به نام خدا

Title: Video Based Palmprint Recognition

Authors: Chhaya Methani and Anoop M Namboodiri

Abstract: The use of camera as a biometric sensor is desir-able due to its ubiquity and low cost, especially for mo-bile devices. Palmprint is an effective modality in suchcases due to its discrimination power, ease of presen-tation and the scale and size of texture for capture bycommodity cameras. However, the unconstrained na-ture of pose and lighting introduces several challengesin the recognition process. Even minor changes in poseof the palm can induce significant changes in the vis-ibility of the lines. We turn this property to our ad-vantage by capturing a short video, where the naturalpalm motion induces minor pose variations, providingadditional texture information. We propose a method toregister multiple frames of the video without requiringcorrespondence, while being efficient. Experimental re-sults on a set of different 100 palms show that the use ofmultiple frames reduces the error rate from 12.75% to4.7%. We also propose a method for detection of poorquality samples due to specularities and motion blur,which further reduces the EER to 1.8%.   

Publish Year: 2010

Publisher: ICPR - IEEE

Number of Pages: 4

موضوع: تشخیص اثر کف دست

مشاهده صفحه اول مقاله

دانلود مقاله

لینک مقاله در سایت ناشر

ایران سای – مرجع مقالات علمی فنی مهندسی

حامی دانش بومی ایرانیان