Signature recognition

About: Signature recognition is a research topic. Over the lifetime, 2138 publications have been published within this topic receiving 37605 citations.

Development of the hidden Markov models based Lithuanian speech recognition system

Abstract: The paper presents a prototype of the speaker-independent Lithuanian isolated word recognition system. The system is based on the hidden Markov models, a powerful statistical method for modeling speech signals. The prototype system can be used for Lithuanian words recognition investigations and is a good starting point for the development of a more sophisticated recognition system. The system graphical user interface is easy to control. Visualization of the entire recognition process is useful for analyzing of the recognition results. Based on this recognizer, a system for Web browser control by voice was developed. The program, which implements control by voice commands, was integrated in the speech recognition system. The system performance was evaluated by using different sets of acoustic models and vocabularies.

Off-line signature verification based on modified dynamic time warping algorithm

Abstract: This work presents an innovative approach for off-line signature verification based on modified dynamic time warping (DTW) algorithm. Conventional DTW does not account for the relative importance regarding the phase difference between a reference point and a testing point, By using the projection profiles of signatures, the scheme employed Projection stability Point and Projection stability factor which acted as weighted DTW, in order to prevent minimum distance distortion caused by outliers. The databases of English signatures are applied to the experiments and the average error rates of 11.4% are obtained to verify the effectiveness of the proposed system.

Efficient recognition of continuously-spoken numbers

Abstract: Automatic recognition of continuously-spoken numbers (e.g., telephone or credit card digit sequences) is possible with excellent accuracy, even in applications using telephone lines and serving a large population. However, even such simple recognition tasks suffer decreased performance in adverse conditions, e.g., significant background noise or fading on portable telephone channels. If we further impose significant limitations on the computing resources for the recognition task, then robust efficient speech recognition is still a significant challenge, even for a vocabulary as simple as the digits. Since connected-digit recognition over telephone lines has very practical applications. The amount of computer resources needed for a given level of recognition accuracy is investigated. Rather than use a traditional hidden Markov model approach with cepstral analysis, which is computationally intensive and does not always work well under adverse acoustic conditions, a simpler spectral analysis is used, combined with a segmental approach. The restricted nature of the digit vocabulary allows this simpler approach. High recognition accuracy can be maintained despite a large decrease in both memory and computation.

A New Method of Signature Verification Based on Biomimetic Pattern Recognition Theory

Abstract: Aim at the difficulty and low recognition rate of signature verification, this paper introduces biomimetic pattern recognition theory and applies it to the problem. According to the features of the signature samples, the coverage in the high-dimension feature space is built, one class of samples are all covered with a super-sausage neuron chain. As the radius selection of the super-sausage neurons maybe unreasonable, unwanted area may be covered and correct recognition rate will reduce. So this paper uses the relationship of the distance between the two training samples and the average distance of all the neurons to adjust the radius of the super-sausage neuron automatically. Finally, the experiments show that compared to traditional pattern recognition method, biomimetic pattern recognition theory used in signature verification have a better recognition result and is more effective.

Comparison of function features using modified dtw for dynamic signature verification

Abstract: Many methods are introduced for estimating the similarities or differences of time signals. One of theses methods, DTW algorithm, is also a utility for other domains including classification, data mining and matching regions between two time signals. DTW algorithm minimizes points distance between two signals by contracting or expanding the time axes to find the corresponding points. In this paper, with modification of the local constraints in DTW, a powerful method is proposed for measuring the global or local similarities between two signals. In addition to increasing the accuracy of signals distance measurements and decreasing the classification error, proposed algorithm is more stable than classic DTW against variations of structure and time signal source. The proposed method for dynamic signature verification was applied to a dataset of signatures from Turkish, Chinese and English people. The results of the experiments based on Fisher, Parzen Window and Support Vectors Machine classifications, showed that equal error rate (EER) is 1.46% and 3.51% with universal threshold for random and skilled forgeries, respectively.