Lexicon reduction for printed Farsi subwords using pictorial and textual dictionaries

Davoudi, Homa; Kabir, Ehsanollah

doi:10.1007/s10032-014-0223-x

Cited by 6 publications

(2 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…All previous techniques compare the input word / sub-word image with the lexicon of a defined set of words/ sub-words [1][2][3][4][5][6]. In this work, the lexicon is a particular set of all Arabic character shape descriptors.…”

Section: The Proposed Lexicon Reduction Techniquementioning

confidence: 99%

“…A lexicon reduction using global shape dependent on the characterizing loci was used to supplement the local descriptors of sub-word. Davoudi et al [4] introduced an approach for the reduction of the number of the lexicon entries for printed Farsi sub-words. This method applied the holistic shape with the information of the key character to lower the lexicon size orderly presented using two methods.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Arabic characters descriptors for lexicon reduction in Arabic handwriting

Essa,

El- Daydamony,

Atwan

2018

Mansoura Journal for Computer and Information Sciences

View full text Add to dashboard Cite

This paper introduces an advanced Arabic handwriting recognition technique using lexicon reduction. The lexicon reduction technique stands on extracting the Arabic character shape descriptors. The technique implementation consists of two major stages. The first stage presents a method for extracting the shape descriptor of each character. The second stage suggests Aho-Corasik string searching algorithm for Arabic character recognition. Various stages have been evaluated on the IFN/ENIT database. The results demonstrate the efficiency of the suggested technique.

show abstract

Section: The Proposed Lexicon Reduction Techniquementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Arabic characters descriptors for lexicon reduction in Arabic handwriting

Essa,

El- Daydamony,

Atwan

2018

Mansoura Journal for Computer and Information Sciences

View full text Add to dashboard Cite

show abstract

Discrete representation learning for handwritten text recognition

Davoudi

Traviglia

2023

Neural Comput & Applic

View full text Add to dashboard Cite

Handwritten text recognition, i.e., the conversion of scanned handwritten documents into machine-readable text, is a complex exercise due to the variability and complexity of handwriting. A common approach in handwritten text recognition consists of a feature extraction step followed by a recognizer. In this paper, we propose a novel DNN architecture for handwritten text recognition that extracts discrete representation from the input text-line image. The proposed model is constructed of an encoder–decoder network with an added quantization layer which applies a dictionary of representative vectors to discretize the latent variables. The dictionary and the network parameters are trained jointly through the k-means algorithm and back propagation, respectively. The performance of the suggested model is evaluated through conducting extensive experiments on five datasets, analyzing the effect of discrete representation on handwriting recognition. The results demonstrate that the use of feature discretization improves the performance of deep handwriting text recognition models when compared to the conventional DNN models with continuous representation. Specifically, the character error rate is decreased by $$22\%$$ 22 % and $$21.1\%$$ 21.1 % on IAM and ICFHR18 datasets, respectively.

show abstract