scispace - formally typeset
Search or ask a question

Showing papers on "Signature recognition published in 1999"


Patent
10 Sep 1999
TL;DR: A method and apparatus for biometric recognition of an individual living organism using electric and/or magnetic or acoustic energy was described in this paper. But this method was not suitable for the detection of a single living organism.
Abstract: A method and apparatus for biometric recognition (10) of an individual living organism using electric and/or magnetic or acoustic energy. A biometric recognition system for measuring at least one bio-electrical and/or biomagnetic property. A method and apparatus for recognition of an individual living organism's identity. A method and apparatus for identifying electric and/or magnetic properties of an individual living organism. A method and apparatus for diagnosing a bone. A method and apparatus for sensing an induced current (12) in an individual living organism. A method for using a computer (22). A method for secure communication between an individual at a first location and a second location. A method and apparatus for sensing the electric and/or magnetic properties of an individual living organism using acoustic energy.

381 citations


Proceedings ArticleDOI
01 Sep 1999
TL;DR: A novel approach to ASL recognition that aspires to being a solution to the scalability problems, based on parallel HMMs (PaHMMs), which model the parallel processes independently and can be trained independently, and do not require consideration of the different combinations at training time.
Abstract: The major challenge that faces American Sign Language (ASL) recognition now is to develop methods that will scale well with increasing vocabulary size. Unlike in spoken languages, phonemes can occur simultaneously in ASL. The number of possible combinations of phonemes after enforcing linguistic constraints is approximately 5.5/spl times/10/sup 8/. Gesture recognition, which is less constrained than ASL recognition, suffers from the same problem. Thus, it is not feasible to train conventional hidden Markov models (HMMs) for large-scab ASL applications. Factorial HMMs and coupled HMMs are two extensions to HMMs that explicitly attempt to model several processes occuring in parallel. Unfortunately, they still require consideration of the combinations at training time. In this paper we present a novel approach to ASL recognition that aspires to being a solution to the scalability problems. It is based on parallel HMMs (PaHMMs), which model the parallel processes independently. Thus, they can also be trained independently, and do not require consideration of the different combinations at training time. We develop the recognition algorithm for PaHMMs and show that it runs in time polynomial in the number of states, and in time linear in the number of parallel processes. We run several experiments with a 22 sign vocabulary and demonstrate that PaHMMs can improve the robustness of HMM-based recognition even on a small scale. Thus, PaHMMs are a very promising general recognition scheme with applications in both gesture and ASL recognition.

266 citations


Proceedings ArticleDOI
01 Sep 1999
TL;DR: A new method for comparing planar curves and for performing matching at sub-sampling resolution is presented and the performance of the well-known Dynamic Time Warping algorithm is compared.
Abstract: The problem of establishing correspondence and measuring the similarity of a pair of planar curves arises in many applications in computer vision and pattern recognition. This paper presents a new method for comparing planar curves and for performing matching at sub-sampling resolution. The analysis of the algorithm as well as its structural properties are described. The performance of the new technique applied to the problem of signature verification is shown and compared with the performance of the well-known Dynamic Time Warping algorithm.

179 citations


Proceedings ArticleDOI
01 Jan 1999
TL;DR: Early results by these studies confirm that there is a rich potential in gait for recognition and only continued development will confirm whether its performance can match those of other biometrics.
Abstract: Gait is an emergent biometric aimed essentially to recognise people by the way they walk. Its advantages are that it is non-invasive and that it is less likely to be obscured since it appears to be difficult to camouflage walk, especially in cases of serious crime. Gait has allied subjects which lend support to the view that gait has clear potential as a biometric. Essentially, we use computer vision to find people and to derive a gait signature from a sequence of images. The majority of current approaches derive motion characteristics, which are then used for recognition. Early results by these studies confirm that there is a rich potential in gait for recognition. Only continued development will confirm whether its performance can match those of other biometrics.

144 citations


Proceedings ArticleDOI
20 Sep 1999
TL;DR: This paper uses a graph grammar approach for the structure recognition, also used in off-line recognition process, resulting in a general tree-structure of the underlying input-expression, which can be translated to any desired syntax.
Abstract: This paper presents an approach for the recognition of on-line handwritten mathematical expressions. The hidden Markov model (HMM) based system makes use of simultaneous segmentation and recognition capabilities, avoiding a crucial segmentation during pre-processing. With the segmentation and recognition results, obtained from the HMM recognizer it is possible to analyze and interpret the spatial two-dimensional arrangement of the symbols. We use a graph grammar approach for the structure recognition, also used in off-line recognition process, resulting in a general tree-structure of the underlying input-expression. The resulting constructed tree can be translated to any desired syntax (for example: Lisp, KT/sub E/X, and OpenMath).

58 citations


Proceedings ArticleDOI
Hiroshi Tanaka1, K. Nakajima, K. Ishigaki, K. Akiyama, Masaki Nakagawa 
20 Sep 1999
TL;DR: A hybrid handwritten character recognition system in which the recognition results of the offline and online recognizer are integrated to create an improved product.
Abstract: Describes a handwritten character recognition system that integrates offline recognition requiring a bitmap image and online recognition involving an input pattern as a sequence of x-y coordinates. Offline recognition performs well for painted or overwritten patterns (for which online recognition would not be suited), whereas online recognition is suitable for very deformed patterns (for which offline recognition is not suited). Because each method has different recognition capabilities, the methods complement each other when integrated together. We have implemented a hybrid handwritten character recognition system in which the recognition results of the offline and online recognizer are integrated to create an improved product. After testing several integration methods for a handwritten character database, we found that the best method increased the recognition rate from 73.8% (offline) and 84.8% (online) to 87.6% (integrated).

55 citations


Patent
28 Jan 1999
TL;DR: In this paper, the user can select one or more of a plurality of recognition constraints which temporarily modify the default recognition parameters to decode uncharacteristic and/or special data, enabling the recognition engine to utilize specific information to decode the special data.
Abstract: A data recognition system and method which allows a user to select between a “default recognition” mode and a “constrained recognition” mode via a user interface. In the default recognition mode, a recognition engine utilizes predetermined default recognition parameters to decode data (e.g., handwriting and speech). In the constrained recognition mode, the user can select one or more of a plurality of recognition constraints which temporarily modify the default recognition parameters to decode uncharacteristic and/or special data. The recognition parameters associated with the selected constraint enable the recognition engine to utilize specific information to decode the special data, thereby providing increased recognition accuracy.

47 citations


Patent
TL;DR: In this paper, a method and a device for recognition of isolated words in large vocabularies are described, wherein recognition is performed through two sequential steps using neural networks and Markov models techniques, respectively, and the results of both techniques are adequately combined so as to improve recognition accuracy.
Abstract: A method and a device for recognition of isolated words in large vocabularies are described, wherein recognition is performed through two sequential steps using neural networks and Markov models techniques, respectively, and the results of both techniques are adequately combined so as to improve recognition accuracy. The devices performing the combination also provide an evaluation of recognition reliability.

46 citations


Proceedings ArticleDOI
22 Aug 1999
TL;DR: A method of an off-line signature recognition by using the Hough transform to detect stroke lines from the signature image and the backpropagation neural network is used as a tool to evaluate the performance of the proposed method.
Abstract: This article describes a method of an off-line signature recognition by using the Hough transform to detect stroke lines from the signature image. The Hough transform is used to extract the parameterized Hough space from the signature skeleton as a unique characteristic feature of signatures. In the experiment, the backpropagation neural network is used as a tool to evaluate the performance of the proposed method. The system has been tested with 70 test signatures from different persons. The experimental results reveal a recognition rate 95.24%.

44 citations


Proceedings ArticleDOI
01 Aug 1999
TL;DR: A face recognition system based on 2-D DCT features and pseudo-2D Hidden Markov Models is presented that achieves a recognition rate of 99.5% on the Olivetti Research Laboratory (ORL) face database, much better than a previous pseudo 2D HMM approach.
Abstract: A face recognition system based on 2-D DCT features and pseudo-2D Hidden Markov Models is presented. The system achieves a recognition rate of 99.5% on the Olivetti Research Laboratory (ORL) face database. This recognition rate is much better than the recognition rate of a previous pseudo 2-D HMM approach. Only one single face out of the 200 available test faces was not correctly recognized. The superiority of our approach against the previous approach is analyzed, and the recognition rates are compared to other face recognition systems evaluated on the ORL database.

41 citations


Proceedings ArticleDOI
05 Oct 1999
TL;DR: The proposed parameters are calculated in two stages; first, the preprocessing stage which includes noise reduction and outline detection through a skeletonization or thinning algorithm; and second, a parameterization stage in which the signature is encoded following the signature line and recording the length and direction of the pencil drawing obtaining a vector that includes the signature spatio-temporal information.
Abstract: Signature recognition is a relevant area in secure applications referred to as biometric identification. The image of the signature to be recognized (in off-line systems) can be considered as a spatio-temporal signal due to the shapely geometric and sequential character of the pencil drawing. The recognition and classification methods known to us are based on the extraction of geometric parameters and their classification by either a linear or nonlinear classifier. This procedure neglects the temporal information of the signature. In order to alleviate this, this paper proposes to use signature parameters with spatio-temporal information and its classification by a classifier capable of dealing with spatio-temporal problems as hidden Markov models (HMM). The proposed parameters are calculated in two stages; first, the preprocessing stage which includes noise reduction and outline detection through a skeletonization or thinning algorithm; and second, a parameterization stage in which the signature is encoded following the signature line and recording the length and direction of the pencil drawing obtaining a vector that includes the signature spatio-temporal information. The classification of the above parameters is done by a HMM classifier working in the same way as isolated word recognition systems. To design (train and test) the HMM classifier we have built a database of 24 signatures of 60 different writers.

Book
01 Nov 1999
TL;DR: This paper presents a meta-modelling architecture suitable for pattern recognition and machine intelligence that has been developed at the university level and at the national and international level.
Abstract: 1 Centre for Pattern Recognition and Machine Intelligence Department of Computer Science, Concordia University 1455 de Maisonneuve Boulevard West Suite GM-606, Montreal, Canada H3G 1M8 2 Ecole de Technologie Superieure Laboratoire d’Imagerie, de Vision et d’Intelligence Artificielle (LIVIA) 1100 Notre-Dame Ouest, Montreal, Canada H3C 1K3 3 Service de Recherche Technique de La Poste Departement Reconnaissance, Modelisation Optimisation (RMO) 10, rue de l’ile Mâbon, 44063 Nantes Cedex 02, France 4 Departamento de Informatica (Computer Science Department) Pontificia Universidade Catolica do Parana Av. Imaculada Conceicao, 1155 Prado Velho 80.215-901 Curitiba PR BRAZIL

Proceedings ArticleDOI
26 Sep 1999
TL;DR: Methods for performance improvement of gesture recognition using HMMs using KL transform to compress the input information and a recursive calculation method for the HMMs' probabilities are proposed.
Abstract: HMMs are often used for gesture recognition because of the robustness. However, the computational cost and accuracy of recognition are important for real applications such as gesture recognition, speech recognition or virtual reality. In this paper, we propose methods for performance improvement of gesture recognition using HMMs. For the computational cost, we use KL transform to compress the input information and propose a recursive calculation method for the HMMs' probabilities. For the accuracy of recognition, we use an automaton layered up on HMMs to deal with context information of gestures. We also show experimental results to make the efficiency of our methods clear.

Proceedings ArticleDOI
20 Sep 1999
TL;DR: It can be shown that for both online and offline recognition, the new hybrid approach clearly outperforms the competing traditional HMM techniques and yields superior results for the offline recognition of machine printed multifont characters.
Abstract: The paper deals with the performance evaluation of a novel hybrid approach to large vocabulary cursive handwriting recognition and contains various innovations. 1) It presents the investigation of a new hybrid approach to handwriting recognition, consisting of hidden Markov models (HMMs) and neural networks trained with a special information theory based training criterion. This approach has only been recently introduced successfully to online handwriting recognition and is now investigated for the first time for offline recognition. 2) The hybrid approach is extensively compared to traditional HMM modeling techniques and the superior performance of the new hybrid approach is demonstrated. 3) The data for the comparison has been obtained from a database containing online handwritten data which has been converted to offline data. Therefore, a multiple evaluation has been carried out, incorporating the comparison of different modeling techniques and the additional comparison of each technique for online and offline recognition, using a unique database. The results confirm that online recognition leads to better recognition results due to the dynamic information of the data, but also show that it is possible to obtain recognition rates for offline recognition that are close to the results obtained for online recognition. Furthermore, it can be shown that for both online and offline recognition, the new hybrid approach clearly outperforms the competing traditional HMM techniques. It is also shown that the new hybrid approach yields superior results for the offline recognition of machine printed multifont characters.

Journal ArticleDOI
01 Apr 1999
TL;DR: The idea of combining the network of HMMs and the dynamic programming-based search is highly relevant to online handwriting recognition and one distinguishing feature of the approach is that letter segmentation is obtained simultaneously with recognition but no extra computation is required.
Abstract: The idea of combining the network of HMMs and the dynamic programming-based search is highly relevant to online handwriting recognition. The word model of HMM network can be systematically constructed by concatenating letter and ligature HMM's while sharing common ones. Character recognition in such a network can be defined as the task of best aligning a given input sequence to the best path in the network. One distinguishing feature of the approach is that letter segmentation is obtained simultaneously with recognition but no extra computation is required.

Journal ArticleDOI
A. S. Tolba1
TL;DR: A hybrid technique, which is based on both the most discriminating eigenfeatures and the self-organizing maps (SOFMs) for signature recognition is demonstrated, which retains the power to discriminate against forgeries.

Proceedings ArticleDOI
08 Aug 1999
TL;DR: This paper proposes a recognition method based on handwriting acceleration, line-crossing points segmentation, macrostructures (isolated traces), chain coding and time-frequency analysis that represents an identification index for the signer.
Abstract: Automatic signature recognition or verification have many practical applications. In this paper we propose a recognition method based on handwriting acceleration, line-crossing points segmentation, macrostructures (isolated traces), chain coding and time-frequency analysis. The acceleration information is integrated twice to get a visual representation of the signature. Further processing generates coefficients and images whose characteristics can be used as a representation. These coefficients, along with dynamic information, are applied as inputs to a 3-layered neural network, to train it. The output patterns are selected to be a binary number that represents an identification index for the signer.

Proceedings ArticleDOI
05 Oct 1999
TL;DR: This paper has proposed a multi-template matching approach to identify the individual via few training samples to achieve a better performance if more training samples are collected.
Abstract: Signature verification is a natural and friendly approach in biometrics-based verification. As we know, the system can achieve a better performance if more training samples are collected. However, routinely signing the patterns is a boring and inconvenient process to obtain enough training samples at the initial enrolment. In this paper, we have proposed a multi-template matching approach to identify the individual via few training samples. Some experimental results were conducted to show the effectiveness of our proposed methods.

Proceedings ArticleDOI
15 Mar 1999
TL;DR: A multi-model approach seeks a combination of different types of acoustic models, thereby integrating the capabilities of each individual model for capturing discriminative information.
Abstract: Most current speech recognition systems are built upon a single type of model, e.g. an HMM or certain type of segment based model, and furthermore typically employs only one type of acoustic feature e.g. MFCCs and their variants. This entails that the system may not be robust should the modeling assumptions be violated. Recent research efforts have investigated the use of multi-scale/multi-band acoustic features for robust speech recognition. This paper described a multi-model approach as an alternative and complement to the multi-feature approaches. The multi-model approach seeks a combination of different types of acoustic models, thereby integrating the capabilities of each individual model for capturing discriminative information. An example system built upon the combination of the standard HMM technique with a segment-based modeling technique was implemented. Experiments for both isolated-word and continuous speech recognition have shown improved performances over each of the individual models considered in isolation.

Proceedings ArticleDOI
10 Jul 1999
TL;DR: A shifting algorithm which can be used in a pattern recognition system to improve the system's performance in the presence of shifted input patterns is presented.
Abstract: This paper presents a shifting algorithm which can be used in a pattern recognition system to improve the system's performance in the presence of shifted input patterns. The algorithm is outlined and simulation results are presented for some human face recognition experiments. It is shown that the shifting algorithm improves recognition performance for several different face recognition algorithms.

Patent
30 Mar 1999
TL;DR: In this paper, the problem of detecting the alteration of a general text document and a word processor document in an electronic signature and recognition device for writing a signature which is the proof of an individual in the electronic document and performing authentication is addressed.
Abstract: PROBLEM TO BE SOLVED: To not only judge whether or not alteration is performed but also detect which part of a general text document and the other so-called word processor document is altered especially in the alteration detection of an electronic document in an electronic signature and recognition device for writing a signature which is the proof of an individual in the electronic document and performing authentication. SOLUTION: For the text document, the line feed codes are retrieved, a message digest which is electronic signature data is prepared from the text document between the respective line feed codes and the message digest is ciphered and transmitted through a network along with the original text document. In the meantime, on a computer side for instance which receives the text document and the ciphered electronic signature data, the message digest is prepared from the text document first, the ciphered electronic signature data are deciphered and turned to the message digest, both data are compared, and when both data do not match, it is defined that they are altered and the text document between the corresponding line feed codes is recognized.

Patent
24 Mar 1999
TL;DR: In this paper, a method for access to a digital mobile phone comprising the steps of: entering a signature in a signature recognition means having a stored reference signature (S100); comparing the entered signature and the stored reference signatures (S200); if there is a match, confirming the access; and if there are a mismatch, denying the access.
Abstract: The present invention provides a method for getting access to a digital mobile phone comprising the steps of: entering a signature in a signature recognition means having a stored reference signature (S100); comparing the entered signature and the stored reference signature (S200); if there is a match, confirming the access; and if there is a mismatch, denying the access.

Proceedings ArticleDOI
23 Aug 1999
TL;DR: An adaptive metric learning procedure with improved generalization of missing training data for facial signature recognition for use in a smart card system and examines the margin structure and selects the minimum number of support vectors to represent mixture distributions by using the generalized portrait method.
Abstract: We present an adaptive metric learning procedure with improved generalization of missing training data for facial signature recognition for use in a smart card system. The conventional learning models suffer from degraded recognition rate due to poor estimation of the margin of a decision boundary. Our model employs an image synthesis method to represent missing patterns of unknown classes by using a mixture distribution. The margin of a decision boundary is dynamically adjusted to input patterns obtained from synthesized images with a time-varying mixing ratio. The metric parameters of mixture distributions have been derived from minimization of the negative log-likelihood probability function. The present method effectively reduces the margin of a class with an improved recognition rate from 81.3% to 100%. Furthermore, we examine the margin structure and select the minimum number of support vectors to represent mixture distributions by using the generalized portrait (GP) method.

Proceedings ArticleDOI
12 Oct 1999
TL;DR: An improved method is proposed that plural signature is able to embed and no one can recognize signature without key image and up to three images can be embedded into an original image.
Abstract: A "signature checking", one can recognize the signature pattern by piling up the two images, is one of the interesting application. The piling up is carried out by coded image or printed image. Matsui proposed information embedding technique for "signature checking" on making binary image from gray scale image by using pattern dither method. We extended Matsui's method to plural signature patterns. However, in simple method, we can recognize the signature patterns by piling up not only key image and public image but also among public images. We propose an improved method that plural signature is able to embed and no one can recognize signature without key image. In the improved method, we inverse a 25% part of pixel which is in each signature pattern, and the locations of inverted pixel are exclusively in four signature patterns. In this method, four images can be embedded into an original image. However, if any information contained in signature patterns must disappear without key image, up to three images can be embedded into an original image.

Proceedings ArticleDOI
16 Nov 1999
TL;DR: An algorithm is proposed for pen-input online signature verification, incorporating pen-position, pen-pressure and pen-inclination trajectories, which considers the writer's signature as a trajectory which evolves over time, so that it is dynamic and biometric.
Abstract: An algorithm is proposed for pen-input online signature verification, incorporating pen-position, pen-pressure and pen-inclination trajectories. The algorithm considers the writer's signature as a trajectory of pen-position, pen-pressure and pen-inclination which evolves over time, so that it is dynamic and biometric. Since the algorithm uses pen-trajectory information, it naturally needs to incorporate stroke number (number of pen-ups/pen-downs) variations as well as shape variations. The proposed scheme first generates templates from several authentic signatures of individuals. In the verification phase, the scheme computes a distance between the template and an input trajectory. Care needs to be taken in computing the distance function because: (i) the length of a pen input trajectory may be different from that of a template even if the signature is genuine; (ii) the number of strokes of a pen input trajectory may be different from that of the template, i.e., the number of pen-ups/pen-downs obtained may differ from that of the template even for an authentic signature. If the computed distance does not exceed a threshold value, the input signature is predicted to be genuine, otherwise it is predicted to be forgery. Preliminary experimental results look encouraging.


Proceedings ArticleDOI
12 Oct 1999
TL;DR: The experimental results revealed that the robust face identification system could be implemented by using the Mahalanobis-Taguchi system and the effectiveness of the proposed system was confirmed through computational experiments.
Abstract: We propose a face identification system based on the Mahalanobis-Taguchi system (MTS). The MTS is one of the pattern recognition methods frequently used in quality engineering. The MTS can perform robust pattern recognition by using various training data including noise. It is likely that this advantage is effective for implementing the robust face identification system against the fluctuation of lighting and the position of the face. We confirm the effectiveness of the proposed system through computational experiments. The experimental results revealed that the robust face identification system could be implemented by using the MTS.

01 Jan 1999
TL;DR: A new topology, called circular HMM, is proposed and tested on the handwritten character recognition problem and indicates excellent performance compared to the classical temporal and ergodic HMM models.
Abstract: This study deals with the shape recognition problem using the Hidden Markov Model (HMM). In many pattern recognition applications, selection of the size and topology of the HMM is mostly done by heuristics or using trial and error methods. It is well known that as the number of states and the non-zero state transition increases, the complexity of the HMM training and recognition algorithms increases exponentially. On the other hand, many studies indicate that increasing the size and non-zero state transition does not always yield better recognition rate. Therefore, designing the HMM topology and estimating the number of states for a specific problem is still an unsolved problem and requires initial investigation on the test data. This study addresses a specific class of recognition problems based on the boundary of shapes. The paper investigates the affect of the HMM topology on the recognition rate. A new topology, called circular HMM, is proposed and tested on the handwritten character recognition problem. The proposed topology is both ergodic and temporal. It eliminates the starting and ending states with the circular state transitions. The experiments indicate excellent performance compared to the classical temporal and ergodic HMM models.


Journal Article
TL;DR: A multi stage character classifier model is presented that combines four independent algorithms into an unique pattern recognition system that has the merit of high recognition rate.
Abstract: A multi stage character classifier model is presented. By combining four independent algorithms, we integrate on line recognition method, off line recognition method, neural network method, tradition method into an unique pattern recognition system. The results and experiments show that our system has the merit of high recognition rate.