Proceedings ArticleDOI
A comparison between continuous and discrete density hidden Markov models for cursive handwriting recognition
Gerhard Rigoll,A. Kosmala,J. Rattland,C. Neukirchen +3 more
- Vol. 2, pp 205-209
Reads0
Chats0
TLDR
The surprising result of the investigation was the fact that discrete density models led to better results than continuous models, although this is generally not the case for HMM-based speech recognition systems.Citations
More filters
Journal ArticleDOI
Markov models for offline handwriting recognition: a survey
Thomas Plötz,Gernot A. Fink +1 more
TL;DR: A comprehensive overview of the application of Markov models in the research field of offline handwriting recognition, covering both the widely used hidden Markov model and the less complex Markov-chain or n-gram models is provided.
Journal ArticleDOI
Training hidden Markov models with multiple observations-a combinatorial method
TL;DR: In this treatment, the multiple observation probability is expressed as a combination of individual observation probabilities without losing generality and it is proven that the derived training equations guarantee the maximization of the objective function.
Proceedings ArticleDOI
A systematic comparison between on-line and off-line methods for signature verification with hidden Markov models
Gerhard Rigoll,A. Kosmala +1 more
TL;DR: This paper may be the first systematic comparison of online and off-line methods for signature verification using exactly the same database, and leading to the surprising result that the difference in performance for both approaches is relatively small.
Proceedings ArticleDOI
A survey of off-line signature verification
Weiping Hou,Xiufen Ye,Kejun Wang +2 more
TL;DR: This paper presents a survey of off-line siwature verification, discussing many approaches of verification in details and proposing some problems existed in the Off-line signature verification system.
Proceedings ArticleDOI
On-line handwritten formula recognition using hidden Markov models and context dependent graph grammars
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.
References
More filters
Journal Article
Vector quantization
TL;DR: During the past few years several design algorithms have been developed for a variety of vector quantizers and the performance of these codes has been studied for speech waveforms, speech linear predictive parameter vectors, images, and several simulated random processes.
Journal ArticleDOI
Continuous speech recognition by statistical methods
TL;DR: Experimental results are presented that indicate the power of the methods and concern modeling of a speaker and of an acoustic processor, extraction of the models' statistical parameters and hypothesis search procedures and likelihood computations of linguistic decoding.
Proceedings ArticleDOI
On-line cursive handwriting recognition using speech recognition methods
TL;DR: A hidden Markov model (HMM) based continuous speech recognition system is applied to on-line cursive handwriting recognition and the handwriting database collected over the past year is described and specific implementation details of the handwriting system are discussed.
Proceedings ArticleDOI
Real-time on-line unconstrained handwriting recognition using statistical methods
TL;DR: A general recognition system for large vocabulary, writer independent, unconstrained handwritten text, that performs recognition in real-time on 486 class PC platforms without the large amounts of memory required for traditional HMM based systems.
Journal ArticleDOI
Maximum mutual information neural networks for hybrid connectionist-HMM speech recognition systems
TL;DR: This paper proposes a novel approach for a hybrid connectionist-hidden Markov model (HMM) speech recognition system based on the use of a neural network as vector quantizer and demonstrates how the new learning approach can be applied to multiple-feature hybrid speech recognition systems, using a joint information theory-based optimization procedure.