高效能非正向視角虹膜辨識系統

近年來, 由於傳統身份辨識方法的諸多問題, 使得利用個人生物特徵來辨識身份的技術逐漸受到重視。由於虹膜辨識技術有良好的便利性與辨識率, 使其在眾多的生物辨識技術當中脫穎而出, 廣為使用。傳統的虹膜辨識系統均是假設使用者在正視攝影機的情況下擷取虹膜影像, 然而此項假設會限制住虹膜辨識的應用範圍。本論文將深入探討使用在非正向的情形下所攝得之虹膜來辨識個人身份的技術。本研究的目標有二: 一為利用即有的硬體架構取得不同視角的近紅外光與可見光彩色的四頻譜虹膜影像, 並且改良舊有系統中虹膜影像品質評估的方法, 對虹膜定位與特徵比對的部份做最佳化, 使得系統能夠快速的進行辨識, 並且依舊擁有良好的辨識率。二為驗証仿射矯正(Affine Rectification) 是否適用於將非正向的虹膜影像轉換為正向視角的虹膜影像。實驗結果顯示, 改進後的系統不但有較高的效率, 也具有較好的辨識效果。

關鍵字

生物特徵；虹膜辨識；非正向視角；仿射轉換

並列摘要

In recent years, techniques for personal identification with biometrics have drawn more and more attention from people due to the insecurity of the traditional authentication methods. Because of the convenience and high recognition accuracy, iris recognition techniques have stood out among the biometric personal identification techniques and are widely adopted in many applications. Most of the existing iris recognition systems are developed under the assumption that the iris images are acquired with the orthogonal view imaging condition (OVIC). This assumption has restricted the applicability of iris recognition systems. In this thesis, we will discuss in depth an iris recognition technique that considers the non-orthogonal view iris image for personal identification. The purpose of this study is twofold. The first is to improve the iris image quality assessment method and to optimize the iris localization and classification process of our pervious four-spectral iris recognition system. The second is to verify whether the affine rectification is appropriate to convert a non-orthogonal view iris image into an orthogonal view image. Experiments show that the improved system has higher efficiency and better recognition accuracy.