透過您的圖書館登入
IP:3.17.156.160
  • 學位論文

鑑別個人身份的手掌影像辨識之演算法

Palm Image Recognition Algorithm for Personal Identification

指導教授 : 許志宇
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


安全認證是近年來越來越受關注的議題,而生物識別技術是應用於各種不同領域的安全認證,現今已提出來的生物識別技術大多是以提取掌紋和掌形等不同的型式來做為生物特徵值,而且須要有固定手掌位置或是手掌正規化問題,這些問題局限了手掌識別的應用範圍。 本論文中我們提出了一個手掌影像辨識鑑別個人身份的方法,希望利用手掌影像能夠分辨出是否為同一人,讓手掌影像辨識能減少限制而有更多的應用。個人身份的鑑別技術是利用手指指谷的四個點所形成的3個三角形面積做為影像特徵值,將收集來的手掌影像進行前置處理,利用HSV(Hue, Saturation, Value)彩色模型,把收集來的手掌影像進行轉換得到S (Saturation)頻道的影像,再用Otsu’s選取門檻值將灰階影像轉為二值影像,接著利用數學形態學的運算如: 二值化、侵蝕、膨脹…等方法去除雜訊後,最後使用LOG(Laplacian of the Gaussian)邊緣檢測將手掌邊緣圈選出來,取得手指指谷所形成的4個三角形面積當作特徵值並且存入資料庫,以供資料做比對,進而做到鑑別個人身份。

並列摘要


Security certification is the issue of increasing concern in recent years, the Biometric is used in a variety of different areas of safety certification, Has now been raised to biometric technology mostly is done in different type of extraction of palm and palm biometric value, and need to have a fixed position of the palm of your hand or palm regularized problems, which limited the scope of application of the palm recognition. In this paper we propose palm image recognition to identify their personal capacity, we hope to use the palm of your hand image to be able to distinguish whether the same person, so that the palm image recognition can be less restrictive and more applications. This personal identity authentication technology is finger four points of the valley formed by four triangular area as the image feature will be collected palm image pre-processing, the use of HSV (Hue, Saturation, Value) color space , we are collected the palm image conversion S (Saturation) image of the channel, then select Otsu’s threshold to the grayscale image to binary image, and then we using mathematical morphology operations such as: binariy, erosion, expansion and so on, they can to remove the noise, and finally use the LOG ( Laplacian of the Gaussian) edge detection hand edge of the circle, made of four triangular area formed by the finger valley as a feature and stored it in the database, it has be for information and then do to identify their personal capacity.

參考文獻


[15]王蔚強,基於形態學之手掌身份辨識,碩士論文,暨南國際大學電機工程系,南投,2008。
[1]D. Kumar and Y. Ryu, “A Brief Introduction of Biometrics and Fingerprint Payment Technology,” IJAST, 2009, pp.25-38.
[2]A. Kong, D. Zhang and M. Kamel, “Palmprint Identification Using Feature-Level Fusion,” Pattern Recognition, 2006, pp. 478-487.
[3]A. Kumar, C. M. Wong, C. Shen and K. Jain, “Personal Verification using Palmprint and Hand Geometry Biometric,” AVBPA, 2003.
[4]A. Kumar and A. Passi, “Comparison and Combination of Iris Matchers for Reliable Personal Identification,” CVPRW, 2008, pp. 1-7.

延伸閱讀