在這多元化的社會中，人們生活的方式也越來越複雜，在很多情況下，安全保密以及身分辨識是一項相當重要的課題。早期是使用密碼和鑰匙來當作個人的保密工具，印章和簽名來當作身分辨識。但是這類的工具存在著不方便性、易遺失、遺忘、盜用等問題因素。由於人的身體特徵具有不可複製的特點，因此學者們都在探討，能否以生物特徵，來當作個人身份識別的工具？在生物識別的領域中，指紋因為本身的唯一性和不變性而具有相當高的可靠度。此外，指紋辨識系統開發成本較低、使用方便等種種優越之處使得它已被廣泛的運用在資料保全系統上。 大部分的指紋辨識方法是以指紋的奇異點(核心點、三角點)及紋線的特徵點(端點、叉點)的分佈位置、數目作為判別的依據。而若能將指紋流向場精確的求得，那要擷取出特徵點以及核心點將是一件容易的事情。因此許多的學者都提出了計算流向的方法，最常使用的方法不外乎Stock與Swonger所提出的Slit-sum，以及利用梯度方向來得到的指紋流向。而在本文將提出一種新的方法來得到指紋流向場。此方法可以加速指紋流向的計算，並更準確的得到流向，加速指紋比對的速度以及成功率。最後在本論文中，將所提出之方法，利用可程式化系統晶片(SOPC)來實現個人身份辨識系統。 In this diversified society, people''s life are getting more and more complicated. In many situations, keeping secret safely and identifying recognition are quite important lessons. In early days, people kept secret by using secret code and keys and recognized identifications by seals and signatures. But these ways are inconvenient and easily lost, forgotten and stolen. Because the characters of human bodies can not be copied, scholars discuss about that if we can use these characters to recognize the identifications. In the area of Biometric recognitions, the fingerprint is so authentic because of its singularity and constancy. Besides, the fingerprint is used generally in the information saving system because of the advantages of the fingerprint recognition system, such as cheaper cost in development and convenient using way.
Most identification methods are dependent on the locations and numbers of ridge endings and bifurcations. If we can get the fingerprint flow field exactly, it will be easy to find out the minutiae and singular points. As a result there are many scholars propose many ways to calculate flow. The ways mostly used are Slit-sum methods proposed by Stock and Swonger. The fingerprint flow is got by using gradient. In this thesis, we propose a novel way to calculate the fingerprint flow field. This method can accelerate the speed of the calculation of fingerprint flow and help us gain the flow more accurately. Therefore the speed of fingerprint recognition is accelerated and the success rate is increased. Finally, we implement it to build an automatic fingerprint identification system via a Nios embedded processor.
