以簡單線性迴歸分析實作掌紋辨識系統

PALMPRINT RECOGNITION BY SIMPLE LINEAR REGRESSION

李佳勳(Chia-Hsun Lee)

指導教授 ： 虞台文

大同大學/電機資訊學院/資訊工程學系所/碩士(2007年)

摘要

現今，以生物辨識做為身分識別越來越被被人們所接受，例如：指紋、掌紋、虹膜、臉部等特徵都可以做為辨識依據，其中手部辨識擁有高度方便性，未來勢必以手部辨識為主。
本研究將實作一個掌紋辨識系統，並且提出一些新的想法用於掌紋辨識，首先改進USAN這個擷取掌紋的方法，再提出一個快速搜尋掌紋主線的方法，並且利用簡單線性迴歸將掌紋特徵數值化。辨識系統使用USAN將原始掌紋影像轉換成一個三維空間，稱為USANarea，再使用threshold將掌紋擷取出來，擷取出來的掌紋經過去雜訊與close處理後，轉換成以線段表示的圖形，再搜尋出三條掌紋主線，掌紋主線分別經過簡單線性迴歸分析，最後整張手掌影像將以六個參數代表，大幅降低資料庫的複雜度，處理過程儘量減少複雜的運算，整個辨識過程可real-time完成。本研究以實際手掌影像進行實驗，並且將辨識率與執行時間與其他研究做比較。

關鍵字

掌紋辨識 ； 簡單線性迴歸分析 ； 掌紋主線

並列摘要

This paper proposes a palmprint recognition method based on simple linear regressionto typical principal lines on a palm. By USAN line finder, the original palmprintimages are transformed into a three dimensional space, called “USAN area”, whichcan extract palmprint by threshold. Then, palmprint are classified into three typicalprincipal lines and several wrinkles.Finally, typical principal lines are fit into linefunctions respectively, which represent features of a palmprint image. Thus, we can usesix floating points to represent the global pattern of one palmprint image. Complexityof database can be simplified effectively.

並列關鍵字

Simple linear regression ； Typical principal lines ； Palmprint recognition ； USAN

參考文獻

[1] A.K. Jain and N. Duta, “Deformable matching of hand shapes for verification”, InternationalConference on Image Processing, vol. 2 ,pp. 857-861, 1999.

[2] J. Doi and M. Yamanaka, “Discrete Finger and Palmar Feature Extraction for Personal Authentication”, IEEE Transactions on Instrumentation and Measurement, vol. 54, pp. 2213-2219, 2005.

[3] D. Zhang, W.K. Kong, J. You and M. Wong, “Online Palmprint Identification”, IEEE Transactions on Instrumentation and Measurement, vol. 25, pp. 1041-1050, 2003.

[4] “PolyU Palmprint Database”, http://www.comp.polyu.edu.hk/ biometrics.

[5] D. Zhang and W. Shu, “Two novel characteristics in palmprint verification: datum point invariance and line feature matching”, Pattern Recognition Letters, vol. 32, pp. 691-702, 1999.