Some portions of dorsal hand may be occluded due to injuries, pigmentation, or tattoos, which significantly affects the performance of dorsal hand vein recognition systems. Biometric graph matching is a common shape-based feature extraction algorithm for vein recognition. However, this method does not consider edge attributes, which can provide additional discrimination ability. We present an improved biometric graph matching method that includes edge attributes for graph registration and a matching module to extract discriminating features. Moreover, we propose a recognition system for partially occluded dorsal hand vein. A database of normal hand vein images, three databases of images with artificially occluded dorsal hand vein with occlusions in different positions and ratios, and a database of images with tattooed hands are established to verify the validity of the proposed method. The experimental results demonstrated that the equal error rates and the accuracies were 0.0202 and 98.09% ± 0.28%, respectively for the normal hand vein images, 0.0453 and 96.58% ± 0.34%, respectively for images of artificially occluded dorsal hand vein with occlusion at all positions and area ratios (0 − 20%, mean occluded area ratio = 9.3%), and 0.0343 and 97.14% ± 0.29%, respectively for the images of tattooed hands. INDEX TERMS Dorsal hand vein recognition, biometric graph matching, occlusion, databases.
Abstract. In order to extract better vein feathers, image preprocessing is necessary. So this paper propose a new approach to enhance images. The enhancement algorithm uses guided filter (GF) to process hand vein images. The guided filter is used as an edge-preserving smoothing operator. The guided filter enhancement algorithm is effective comparing with bilateral filter (BF), histogram equalization (HE), adaptive histogram equalization algorithm (AHE) and contrast limited adaptive histogram equalization (CLAHE). We use several methods to enhance dorsal hand vein images, the recognition rate with guided filter is the best. For the security, a fake vein detection algorithm is used to discriminate the real vein and fake vein images.
Vein texture elements in images are sometimes degraded due to centrally symmetric blur, e.g. due to motion of hand or camera not being in focus, which significantly affects recognition systems performance. Multiscale local phase quantisation (MLPQ) can effectively improve system accuracy in the presence of blur by dividing vein images into non-overlapping blocks and extracting multiscale local phase information. However, MLPQ misses some important discriminant information at the intersections of different regions, which leaves room for improving the system performance. In the present work, a multiscale overlapping blocks local phase quantisation (MOLPQ) histogram algorithm is presented to divide the vein image into overlapping blocks and extract multiscale local phase information, which includes the discriminant information lost in MLPQ. MOLPQ is validated via comparison with state-of-the-art recognition algorithms on a normal hand vein database, artificial-blur databases, and a normal-blur database, and the accuracies on the normal hand vein and normal-blur databases are 99.57 and 98.07%, respectively. Additionally, MOLPQ outperforms other methods on all databases in terms of accuracy.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.