摘 要:指紋圖像采集過程常會造成對比度不強等非線性失真,基于模糊邏輯的處理方法常用于改善指紋圖像質量。研究了模糊特征平面增強算法和基于廣義模糊算子的圖像增強算法,將兩種算法應用于指紋圖像對比度增強,并對增強結果進行比較分析。實驗結果表明,采用這2種方法均可以在一定程度上提高指紋圖像低灰度區域和高灰度區域之間的對比度,從而提高圖像的質量,使增強后的指紋圖像結構更清晰。關鍵詞:指紋; 對比度增強; 模糊特征平面; 廣義模糊算子
中圖分類號:TP911-34; TP39141文獻標識碼:A
文章編號:1004-373X(2010)16-0140-03
Algorithms of Fingerprint Image Contrast Enhancement Based on Fuzzy Logic
CAI Xiu-mei
(School of Automation, Xi'an University of Posts and Telecommunications, Xi’an 710121, China)
Abstract: The acquisition process of fingerprint image often causes the nonlinear distortion such as low contrast. The algorithms based on the fuzzy logic are often used to improve the quality of fingerprint image. The image enhancement algorithms based on fuzzy property plane and Generalized Fuzzy Operator (GFO) are researched respectively. They are used to enhance the contrast of fingerprint images. The results of contrast enhancement are analyzed contrastively. The experimental results show that these two methods based on fuzzy logic can increase the contrast between the low gray level area and the high gray level area of a fingerprint image to a certain extent, and make the construct of the fingerprint image more clearly.Keywords: fingerprint; contrast enhancement; fuzzy property plane; generalized fuzzy operator
0 引 言
指紋識別是指指尖表面紋路的脊谷分布模式識別,這種脊谷分布模式是由皮膚表面細胞死亡、角化及其在皮膚表面積累形成的。人的指紋特征是與生俱來的,在胎兒時期就已經決定了。人類使用指紋作為身份識別的手段已經有很長歷史,使用指紋識別身份的合法性也己得到廣泛的認可。自動指紋識別系統通過比對指紋脊線和谷線結構以及有關特征,如紋線的端點和分歧點等來實現個人身份認證。然而,要從原始指紋圖像上準確地提取特征信息,這是十分困難的,在很大程度上特征提取的精確性依賴于圖像質量。因此,在指紋特征提取和匹配之前有必要對指紋圖像進行增強處理。指紋圖像增強就是對指紋圖像采用一定算法進行處理,使其紋理結構清晰化,盡量突出和保留固有的指紋特征信息,并消除噪聲,避免產生虛假特征。……