussed. The coin learning and recognition programs are developed by NI LabVIEWR software. To achieve a high recognition rate, a 2D voltage map created by ten thousand times statistics for each set of coins was proposed. On the other hand, to solve the voltage offset caused by temperature dependence, two sets of temperature compensation circuit were designed to sense the environmental temperature and the circuit box temperature, respectively. With thermal compensation, the recognition rate for counterfeit tokens can be further increased.

並列關鍵字

coin recognition ； Eddy current ； temperature compensation

參考文獻

[6] Fukumi, M., Omatu S, Takeda F, Kosaka T, " Rotation-invariant neural pattern recognition system with application to coin recognition," IEEE transaciions on neural networks, volume 3, no. 2, March 1992, Pages 272-279.

[8] Minemasa Hida, Hiroyasu Sato, Hiroshi Sugawara, Toshiyuki Mitsui," Classification of counterfeit coins using multivariate analysis with X-ray diffraction and X-ray fluorescence methods," Forensic Science International, Volume 115, Issues 1-2, January 2001, Pages 129-134.

[9] Adnan Khashman, Boran Sekeroglu, Kamil Dimililer," Intelligent Coin Identification System," International Symposium on Intelligent Control, October, 2006, Pages 1226-1230.

[10] Marcel Tresanchez, Tomas Palleja, Merce Teixido, Jordi Palacin," Using the optical mouse sensor as a two-Euro counterfeit coin detector," Sensors, September 2009, Pages 7083-7096.

[11] Vaibhav Gupta, Rachit Puri, Monir Verma," Prompt Indian Coin Recognition with Rotation Invariance Using Image Subtraction Technique," Devices and Communications (ICDeCom), 2011 International Conference, February, 2011, Pages 1-5.