Dry fingerprint detection for multiple image resolutions using ridge features

摘要: Dry and wet fingers lead to poor fingerprint quality, which means that it has impact for recognition matching. Recognition methods are based on the feature of ridge, valley, minutiae or pore affected by skin conditions. In this paper, we propose a novel dry detection method images with different resolutions using ridge features. The fingerprints have vague pores discontinuous fragmented ridges. Therefore, features adopt continuity, fragmentation ridge/valley ratio. These can be observed clearly under image resolutions, so our proposed work 500∼1200 dpi. We several use support vector machine classify into two groups, normal. NASIC database (1200dpi) FVC2002 DB1 (500dpi) used in experiments, SVM classification accuracy 99.00%, 99.09% relatively.