Context-Dependent Confusions Rules for Building Error Model Using Weighted Finite State Transducers for OCR Post-Processing

Azawi, Mayce Al; Breuel, Thomas M.

doi:10.1109/das.2014.75

Cited by 7 publications

(2 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The classical technique, widely used in different fields, to find the maximum likelihood path on a finite-state machine and to perform error-correcting parsing on a regular grammar is the Viterbi Algorithm [10,11]. In [12], OCR errors are corrected using contextdependent confusion rules and a language model, both represented by means of Finite-State Transducers (WFST). The confusion rules are extracted from a training corpus by aligning the misrecognized words of the OCR output with their corresponding ground truth.…”

Section: Introductionmentioning

confidence: 99%

Composition of Constraint, Hypothesis and Error Models to improve interaction in Human–Machine Interfaces

Navarro-Cerdan¹,

Llobet²,

Arlandis³

et al. 2016

Information Fusion

View full text Add to dashboard Cite

a b s t r a c tAlthough there are many tasks where output strings are automatically generated from a set of evidence, they are not perfect and human intervention is often required to correct the result. In this paper we present a generic Symbol Input Interaction Method for Human-Machine Interfaces that combines multisource information: an input Hypothesis Model, an Error Model, a Constraint Model and a user interaction scheme.We use Weighted Finite-State Transducers (WFSTs) to represent the different sources of information available: the initial hypotheses, the possible errors, the constraints imposed by the task (interaction language) and the user input. The fusion of these models to find the most probable output string can be performed efficiently by using carefully selected transducer operations. The proposed system initially suggests an output based on the set of hypotheses, possible errors and Constraint Models. Then, if human intervention is needed, a multimodal approach, where the user input is combined with the aforementioned models, is applied to produce, with a minimum user effort, the desired output. This approach offers the practical advantages of a de-coupled model (e.g. input-system + parameterized rules + post-processor), keeping at the same time the error-recovery power of an integrated approach, where all the steps of the process are performed in the same formal machine (as in a typical HMM in speech recognition) to avoid that an error at a given step remains unrecoverable in the subsequent steps. After a presentation of the theoretical basis of the proposed multisource information system, its application to two real world problems, as an example of the possibilities of this architecture, is addressed. The experimental results obtained demonstrate that a significant user effort can be saved when using the proposed procedure. A simple demonstration, to better understand and evaluate the proposed system, is available on the web https://demos.iti.upv.es/hi/.

show abstract

Section: Introductionmentioning

confidence: 99%

Composition of Constraint, Hypothesis and Error Models to improve interaction in Human–Machine Interfaces

Navarro-Cerdan¹,

Llobet²,

Arlandis³

et al. 2016

Information Fusion

View full text Add to dashboard Cite

show abstract

“…Hassan Awadallah et al [2008] hace uso de autómatas de estados finitos (FSA) a la hora de proponer correcciones candidatas con una distancia de edición específica con las correcciones con errores ortográficos. Al Azawi and Breuel [2014] hace uso de autómatas transductores de estados finitos (WFST) para modelizar la información de confusiones contextuales de símbolos obtenidas a partir del algoritmo Levenshtein [1966] y fusionar dicha información con la información procedente de un OCR y el modelo de lenguaje de salida ambos modelizados también mediante autómatas WFST.…”

Section: Estado Del Arteunclassified

Aplicaciones de los autómatas transductores finitos con pesos (WFST) en la corrección simbólica en interfaces persona-máquina

Cerdán¹,

Ramón²

View full text Add to dashboard Cite

Transformer-Based Neural Machine Translation for Post-OCR Error Correction in Cursive Text

Yasin

Siddiqi²,

Moetesum³

et al. 2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Context-Dependent Confusions Rules for Building Error Model Using Weighted Finite State Transducers for OCR Post-Processing

Cited by 7 publications

References 9 publications

Composition of Constraint, Hypothesis and Error Models to improve interaction in Human–Machine Interfaces

Composition of Constraint, Hypothesis and Error Models to improve interaction in Human–Machine Interfaces

Aplicaciones de los autómatas transductores finitos con pesos (WFST) en la corrección simbólica en interfaces persona-máquina

Transformer-Based Neural Machine Translation for Post-OCR Error Correction in Cursive Text

Contact Info

Product

Resources

About