Ratha N., Karu K, Chen S. and Jain A., "A Real-Time Matching System for Large Fingerprint Databases", IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 18, no. 8, pp.799-813, 1996  Home/Search   Document Details and Download   Summary   Related Articles   Check

....into background (clear gray) noisy areas (dark gray) ridges (black) and valleys (white) The performance of the segmentation process was numerically assessed using the accuracy of our verification system. Indeed, we have developed a matching algorithm [8] based on a generalized Hough transform [9] and a similarity metric that takes the geometric relationships between minutiae into account. For a given database, the distribution of the matching scores for the same fingers and for different fingers is computed. Setting different values of the threshold on the matching score, we obtain the ....

- N.K Ratha, K. Karu, S. Chen and A.K Jain, "A Realtime Matching System for Large Fingerprint Databases", IEEE Trans. PAMI, vol. 18, no. 8, pp.799-813, 1996.

....[1, 2] b) Criminal Investigation: Corresponding author. On leave at Carnegie Mellon. Partially supported by the National Science Foundation under grants EEC 94 02384, IRI 9625428. A police officer would like to retrieve images resembling the face or the fingerprint image of a suspect [3, 4], c) Trademark Copyright Detection: Given a trademark image, we would like to detect if it already exists in the database [5] Education, home entertainment systems, remote sensing, astronomy, cartography and defense, are just a few more applications with great interest. Consequently, ....

....images. A desirable feature common to many systems is the adaptive behavior to retrievals through user relevance feedback and iterative query refinement (e.g. 8] Additional work on IDB systems and content based image retrieval include, the work by Ratha, Karu, Chen and Jain for fingerprints [4], the work by Mehrota and Gary for shapes [18] and the work by Mehtre, Kankanhalli and Lee for trademark authentication [5] The methods referred to above can be extended to handle video: A video is regarded as a sequence of related image frames from which the keyframes (i.e. frames ....

Nalini K. Ratha, Kalle Karu, Shaoyun Chen, and Anil K. Jain. A Real Time Matching System for Large Fingerprint Databases. IEEE Transactions on Pattern Analysis and Machine Intelligence, 18(8):799--813, August 1996.

....Lower Bounding Principle. Index terms: image database, image indexing, range query, nearest neighbor query, attributed relational graph, spatial access method. 1 Introduction The management of large volumes of digital images in many application fields (e.g. medicine [1] criminal investigation [2], trademark copyright detection [3] etc. has generated additional interest in methods and tools for real time archiving and retrieval of images by content. Consequently, content based image retrieval has become the object of intensive research activities over the past few years [4, 5] Several ....

....images by spatial and temporal content. A desirable feature common to many systems is the adaptive behavior to retrievals through user relevance feedback and iterative query refinement (e.g. 20] Additional work on IDB search and retrieval includes the work by Ratha et.al. for fingerprints [2]. Das et.al. 21] and Huang et.al. 22] suggest using the color image attributes as measurements of image content in image databases. Mojsilovic et.al. 23] focuses on the perception of color by humans and proposes a method that attempts to simulate human behavior in matching and retrieving color ....

[Article contains additional citation context not shown here]

N. K. Ratha, K. Karu, S. Chen, and A. K. Jain. A Real Time Matching System for Large Fingerprint Databases. IEEE Trans. on Patt. Anal. and Mach. Intell., 18(8):799--813, Aug. 1996.

....that the ARG editing distance, although the slowest, is the most accurate method. Key Words: Image DataBase, Image retrieval, Attributed Relational Graphs. 1. Introduction The management of large volumes of digital images in many application fields (e.g. medicine [1] criminal investigation [2], trademark copyright detection [3] etc. has generated additional interest in methods and tools for real time archiving and retrieval of images by content. Consequently, content based image retrieval has become the object of intensive research activities over the past few years [4, 5] Several ....

.... two ARGs can also be formulated as an assignment problem, that is a problem of finding the best association between the nodes (objects) of a query Q and the nodes (objects) of a reference [0] 0] 1] 1] 1] 1] 1,2) 1,3) 2,2) 2,1) 2,3) 2,1) 2,2) 3,1) 3,2) 3,1) 3,3) 3,3) 1] 1] [2] [2] 1] 2] 2] 2] 2] 2] 0] 2] 7] 6] 5] 7] 0 1] 0 1] 0] 0] 1,1) 1] 2,3) 1] 1] 3,2) 0] 4] 0 1] 0 1] BEST COST [0 1] 0 0] 5] Figure 4. Matching tree. image I (the relationships are ignored) To compute this association, we need the costs ....

[Article contains additional citation context not shown here]

Nalini K. Ratha, Kalle Karu, Shaoyun Chen, and Anil K. Jain. A Real Time Matching System for Large Fingerprint Databases. IEEE Transactions on Pattern Analysis and Machine Intelligence, 18(8):799--813, August 1996.

....that match. A typical minutiae extraction technique performs the following sequential operations on the fingerprint image: i) fingerprint image enhancement, ii) binarization (segmentation into ridges and valleys) iii) thinning, and (iv) minutiae detection. Several commercial [112] and academic [131, 11] algorithms follow these sequential steps for minutiae detection. Alternative techniques for minutiae detection directly operate 152 on the gray scale fingerprint image itself and detect minutiae by adaptively tracing the gray scale ridges in the fingerprint images [56, 181] The alignment ....

....obtained using one or more of the fingerprint features. For example, an alignment can be achieved based on the orientation field of the fingerprints, the location of singular points such as the core and the delta [95] ridges [11] inexact graph matching on the minutiae graphs [5] Hough transform [131], point patterns [128] etc. The number of matched minutiae in certain tolerances is typically normalized by the total number of minutiae in the two sets to account for the falsely detected and missed minutiae during the feature extraction. One of the main di#culties in the minutiae based ....

[Article contains additional citation context not shown here]

N. Ratha, K. Karu, S. Chen, and A. K. Jain, "A Real-Time Matching System for Large fingerprint Databases," IEEE Trans. Pattern Anal. and Machine Intell., Vol. 18, No. 8, pp. 799-813, 1996.

....the lit 1 E mail: carlotta ee.ucr.edu, stari ee.ucr.edu, liang ee.ucr.edu. Research is in part funded by the TRW fund of the Cleveland foundation. In Proceedings of ICASSP 98. erature; although rather different from each other, all these methods transform fingerprint images into binary images [7, 6]. The main difficulty of this approach is due to the fact that fingerprints quality is often too low, and when binarization is applied to noisy and low contrast images often produces unsatisfactory results. Noise and contrast deficiency can produce false minutiae, which are impossible to detect ....

Ratha, N.K., Karu, K., N.K., Chen, S., and Jain, A.K., 1996, "A Real-Time Matching System for Large Fingerprint Databases", IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol. 18, No. 8, pp. 799--813.

....systems, a very high degree of parallelism is utilized by mapping the parallel computing jobs onto the array of the FPGAs. The range of applications investigated by a number of researchers in this area cover image signal processing[ID94] encryption[Nor95] pattern matching[GJL95] database search[RKCJ96, CAC 95] etc. On the other side, reconfigurability can be used for circuit emulation. In a circuit emulation system, a netlist of the circuit is mapped into the emulator. The reconfigurability of the system offers the user chances to debug the circuit before going to real silicon. There has ....

N. K. Ratha, K. Karu, Shaoyun Chen, and A. K. Jain. A real-time matching system for large fingerprint databases. IEEE Transactions on Pattern Analysis and Machine Intelligence, pages 799--813, 1996.

.... may introduce several problems in a later feature detection phase, and hence many mechanisms have been proposed to avoid them [12] 13] 14] Among these approaches, the most flexible one consists of estimating orientations by means of a structure tensor or second moment descriptor [15] 16] [17], 7] and by substituting the contextual filtering step by a special kind of anisotropic di#usion scheme [18] 15] 16] which was first proposed for fingerprint image processing in [16] These ideas have to be further developed in order to design a fully automatic and highly accurate module for ....

....scale) necessary to obtain reliable orientation estimates. These methods are based on a measure of consistency of the orientation map ALMANSA AND LINDEBERG 26 after smoothing, and the smoothing level is selected by maximizing this measure in a coarse to fine [14] 7] or fine to coarse [17] manner. In this work the integration scale is determined as a function of the local scale and ridgeness. This avoids unreliable estimates near scars and other noisy situations, which can be expected to occur in the methods in [14] 7] On the other hand, the proposed method has the disadvantage ....

N.K. Ratha, K. Karu, S. Chen, and A. Jain, "A real-time matching system for large fingerprint databases," IEEE Trans. Pattern Analysis and Machine Intell., vol. 18, no. 8, pp. 799--813, Aug. 1996.

No context found.

N. Ratha, K. Karu, S. Chen and A. K. Jain, A Real-time Matching System for Large Fingerprint Database, IEEE Trans. on Pattern Anal. Machine Intell., Vol. 18, No. 8, pp. 799-813, 1996.

No context found.

N. Ratha, K. Karu, S. Chen and A. K. Jain, A Real-time Matching System for Large Fingerprint Database, IEEE Trans. on Pattern Anal. Machine Intell., Vol. 18, No. 8, pp. 799-813, 1996.

No context found.

N. K. Ratha, K. Karu, S. Chen, and A. K. Jain, \Real-time Matching System for Large Fingerprint Database", IEEE Trans. on Patt. Anal. and Machine Intell., Vol. 18, No. 8, pp. 799-813, 1996.

No context found.

N. Ratha, K. Karu, S. Chen and A . K. Jain, "A real-time matching system for large fingerprint databases," IEEE Trans. pattern Analysis and Machine Intell., Vol. 18, No. 8, pp. 799-813, 1996.

No context found.

# N. Ratha, K. Karu, S. Chen, and A.K. Jain, "A Real-Time Matching System for Large Fingerprint Databases," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 18, no. 8, pp. 799-813, Aug. 1996.

....veri cation systems which can be broadly classi ed into two categories: i) minutiae based, and (ii) lter based. The three minutiae based and one lter based algorithms are summarized in this section. 4. 1 Matcher Hough The ngerprint matching problem can be regarded as template matching [10]: given two sets of minutia features, compute their matching score. The two main steps of the algorithm are: 1) Compute the transformation parameters x , y , and s between the two images, where x and y are translations along x and y directions, respectively, is the rotation angle, and ....

....within a bounding box; 3) Repeat the previous two steps for the set of discretized allowed transformations. The transformation that results in the highest matching score is believed to be the correct one. The nal matching score is scaled between 0 and 99. Details of the algorithm can be found in [10]. 4.2 Matcher String Each set of extracted minutia features is rst converted into polar coordinates with respect to an anchor point. The two dimensional (2D) minutia features are, therefore, reduced to a one dimensional (1D) string by concatenating points in a increasing order of radial angel ....

N. K. Ratha, K. Karu, S. Chen, and A. K. Jain, \Real-time Matching System for Large Fingerprint Database", IEEE Trans. on Patt. Anal. and Machine Intell., Vol. 18, No. 8, pp. 799-813, 1996.

....of the block in several ways. For instance, one could determine the block gradient orientation by averaging the pixel gradient orientations [46] An alternative method of determining block gradient orientation may rely on a voting scheme involving pixel gradient orientations [47] Another method [48] uses a least square optimization scheme involving the pixel gradient orientations. 27 The rationale for determining a single orientation for each block of w Theta w pixels (rather than for each pixel) is computational efficiency. Consequently, in regions of a fingerprint with smoothly flowing ....

....These straightforward approaches generally do not work well for noisy and low contrast portions of the image. A more reliable property of the ridges in a fingerprint image is that the gray level values on ridges attain their local minima 9 along a direction normal to the local ridge orientation [48, 9]. Pixels can be identified to be ridge pixels based on this property. Given the local ridge orientation at a pixel (i; j) in the foreground portion of the image, a simple test can be devised to determine whether the gray level values in the fingerprint image attain a local minima at (i; j) along a ....

[Article contains additional citation context not shown here]

S. Chen N. Ratha, K. Karu and A. K. Jain. A real-time matching system for large fingerprint database. IEEE Trans. Pattern Anal. and Machine Intell., 18(8):799--813, 1996.

....say, in the color of the sky. In the latter case the images themselves are not restricted, but the isomorphism between images and meanings is. An unrestricted database, of course, is composed of images that are not restricted neither in content nor in interpretation. For instance, 38] [37], 29] and [2] can be considered restricted in our sense, while [23] 12] and [15] are unrestricted. There is no place in the image to which we could ascribe the meaning or parts of it (from which particular part of a tree does its shape come from ) In an image of the sea at twilight, there ....

N.K. Ratha, K. Karu, Shaoyun Chen, and A.K. Jain. A real-time matching system for large fingerprint databases. IEEE Transactions on Pattern Analysis and Machine Intelligence, 18(8):799-- 813, 1996.

No context found.

N. Ratha, K. Karu, S. Chen and A. K. Jain, A Real-time Matching System for Large Fingerprint Database, to appear in the IEEE Trans. on PAMI, 1996.

....say, in the color of the sky. In the latter case the images themselves are not restricted, but the isomorphism between images and meanings is. An unrestricted database, of course, is composed of images that are not restricted neither in content nor in interpretation. For instance, 38] [37], 29] and [2] can be considered restricted in our sense, while [23] 12] and [15] are unrestricted. There is no place in the image to which we could ascribe the meaning or parts of it (from which particular part of a tree does its shape come from ) In an image of the sea at twilight, there ....

N.K. Ratha, K. Karu, Shaoyun Chen, and A.K. Jain. A real-time matching system for large fingerprint databases. IEEE Transactions on Pattern Analysis and Machine Intelligence, 18(8):799--813, 1996.

....a, b, and c) they tolerate both spurious minutiae as well as missing genuine minutiae. Elastic matchers in the literature accommodate a small bounded local perturbation of minutiae from their true location but can not handle large displacements of the minutiae from their true locations [59]. Figure 5 illustrates a typical situation of aligned ridge structures of mated pairs. Note that the best alignment in one part (center) of the image may result in a large amount of displacements between the corresponding minutiae in the other regions (bottom right) In addition, observe that the ....

.... an elastic matching can be achieved by placing a bounding box around each template minutiae, which specifies all the possible positions of the corresponding input minutiae with respect to the template minutiae, and restricting the corresponding minutiae in the input image to be within this box [59]. This method does not provide a satisfactory performance in practice, because local deformations may be small while the accumulated global deformations can be quite large. We 1. For each ridge d 2 R d , represent it as an one dimensional discrete signal and match it against each ridge, D 2 R ....

[Article contains additional citation context not shown here]

N. Ratha, K. Karu, S. Chen and A. K. Jain, A Real-time Matching System for Large Fingerprint Database, IEEE Trans. on PAMI, Vol. 18, No. 8, pp. 799-813, 1996.

....Abstract We describe the design and implementation of an automatic identity authentication system which uses fingerprint to establish the identity of an individual. An improved minutia extraction algorithm that is much faster and more accurate than our earlier algorithm [12] has been implemented. An alignment based elastic matching algorithm has been developed. This algorithm is capable of finding the correspondences between input minutia pattern and the stored template minutia pattern without resorting to exhaustive search and has the ability to adaptively ....

....pattern, a reliable minutia extraction algorithm is critical to the performance Figure 3. GUI of the authentication system of an automatic identity authentication system. In our system, we have implemented a minutia extraction algorithm which is an improved version of the technique proposed in [12]. Experimental results show that this algorithm performs very well in operation. The overall flowchart of our minutia extraction algorithm is depicted in Figure 4. Orientation Ridge Extraction Estimation Locator Fingerprint Thinning Minutia Extraction Orientation Field Region of Interest Thinned ....

N. Ratha, K. Karu, S. Chen and A. K. Jain, A Real-time Matching System for Large Fingerprint Database, IEEE Trans. on PAMI, Vol. 18, No. 8, pp. 799-813, 1996.

....[29] and DNA sequence matching [8] The CCMs can be programmed with varying granularity of instructions. Further the availability of a variety of communication support makes them highly suitable for all the three layers (low level, intermediate level and high level) of computer vision tasks [31, 30, 32]. 3 Splash 2 Architecture and Programming Flow Splash 2 is one of the leading FPGA based custom computing machine designed and developed by the Supercomputing Research Center [8] The Splash 2 system consists of an array of Xilinx 4010 FPGAs, improving on the design of the Splash 1 which was ....

N. K. Ratha, K. Karu, S. Chen, and A. K. Jain. A real-time matching system for large fingerprint database. IEEE Trans. on PAMI, 18(11):to appear, Nov. 1996.

No context found.

Ratha N., Karu K, Chen S. and Jain A., "A Real-Time Matching System for Large Fingerprint Databases", IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 18, no. 8, pp.799-813, 1996

No context found.

N. K. Ratha and K. Karu, "A real time matching system for large fingerprint databases," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 18, no. 8, pp. 799-813, 1996.

No context found.

Ratha,N., Karu,K., Chen,S., and Jain,A.: A Real Time Matching System for Large Fingerprint Databases. IEEE-PAMI, Vol. 18, No. 8, pp. 799-813, Aug. 1996. 10

No context found.

N. K. Ratha, K. Karu, S. Chen, and A. K. Jain, "A real-time matching system for large fingerprint databases," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 18, no. 8, pp. 799--813, 1996.

No context found.

Ratha, N., Karu, K., Chen, S., Jain, A.K.: A real-time matching system for large fingerprint database. IEEE Trans. on PAMI 18 (1996) 799--813

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC