Handwriting based writer recognition using implicit shape codebook

Bennour, Akram; Djeddi, Chawki; Gattal, Abdeljalil; Siddiqi, Imran; Mekhaznia, Tahar

doi:10.1016/j.forsciint.2019.05.014

Cited by 29 publications

(13 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Writer recognition systems try to assign test samples (e.g., a page of handwriting) to a particular writer given an existing database. Many methods use codebook approaches [8,11,22,7] to catalogue characteristic patterns such as graphemes, stroke junctions, and key-points from offline handwriting images and compare them to test samples. Zhang et al [45] extend this idea to online handwriting, and Adak et al study idiosyncratic character style per person and extract characteristic patches to identify the writer [1].…”

Section: Methodsmentioning

confidence: 99%

Generating Handwriting via Decoupled Style Descriptors

Kotani

Tellex

Tompkin

2020

Preprint

View full text Add to dashboard Cite

Representing a space of handwriting stroke styles includes the challenge of representing both the style of each character and the overall style of the human writer. Existing VRNN approaches to representing handwriting often do not distinguish between these different style components, which can reduce model capability. Instead, we introduce the Decoupled Style Descriptor (DSD) model for handwriting, which factors both character-and writer-level styles and allows our model to represent an overall greater space of styles. This approach also increases flexibility: given a few examples, we can generate handwriting in new writer styles, and also now generate handwriting of new characters across writer styles. In experiments, our generated results were preferred over a state of the art baseline method 88% of the time, and in a writer identification task on 20 held-out writers, our DSDs achieved 89.38% accuracy from a single sample word. Overall, DSDs allows us to improve both the quality and flexibility over existing handwriting stroke generation approaches.

show abstract

Section: Methodsmentioning

confidence: 99%

Generating Handwriting via Decoupled Style Descriptors

Kotani

Tellex

Tompkin

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…These include run-length features (Djeddi et al, 2013b), edge-direction distribution (Bulacu & Schomaker, 2007), edge-hinge distribution (Bulacu & Schomaker, 2007), polygon based features (Siddiqi & Vincent, 2010), chain code based global and local features (Siddiqi & Vincent, 2010) and AR-coefficients (Garain & Paquet, 2009). Furthermore, local handwriting attributes (Halder et al, 2018), implicit shape codebook (Bennour et al, 2019), delta-hinge features (He & Schomaker, 2014) and COLD features (He & Schomaker, 2017) are also investigated. A summary of these features is presented in Table 4 while the realized identification and equal error rates on the three databases are summarized in Table 5.…”

Section: Comparison With State-of-the-art Featuresmentioning

confidence: 99%

“…In another similar study, Jain and Doermann (2011) introduce a codebook of K‐adjacent segments (KAS) to identify the writer of a given sample. A recent study (Bennour et al, 2019) extends the same idea to an implicit shape codebook where patches around key points in handwriting are extracted to produce the codebook.…”

Section: Introductionmentioning

confidence: 99%

“…In another similar study, Jain and Doermann (2011) introduce a codebook of K-adjacent segments (KAS) to identify the writer of a given sample. A recent study (Bennour et al, 2019) extends the same idea to an implicit shape codebook where patches around key points in handwriting are extracted to produce the codebook. This research is aimed at studying the influence of codebook patterns on characterizing the writer of a handwritten sample extending the ideas presented in our previous work (Djeddi et al, 2013a).…”

mentioning

confidence: 99%

See 1 more Smart Citation

Influence of codebook patterns on writer recognition: An experimental study

et al. 2021

Self Cite

View full text Add to dashboard Cite

Codebook‐based writer characterization is an effective technique that has been investigated in a number of recent studies on identification and verification of writers. These methods divide a set of writing samples into small units (fragments or graphemes) and cluster these patterns to produce a codebook. Writer of a handwritten sample is then characterized by the probability (distribution) of producing the codebook patterns. In most cases, a small subset of the database under study is employed to produce the codebook while the rest of the database is used in evaluations. This work aims to validate the hypothesis that the codebook simply serves as a representation space to compare different writings and, in most cases, the patterns in the codebook do not significantly influence the identification and verification performance. The hypothesis is validated by generating a number of codebooks using Greek, Arabic and Chinese handwritten samples. Moreover, codebooks using fragments of handwritten music scores, printed text and synthetic data are also investigated. Evaluations on three well‐known handwriting databases (CVL, BFL and IAM) validate the idea that, in general, the codebook patterns do not have a significant impact on characterizing writer from handwriting.

show abstract

“…Another approach for writer identification is based on making a codebook of shapes which people usually used in their handwritings. Then, by computing the histogram of codebook members in the handwritings, appropriate feature vectors [10, 11] are extracted. In a previous work by the authors of this paper, the codebook method was used with a new algorithm for extraction of smaller parts [12, 13].…”

Section: Related Workmentioning

confidence: 99%

Writer identification with n ‐tuple direction feature from contour

et al. 2020

View full text Add to dashboard Cite

This study introduces an effective solution for text‐independent writer identification by generalising contour‐hinge feature, which is called n‐tuple direction feature. For extracting n‐tuple direction feature, the authors first obtain all contours from connected components, then n + 1 points are considered on the contour with a certain distance apart, and next, the directions of the fragments connecting two successive points are computed. The n + 1 points move on the contour and the n‐dimensional histogram of directions is computed. The proposed method is evaluated on large Farsi and English databases. A correct writer identification rate of 92.2% for English handwritings from 900 persons and 97.7% for Farsi handwritings from 600 persons are achieved. Comparison between the proposed method and other studies shows the promising performance and superiority of the proposed method.

show abstract

Handwriting based writer recognition using implicit shape codebook

Cited by 29 publications

References 28 publications

Generating Handwriting via Decoupled Style Descriptors

Generating Handwriting via Decoupled Style Descriptors

Influence of codebook patterns on writer recognition: An experimental study

Writer identification with n ‐tuple direction feature from contour

Contact Info

Product

Resources

About