A. Ranade and A. Rosenfeld, "Point pattern matching by relaxation," Pattern Recognition, vol. 12, no. 2, pp. 269-275, 1993.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....A good quality fingerprint contains between 60 and 80 minutiae, but different fingerprints have different number of minutiae. The variable sized minutiae based representation does not easily lend itself to indexing mechanisms. Further, typical graph based [9] 11] and point pattern based [1] [12], 13] approaches to match minutiae from two fingerprints need to align the unregistered minutiae patterns of different sizes which makes them computationally expensive. Correlation based techniques [14] 15] match the global patterns of ridges and valleys to determine if the ridges align. The ....

A. Ranade and A. Rosenfeld, "Point pattern matching by relaxation," Pattern Recognit., vol. 12, no. 2, pp. 269--275, 1993.

....[7, 8, 58, 31] The time complexity of matching increases exponentially with the number of objects in the images which are compared. Various other techniques with lower time complexity, can be used to resolve such queries. Such a technique is matching based on 2 D strings [9] Techniques such as [68, 69, 70, 71, 72, 73] provide alternative solutions to reduced complexity matching. The similarity between two images (e.g. a query and a stored image) whose content is represented by 2 D strings, can be determined based either on exact or approximate matching techniques. In the first case, in order for two images ....

....is best suited in the case of partial match queries, since such kinds of queries require more than one accesses to secondary storage and large amounts of data are expected to be retrieved and processed. A number of techniques, such as cross correlation (or point pattern) matching techniques [68, 69, 71], techniques based on graph matching [8, 58, 70, 31] etc. are known to exist and can be used to retrieve images by content (see Section 2.7) The performance of the proposed 91 methodology must also be compared against such techniques. Furthermore, the retrieval of images based on characteristics ....

Sanjay Ranade and Azriel Rosenfeld. Point Pattern Matching by Relaxation. Pattern Recognition, 12:269--275, 1980.

....di#erent 14 acquisitions of the same finger have di#erent numbers of minutiae. A graph based representation [118, 155, 5] constructs a nearest neighbor graph from the minutiae patterns. The matching algorithm is based on inexact graph matching techniques. The point pattern based representation [11, 26, 96] considers the minutiae points as a two dimensional pattern of points. Correlation based techniques [61, 31] consider the gray level information in the fingerprint as features and match the global patterns of ridges and valleys to determine if the ridges align. The global representation of ....

A. Ranade and A. Rosenfeld, "Point Pattern Matching by Relaxation," Pattern Recognition, Vol. 12, No. 2, pp. 269-275, 1993.

....extraction architecture (x5.2) Iterative point matching algorithms have repeatedly demonstrated considerable robustness towards noise, local as well as global landmark variation [Kitchen, 1985] Hence, these methods are better suited for our need. We utilise the basic point matching strategy [Ranade and Rosenfeld, 1980], but incorporate feature type information [Wang et al. 1983, Ton and Jain, 1989] and two way matching [Ton and Jain, 1989] The adapted, heuristic algorithm is presented in the remaining part of this section. Let A = fA 1 ; Delta Delta Delta ; Am g and B = fB 1 ; Delta Delta Delta ; Bn g ....

....1 = ff( w j u w r v kw j ukkw r v k Gamma 1) 1 (8.3) The norm of a vector is denoted by k Delta k and 0 ff 1 represents the feature weighting factor. Local feature information is incorporated in the matching process if ff 0, otherwise, matching is reduced to the original scheme [Ranade and Rosenfeld, 1980]. Only if landmark types at position k and j are identical, does the landmark type correspondence equal unity, i.e. k = 1; ff 0. As the variation between types is increased, k decreases and local support of feature types is withdrawn. The support g that pair (A i ; B h ) receives from (A j ; B r ....

[Article contains additional citation context not shown here]

Ranade, S. and Rosenfeld, A. (1980). Point pattern matching by relaxation. Pattern Recognition, 12:269--275.

....in the next subsection. 1. 1 Related work Matching has been approached in a number of ways, including tree pruning [55] the generalized Hough transform [8] or pose clustering [51] geometric hashing [59] the alignment method [27] statistics [40] deformable templates [50] relaxation labeling [44], Fourier descriptors [35] wavelet transform [31] curvature scale space [36] and neural networks [21] The following subsections treat a few methods in more detail. They are based on shape representations that depend on the global shape. Therefore, they are not robust against occlusion, and do ....

S. Ranade and A. Rosenfeld. Point pattern matching by relaxation. Pattern Recognition, 12:269--275, 1980.

....and within it individual point correspondences. This is a point pattern matching problem. The justification of most point pattern matching algorithms includes assumptions about the mapping that relates the two sets of points. Solutions have been proposed to problems involving translation, 21] [22], translation combined with rotation, 23] general affine transformation, 24] and affine transformation combined with random noise, 25] The problem of matching point patterns extracted from mammograms is more difficult because the transformation relating two sets of points cannot be ....

....has shown that the signature formation is robust and that the distributions do not show significant changes with different parameter selection. 4. 3 Establishing point correspondence The algorithm for establishing point correspondence uses an accumulator matrix (based on work described in [21] [22], and [25] and signatures to tally votes for a particular match. Given a set of m potential control points in the older mammogram, M o , and a set of n potential control points in the newer mammogram, M n , the accumulator matrix is an m Theta n array, where an entry e(p; q) corresponds to ....

S. Ranade and A. Rosenfeld, "Point Pattern Matching by Relaxation," Pattern Recognition, vol. 12, pp. 269-275, 1980.

....pattern matching has been an important problem in machine vision for some time. A number of di erent general strategies have been used to approach the problem. Four such strategies, along with their advantages and disadvantages are outlined below. The Cluster Approach. The clustering approach ([28, 29, 31, 34, 36, 38]) involves associating con dence values with locations in a discretized con guration space of possible orientations of the pattern with respect to the background and then choosing the match that is associated with the largest cluster or peak in the con dence values in the con guration space. These ....

S. Ranade and A. Rosenfeld. Point pattern matching by relaxation. Pattern Recognition, 12:269-275, 1980.

....88] Instead of mapping each point individually, these techniques map the set of points in one image onto the corresponding set in the second image. Consequently the matching solution uses the information from all points and their 26 relative locations. The relaxation technique described by [Ranade 80], can be used to register images under translation. In this case, the point matching and the determination of the best spatial transformation are accomplished simultaneously. Each possible match of points defines a displacement which is given a rating according to how closely other pairs would ....

.... Examples Decision Sequencing Improved efficiency for similarity optimization for rigid transformations [Barnea 72] Relaxation Labeling Practical approach to find global transformations when local distortions are present, exploits spatial relations between features [Hummel 83] Price 85] [Ranade 80], Shapiro 90] Dynamic Programming Good efficiency for finding local transformations when an intrinsic ordering for matching is present [Guilloux 86] Maitre 87] Milios 89] Ohta 87] Generalized Hough Transform For shape matching of rigidly displaced contours by mapping edge space into ....

[Article contains additional citation context not shown here]

S. Ranade and A. Rosenfeld, "Point Pattern Matching by Relaxation, " Pattern Recognition 12, pp269-275.

....to see wether they are the same or not. The comparison is not an exact process. Two images captured at different times or by different people are generally not exactly the same. To judge wether two images are the same or not is the task of image matching. The literature on this subject is rich [22, 23, 30, 31, 32, 33]. To match two images some common features are needed. Low level image features like the image centroids, principle axes, edges and corners are often used. If two images are the 15 CHAPTER 3. BACKGROUND 16 same, these point sets must be the same. These points can be viewed as invariants under 2 D ....

....may differ because of perspective distortion. It is hard to determine the proper perspective transformation, however, using the landmarks that can be found in both images is a good approach. 4.1. 1 Relaxation Method Relaxation gets its name from the iterative numerical methods which it resembles[31]. It is a bottom up search strategy that involves local ratings (of similarity) which depend on the ratings of their neighbours. These ratings are updated iteratively until the ratings converge or until a sufficiently good match is found. Ranade and Rosenfeld [31] used this technique to register ....

[Article contains additional citation context not shown here]

S. Ranade and A. Rosenfeld, "Point pattern matching by relaxation", Pattern Recognition, V.12, pp. 269-275, 1980.

....reference landmark set and the landmarks in the sensed image(s) is to be established. Common diculty here is that there could be some landmarks having no corresponding counterparts in the other frame(s) Matching methods can be based on parameter clustering [30] graph matching [14] relaxation [26], 33] correlation like techniques [25] 2] 4] taking into account the landmark neighborhood or hybrid invariant approaches [34] 10] 35] 8] 6] This work has been supported by the grant No. 624178 3 of the Grant Agency of the Czech Ministry of Health. The authors thank to Dr. Igor ....

S. Ranade and A. Rosenfeld. Point pattern matching by relaxation. Pattern Recognition, 12:269-275, 1980.

.... [VV99] 2 Approaches Matching has been approached in a number of ways, including tree pruning [Ume93] the generalized Hough transform or pose clustering [Bal81] Sto87] geometric hashing [WR97] the alignment method [HU87] statistics [Sma96] deformable templates [SP95] relaxation labeling [RR80] Fourier descriptors [Lon98] wavelet transform [JFS95] curvature scale space [MAK96] and neural networks [Gol95] Without being complete, in the following subsections we will describe and group a number of these methods together. 2.1 Global image transforms There is a number of techniques ....

S. Ranade and A. Rosenfeld. Point pattern matching by relaxation. Pattern Recognition, 12:269-275, 1980.

....differences (one way versus two way constraints) 17] have been highlighted as well. We now turn to relaxation labeling. 3 Relaxation labeling Relaxation labeling as the name suggests began as a method for solving labeling problems [11] While the framework has been extended to many applications [38, 39, 40, 41, 16, 15] the basic feature of the framework remains: Start with a set of nodes i (in feature or image space) and a set of labels . Derive a set of compatibility coefficients (as in Section 2.3.1) r for each problem of interest and then apply the basic recipe of relaxation labeling for updating the ....

S. Ranade and A. Rosenfeld. Point pattern matching by relaxation. Pattern Recognition, 12:269--275, 1980.

....Generally, an automatic fingerprint verification identification is achieved with point pattern matching (minutiae matching) instead of a pixel wise matching or a ridge pattern matching of fingerprint images. A number of point pattern matching algorithms have been proposed in the literature [23, 1, 21, 16]. Because a general point matching problem is essentially (a) input image (b) orientation field (c) fingerprint region (d) ridge map (e) thinned ridge map (f) extracted minutiae Figure 9: Results of our minutia extraction algorithm on a fingerprint image (512 Theta 512) captured with an inkless ....

....and their orientations superimposed on the input image. intractable, features associated with each point and their spatial properties such as the relative distances between points are often used in these algorithms to reduce the exponential number of search paths. The relaxation approach [16] iteratively adjusts the confidence level of each corresponding pair based on its consistency with other pairs until a certain criterion is satisfied. Although a number of modified versions of this algorithm have been proposed to reduce the matching complexity [23] these algorithms are inherently ....

A. Ranade and A Rosenfeld, Point Pattern Matching by Relaxation, Pattern Recognition, Vol. 12, No. 2, pp. 269-275, 1993.

....feasible heuristics have been employed. For example, the assumption is made that features move in a way similar to features in their neighborhoods. The matching process is then solved locally by means of iterative relaxation algorithms ( Barnard and Thompson, 1980; Fang and Huang, 1984; Ranade and Rosenfeld, 1980] The correspondence problem is ill posed by its nature. One of the assumptions underlying any correspondence technique is that features in the image plane correspond to moving features in the scene. In the general case, however no operator can be constructed that solves this problem. ....

S. Ranade and A. Rosenfeld. Point pattern matching by relaxation. Pattern Recognition, 12:269--275, 1980.

....30] Current computational models for human stereopsis are based on the matching of corresponding luminance edges in the pair of stereo images. Edge points and edge segments are the most popular matching tokens. Low level tokens such as edge points have been used in early work in stereo vision [31, 32, 33, 34, 35, 36, 37]. Feature points based on gray level, intensity gradient, disparity, etc. are extracted and later used as attributes for point based matching. Marr and Poggio [28] used lters in di erent orientations to extract the zero crossing points and recorded their contrast sign and orientation as ....

....matches is motivated by the existence of cooperative processes in the biological vision systems postulated by Julesz [27] Marr and Poggio [31] and others. Figure 6: The basic structure of the relaxation labeling algorithm A general matching score updating is formulated by Ranade and Rosenfeld [33]: S (r 1) N i l ; N j r ) 1 n N l X N m l 2K N l X m=1 h N l max n=1 C(N i l ; N j r ; N m l ; N n r ) S (r) N m l ; N n r ) i where C(N i l ; N j r ; N m l ; N n r ) 1 if j = n, and 0 otherwise. In Ranade and Rosenfeld s relaxation technique [33] the initial ....

[Article contains additional citation context not shown here]

S. Ranade and A. Rosenfeld, \Point pattern matching by relaxation," Pattern Recognition, vol. 12, pp. 267-275, 1980.

....with an inkless scanner. matching of fingerprint images. Because a general point matching problem is essentially intractable, features associated with each point and their relative positions are widely used in the point pattern matching algorithms to reduce the exponential number of search paths [3, 1, 6, 9]. However, these algorithms are inherently slow and are unsuitable for an on line fingerprint verification system. In our system, an alignment based matching algorithm is implemented, which decomposes the minutia matching into two stages: i) alignment stage, where transformations such as ....

A. Ranade and A Rosenfeld, Point Pattern Matching by Relaxation, Pattern Recognition, Vol. 12, No. 2, pp. 269-275, 1993.

....algorithm is, instead, based on point pattern matching (minutiae matching) The reason for this choice is our need to design a robust, simple, and fast verification algorithm and to keep a small template size. A number of point pattern matching algorithms have been proposed in the literature [69, 1, 66, 55]. A general point matching problem is essentially intractable. Features associated with points and their spatial properties such as the relative distances between points are widely used in these algorithms to reduce the exponential number of search paths. The relaxation approach to point pattern ....

....point matching problem is essentially intractable. Features associated with points and their spatial properties such as the relative distances between points are widely used in these algorithms to reduce the exponential number of search paths. The relaxation approach to point pattern matching [55] iteratively adjusts the confidence level of each corresponding pair based on its consistency with other pairs until a certain criterion is satisfied. Although a number of modified versions of this algorithm have been proposed to reduce the matching complexity [69] these algorithms are inherently ....

A. Ranade and A. Rosenfeld, Point Pattern Matching by Relaxation, Pattern Recognition, Vol. 12, No. 2, pp. 269-275, 1993.

....techniques iteratively assign values to mutually constrained objects in such a way as to ensure that the values remain consistent. A solution is found when these values converge. In PPM the objects are point or primitive mappings, and the constraints are the match criteria. Ranade et al. [10] propose relaxation for a translation invariant technique that is more tolerant of global distortions. Each point mapping, p i ; q j ) is assigned a merit score according to how well all other points (p h ; q k ) match when p i is mapped onto q j . Subsequent iterations use the merit scores ....

.... Deletion tolerant matching Rotation, translation O(n 2 log n) Minimal spanning tree Lavine et al. 8] Deletion tolerant matching Rotation, translation O(n 2 log n) Minimal cost search Wong et al. 9] Deletion tolerant matching Rotation, translation O(n 2 log n) Relaxation Ranade et al. [10] Deletion tolerant matching Translation O(n 4 ) Fuzzy relaxation Ogawa [11] Deletion tolerant matching Rotation, scale, translation Canonical forms Hong et al. 12] Matching Rotation, scale, translation, stretching O(n) Centroid bounding Griffin et al. 13] Matching Rotation, scale, ....

S. Ranade and A. Rosenfeld. Point pattern matching by relaxation. Pattern Recognition, 12(4):269--75, 1980.

No context found.

A. Ranade and A. Rosenfeld, "Point pattern matching by relaxation," Pattern Recognition, vol. 12, no. 2, pp. 269-275, 1993.

No context found.

S. Ranade and A. Posenfeld. Point pattern matching by relaxation. Pattern Recognition, 12:269--275, 1980.

No context found.

S. Ranade and A Rosenfeld, "Point pattern matching by relaxation", Pattern recognition, 12:269275, 1980.

No context found.

A. Ranade and A. Rosenfeld, \Point Pattern Matching by Relaxation," Pattern Recognition, Vol. 12, No. 2, pp. 269-275, 1993.

No context found.

Ranade, A., Rosenfeld, A.: Point pattern matching by relaxation. Pattern Recognition 12 (1993) 269--275

No context found.

S. Ranade and A. Rosenfeld. Point pattern matching by relaxation. Pattern Recognition, 12:269--275, 1980.

No context found.

Intell., 7(5):617--623. Ranade, S. and Rosenfeld, A. (1980). Point pattern matching by relaxation. Pattern Recognition, 12:269--275.

First 50 documents

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