An Image Quality Assessment Algorithm for Palm-Dorsa Vein Based on Multi-Feature Fusion

Abstract

In order to acquire more feature information in captured vein image with high-quality, an image quality assessment algorithm for palm-dorsa vein is presented based on multi-feature fusion. According to the statistical and structure properties of image, we can acquire the good vein image by adjusting near-infrared LED light exposure duty cycle combined with computing the four characteristic parameters of gray variance, information entropy, cross point and area of effect and then fusing those according to weights. In the end, an experimental case is given and the assessment results prove that this algorithm is efficient.