دانلود رایگان ترجمه مقاله نمایه سازی اثر انگشت بر اساس همبستگی نقطه منفرد (نشریه IEEE 2005)

شناسایی اثر انگشت بر اساس همبستگی نقاط تکین

Fingerprint Indexing Based on Singular Point Correlation

Abstract

Fingerprint indexing is an efficient technique that greatly improves the performance of Automated Fingerprint Identification Systems. We propose a continuous fingerprint indexing method based on location, direction estimation and correlation of fingerprint singular points. Location and direction estimation are achieved simultaneously by applying a T-shape model to directional field of fingerprint images. The T-shape model analyzes homocentric sectors around the candidate singular points to find lateral-axes and further main-axes. Then a distortion-tolerant filter of Minimum Average Correlation Energy is utilized to obtain a correlation-based similarity measure which gives the evidence of searching priority. The experiment is performed by 400-fingerprint retrieval from 10,000 templates and the mean search space is only 3.46% of the whole dataset.

1 Introduction

The huge amount of data in large fingerprint databases (e.g. several million fingerprints) seriously compromises the efficiency of the fingerprint identification task in Automated Fingerprint Identification Systems (AFIS) for both forensic and civil applications. There are two technical choices to reduce the number of comparisons during fingerprint retrieval and consequently to reduce the response time of the identification process: one is classification and the other is indexing.

Traditional classification techniques [1-3] attempt to classify fingerprints into five classes: Right Loop (R), Left Loop (L), Whorl (W), Arch (A), and Tented Arch (T). Due to the uneven natural distribution, comparatively large inter-class similarity and intra-class difference, the workload reduction resulted from classification is not gratifying.

Fingerprint indexing algorithms select most probable candidates and sort them by the similarity to the input one [4]. For indexing technique performs better than exclusive classification considering the size of space that need to be searched [5]. Many indexing algorithms have been proposed recently. In [4] and [6], the triplets of minutiae are used in the indexing procedure. These methods focus on the detailed information of fingerprints and ignore the macro information which is more robust to local noise. A.K. Jain et al [7] use the features around a core point of a Gabor filtered image to realize indexing. Although this approach makes use of the core point information but the discrimination power of just one core is limited. We also can see the efforts on combining methods, such as [8] and [9].

As a sort of prominent and global feature, singular points (SPs) in fingerprint images can be robustly identified and contain fingerprint intrinsic features. According to this fact, we propose an indexing approach based on SP correlation. SP detection and direction estimation are achieved simultaneously by applying a T-shape model to directional field (DF). The Tshape model reveals the intrinsic nature of SPs including cores and deltas which broadly exist in fingerprint images but are seldom utilized in fingerprint indexing. Then the Minimum Average Correlation Energy (MACE) filter [10], a kind of distortion-tolerant filter, is used to synthesize templates and perform correlation computation to give the similarity measurement. Further indexing is obtained by sorting the similarity between the query image and all stored templates. This paper is organized as follows. In Section II, a so-called T-shape model is introduced and utilized to detect SPs and to estimate their directions. Then the MACE filter is introduced in Section III. In Section IV, some experimental results are presented. Finally, the conclusion is drawn in Section V.