2014年2月18日 星期二

指紋(fingerprint)辨識:最融入你我生活的生物辨識應用



上市沒多久的iPhone 5s最受大家期待的就是新的指紋辨識系統可以想見生物辨識Biometric)的應用技術正融入我們最日常的生活當中目前生物辨識核心技術的發展,指紋辨識佔技術比率54,簽名辨識佔技術比率約21%,臉部辨識則佔技術比率16%,虹膜辨識佔技術比率9%。市面上以「指紋辨識」技術較成熟,市場占有率最高,其次則為成長速度最快的「臉部辨識」技術。「虹膜辨識」的準確度最高,但是由於使用上必需以紅外線掃描眼球,在價格及安全性的考慮下,並不容易發展成為大眾化的產品,相對的市場占有率也就無法迅速拓展。其餘生物辨識科技則仍受一般消費者的使用習慣、可接受度以及經濟價格因素影響,成長較緩慢。(資料來源什麼是「生物辨識(Biometric)系統」?)

首先,我們透過IET Digital Library(電子電機全文資料庫)來從最全面的方式先掌握關於生物辨識(Biometric的發展與應用,因為它收錄IET期刊雜誌1994年以來之技術性論文超過50000篇。高引用率之期刊如Electronics Letters,囊括21份專業期刊、8本專業性雜誌,以及IET會議出版品及研討會摘要。使用者可回溯使用1994年起所出版的期刊並可瀏覽所有內容目錄及IET電子學會議記錄與摘要。而且新版Digital Library所收錄IET 期刊與雜誌等文獻與IET會議記錄、研討會摘要讓讀者保持在電子、電機、通訊、電腦、動力、雷達、電路、材料、生物工程及IT等科技產業發展的前線,獲取最新資訊。

接著,我們就以大家最熟悉「指紋(fingerprint)辨識」為例,一起看看此項技術的發展情況。過去,要收集指紋紀錄必須將十隻手指頭沾滿油墨,再一一拓印到白紙上;現在則是以特殊設備來完成指紋辨識例如光學感應器,搭配相關軟體,就可以儲存自己的指紋檔案或者是更先進的利用矽晶式的感應設備,藉由偵測指紋上所帶的正負電荷,可以描繪出包含紋路形狀、深淺的3 D立體影像同時為了確保接受辨識的指紋是來自活體,有些感應器還加裝溫度、溼度的感測而這些技術的研發與改善。相關查詢結果豐富且具有學術權威性,能成為研究此一科技技術發展與應用時不可或缺的有利參考資料。(以下列舉數篇文獻供參考)



Revisiting the accuracy of the biohashing algorithm on fingerprints
 
Patrick Lacharme , IET Biometrics, Volume 2, issue 3, 2013 , p. 130 – 133
Oriented diffusion filtering for enhancing low-quality fingerprintimages
C. Gottschlich, C.-B. Schönlieb. IET Biometrics, Volume 1, issue 2, 2012 , p. 105 - 113
Operational bio-hash to preserve privacy of fingerprint minutiae templates
Rima Belguechi, Estelle Cherrier, Christophe Rosenberger, Samy Ait-Aoudia. IET Biometrics, Volume 2, issue 2, 2013 , p. 76 - 84
Reducing descriptor measurement error through Bayesian estimation of fingerprint minutia location and direction
N.J. Short, A. Lynn Abbott, M.S. Hsiao, E.A. Fox. IET Biometrics, Volume 1, issue 1, 2012 , p. 82 - 90
Transformation invariant algorithm for automatic fingerprintrecognition
A. Noor, N. Manivanan, W. Balachandran. Electronics Letters, Volume 48, issue 14, July 2012, p. 834 - 835
You need hands [fingerprint-based biometrics]
S. Bains. IEE Review, Volume 51, issue 11, 2005 , p. 30 - 33
Fingerprint-based remote user authentication scheme using smart cards
J.K. Lee, S.R. Ryu, K.Y. Yoo. Electronics Letters, Volume 38, issue 12, June 2002, p. 554 - 555
Biohashing applied to orientation-based minutia descriptor for securefingerprint authentication system
L. Nanni, S. Brahnam, A. Lumini. Electronics Letters, Volume 47, issue 15, July 2011, p. 851 - 853
Image quality and position variability assessment in minutiae-basedfingerprint verification
D. Simon-Zorita, J. Ortega-Garcia, J. Fierrez-Aguilar, J. Gonzalez-Rodriguez. IEE Proceedings - Vision, Image and Signal Processing, Volume 150, issue 6, 2003 , p. 402 - 408
Wavelet based fingerprint liveness detection
Y.S. Moon, J.S. Chen, K.C. Chan, K. So, K.C. Woo. Electronics Letters, Volume 41, issue 20, September 2005, p. 1112 - 1113
Fingerprint recognition using DCT features
T. Amornraksa, S. Tachaphetpiboon. Electronics Letters, Volume 42, issue 9, April 2006, p. 522 - 523
Effective algorithm for rolled fingerprint construction
Jie Zhou, Di He, Gang Rong, Zhao-qi Bian. Electronics Letters, Volume 37, issue 8, April 2001, p. 492 - 494
Adaptive image normalisation based on block processing for enhancement of fingerprint image
Byung-Gyu Kim, Dong-Jo Park. Electronics Letters, Volume 38, issue 14, July 2002, p. 696 - 698
Further cryptanalysis of fingerprint-based remote user authentication scheme using smartcards
W.C. Ku, S.T. Chang, M.H. Chiang. Electronics Letters, Volume 41, issue 5, March 2005, p. 240 - 241
Latent fingerprint segmentation using ridge template correlation
N.J. Short, M.S. Hsiao, A.L. Abbott, E.A. Fox. 4th International Conference on Imaging for Crime Detection and Prevention 2011 (ICDP 2011)2011, p. 28 - 28
Fingerprint verification using wavelet and local dominant orientation
B.D. Patil, J.V. Kulkarni, R.S. Holambe. IET International Conference on Visual Information Engineering (VIE 2006)2006, p. 79 - 82
Combining level-3 features with perspiration pattern for robustfingerprint recognition
A. Abhyankar. National Conference on Signal and Image Processing Applications, 2009, p. 1 - 1
Fingerprint classification by ridge orientation and singular point analysis
S. Mohammadi, S. Fazli, A. Farajzadeh. National Conference on Signal and Image Processing Applications, 2009, p. 14 - 14
A graphical approach for fingerprint verification
S. Jain, S.K. Mitra, A. Banerjee, A.K. Roy. IET International Conference on Visual Information Engineering (VIE 2006), 2006, p. 68 - 73
Assessing image characteristics for user feedback in biometricfingerprint identity verification tasks
M.C. Fairhurst, C. McIntosh. IEE International Conference on Visual Information Engineering (VIE 2005), 2005, p. 135 - 140
Novel fingerprint segmentation with entropy-Li MCET using log-normal distribution
D.H. AlSaeed, A. Bouridane, A. ElZaart, R. Sammouda. IET Conference on Image Processing (IPR 2012), 2012, p. 6 - 6
 


以準確度來說,「生理特徵」在唯一性及安全性上明顯優於「行為特徵」因此我們可以預見這項科技未來將逐漸成為我們生活中不可或缺的應用。而透過IET Digital Library(電子電機全文資料庫)提供的這些學術研究的研討論文或期刊文獻,都可以幫掌握生物辨識系統的發展趨勢同時可洞悉未來的應用層面會是全面了解生物辨識技術的最佳研究資源之一

【撰文●Nicky


網路資源:


4.           Biometrics: Overview. Biometrics.cse.msu.edu. 6 September 2007. Retrieved 2012-06-10.
5.           Introduction to Biometrics. Jain, Anil K.; Ross, Arun (2008).  ISBN 978-0-387-71040-2.
6.           A Survey of Biometric Recognition Methods  Delac, K., Grgic, M. (2004). 
7.           Biomtrics Institute Privacy Code, September 2006
12.        指紋辨識專題介紹