Handwritten document retrieval

Russell, Gregory; Perrone, Michael; Chee, Yi‐Min; Ziq, Airnan

doi:10.1109/iwfhr.2002.1030915

Cited by 15 publications

(8 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[2][3][4][5][6][7] Roughly, the existing methods for handwritten document retrieval can be divided into recognitionbased 2, 6 and recognition-free or word-spotting approaches 3, 4, 7, 8 (see Figure 1). …”

Section: Introductionmentioning

confidence: 99%

Combining approaches to on-line handwriting information retrieval

Saldarriaga¹,

Viard-Gaudin

Morin³

2010

Document Recognition and Retrieval XVII

View full text Add to dashboard Cite

In this work, we propose to combine two quite different approaches for retrieving handwritten documents. Our hypothesis is that different retrieval algorithms should retrieve different sets of documents for the same query. Therefore, significant improvements in retrieval performances can be expected. The first approach is based on information retrieval techniques carried out on the noisy texts obtained through handwriting recognition, while the second approach is recognition-free using a word spotting algorithm. Results shows that for texts having a word error rate (WER) lower than 23%, the performances obtained with the combined system are close to the performances obtained on clean digital texts. In addition, for poorly recognized texts (WER > 52%), an improvement of nearly 17% can be observed with respect to the best available baseline method.

show abstract

Section: Introductionmentioning

confidence: 99%

Combining approaches to on-line handwriting information retrieval

Saldarriaga¹,

Viard-Gaudin

Morin³

2010

Document Recognition and Retrieval XVII

View full text Add to dashboard Cite

show abstract

“…The primary solution to handle online handwritten data is to employ a HWR (handwriting recognizer) to convert the ink into text, and use the results to search and retrieve the documents [1,2]. However, this approach is suited only where the handwritten data is purely text and where a robust HWR is available for the language contained in the document.…”

Section: Introductionmentioning

confidence: 99%

Retrieval of online handwriting by synthesis and matching

Jawahar

Balasubramanian

Meshesha

et al. 2009

Pattern Recognition

View full text Add to dashboard Cite

Search and retrieval is gaining importance in the ink domain due to the increase in the availability of online handwritten data. However, the problem is challenging due to variations in handwriting between various writers, digitizers and writing conditions. In this paper, we propose a retrieval mechanism for online handwriting, which can handle different writing styles, specifically for Indian languages. The proposed approach provides a keyboard-based search interface that enables to search handwritten data from any platform, in addition to pen-based and example-based queries. One of the major advantages of this framework is that information retrieval techniques such as ranking relevance, detecting stopwords and controlling word forms are extended to work with search and retrieval in the ink domain. The framework also allows cross-lingual document retrieval across Indian languages.

show abstract

“…In some cases, the words are searched after the documents have been recognized and several techniques have been proposed to make WS more robust with respect to recognition errors: Kwok et al (2000) convert each handwritten word into a stack of scores related to the dictionary entries. Russell et al (2002) use the N best recognizer outputs to expand the transcriptions and associate a probabilistic score to each term. In other cases, the recognition is avoided and WS is performed by matching query word images with the word images extracted from the documents (Jain and Namboodiri, 2003;Kolcz et al, 2000;Rath and Manmatha, 2003;Tomai et al, 2002;Uchiashi and Wilcox, 1999).…”

Section: Introductionmentioning

confidence: 99%

Application of information retrieval techniques to single writer documents

Vinciarelli

2005

Pattern Recognition Letters

View full text Add to dashboard Cite

Abstract. This work shows Information Retrieval experiments performed over handwritten documents produced by a single writer. The same retrieval task has been performed over both manual (no errors) and automatic (Word Error Rate around 45%) transcriptions of 200 handwritten texts. The results show that the performance loss due to recognition errors is acceptable and that Information Retrieval technologies can be effectively applied to handwritten data.

show abstract

Handwritten document retrieval

Cited by 15 publications

References 7 publications

Combining approaches to on-line handwriting information retrieval

Combining approaches to on-line handwriting information retrieval

Retrieval of online handwriting by synthesis and matching

Application of information retrieval techniques to single writer documents

Contact Info

Product

Resources

About