Hand Vein Authentication System Using Dynamic ROI

Abstract

Abstract. This paper presents an efficient authentication system based on hand vein pattern. The stages involved in vein pattern authentication system are image acquisition, Region of Interest (ROI) Extraction, image enhancement, binarization, thinning, feature extraction and matching. We propose an algorithm for extraction of dynamic ROI from the hand vein image. The advantage of dynamic ROI extraction is that, ROI extracted for different hand images varies in size as the size of the hand varies and is possible to extract more features from a larger hand which otherwise is not possible with fixed ROI. A new thinning algorithm is used to extract one pixel thick medial axis vein network from the dynamic ROI and compared the results with matlab's thinning algorithm. The resulting thinned image may contain some artefacts, and we propose an algorithm to remove these artefacts. The minutiae features that represents the geometric information of the vein pattern is extracted which are bifurcation and ending points. Finally a matching algorithm is applied for authentication. The proposed system is efficient and got the lowest error rate.

Dynamic ROI extraction method for hand vein images

“…Furthermore, area and entropy of our proposed method are always the best compared to others methods. A significant percentage of improvement for area and entropy measures, respectively, is provided by our proposed method: 51.25 and 11.52% compared to the Trabelsi et al method [10], 48 and 15.3% compared to the Yakno et al method [7], and 37.58 and 13.54% compared to the Prasad et al method [8, 9], for Vera database. For Bosphorus database: the percentage improvement accomplished by our proposed method for area and entropy measures, respectively, are ∼5.5 and 4.37% compared to the Trabelsi et al method [10], are ∼22.5 and 2.78% compared with the Prasad et al method [8, 9], 33.9 and 3.14% compared with the Yakno et al method [7].…”

“…9. This figure presents a comparison between our proposed method and the three previous methods: our proposed method and Prasad et al method [8, 9] produce a rectangular ROI, the Yakno et al method [7] produces a square ROI, and Trabelsi et al method [10] produces a circular ROI. All anteriors methods have provided smaller ROI sizes than those extracted by our proposed method while respecting hand borders.…”

“…For Vera database, our proposed method has a greater ROI extraction accuracy rate (90%) than previously described methods: 73% by the Yakno et al method [7], 52% by the Trabelsi et al method [10], and 75% by the Prasad et al method [8, 9]. If we look at the different methods, we notice that our proposed method upgrades the ROI extraction accuracy rate significantly by 17, 38, and 15% in comparison with the Yakno et al method [7], Trabelsi et al method [10], and Prasad et al method [8, 9], respectively.Fig. 8 illustrates the detected and extracted ROI from 12 and 20 samples of the same subject from Bosphorus and Vera databases, respectively.…”

“…Prasad et al [8, 9] proposed a dynamic ROI in order to select more features from the hand. They begin with a global thresholding binarisation.…”

Robust and adaptive ROI extraction for hyperspectral dorsal hand vein images

“…After an extensive literature review, we found that there are four categories of dorsal hand ROI extraction methods. The first category is Centroid-based [12,13,[21][22][23][24], followed by Knuckles- [3,[7][8][9]25], Valley points-based [26][27][28][29][30], and boundary based [31]. In Centroid-based methods, the centroid and shearing factor of the binarised hand image are first computed before drawing a square ROI (see Fig.…”