Morphology-based text line extraction

Wu, Jui-Chen; Hsieh, Jun-Wei; Chen, Yung‐Sheng

doi:10.1007/s00138-007-0092-0

Cited by 34 publications

(17 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The precision and recall rates of our proposed classifier is computed and compared these to other methods (different datasets and data sizes). The performance of Wu's method [2], tested with 84 test images, contained 367 text regions (minimum three characters per region) of the ICDAR2003 showed the recall and precision rates were 76.29% and 78.87% respectively. While, Alvess method [1] showed a better result for precision-recall 97% and 88% respectively with similar datasets.…”

Section: Evaluation Criteriamentioning

confidence: 99%

“…Therefore, text and non-text classification in a scene image is currently particularly challenging and needs further studies. Foreground (FG) and background (BG) classification results in the literature are still not optimal [1,2] due to a diversity of the object in a natural scene (Figures 1). Hence, a new attempt that considers FG and BG as object classes would be useful for the classification.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Explicit foreground and background modeling in the classification of text blocks in scene images

Sriman

Schomaker

2015

2015 3rd IAPR Asian Conference on Pattern Recognition (ACPR)

View full text Add to dashboard Cite

Achieving high accuracy for classifying foreground and background is an interesting challenge in the field of scene image analysis because of the wide range of illumination, complex background, and scale changes. Classifying foreground and background using bag-of-feature model gives a good result. However, the performance of the classifier depends on designed features. Therefore, this paper presents an alternative classification method based on three categories of object-attributes features namely object description, color distribution and gradient strength. Each feature is computed to a classifier model. The robustness of the method has been tested on the ICDAR2015 dataset. The experimental results show that the performance of the proposed method performs competitively against the results of existing methods in term of precision and recall.

show abstract

Section: Evaluation Criteriamentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Explicit foreground and background modeling in the classification of text blocks in scene images

Sriman

Schomaker

2015

2015 3rd IAPR Asian Conference on Pattern Recognition (ACPR)

View full text Add to dashboard Cite

show abstract

“…A recent paper by Wu et al [14] uses a proposed morphology-based text line extraction method to detect text regions in images. To handle with skewed handwritten text lines, the authors apply a moment-based method to estimate the line orientations.…”

Section: Text Line Segmentation Techniquesmentioning

confidence: 99%

Text Line Segmentation in Images of Handwritten Historical Documents

Sánchez

Suárez

Mello

et al. 2008

2008 First Workshops on Image Processing Theory, Tools and Applications

View full text Add to dashboard Cite

This paper describes an original method to segment handwritten text lines from historical document images. After an initial preprocessing, we compute a black/white transition map to achieve a rough detection of the line regions in the image. Using this map, the corresponding line axes are extracted through a skeletonization algorithm and the conflicts between adjacent cutting lines are solved by some heuristics. Our approach was tested on a set of handwritten digitized documents (from the PROHIST Project database) from the end of the 19th century onwards. The proposed method worked well even with difficult images and it achieved an 82.18% of correct segmented lines for our database. The results of comparing our method with other recent proposal for automatic line extraction on the same test images offered more than a 38% of correct segmentation improvement.

show abstract

“…filtrar, caso contrário, (10.1) onde H 1 , H 2 , ..., H 7 são tradicionais heurísticas empregadas em problemas de localização de texto (Alves e Hashimoto, 2010;Neumann e Matas, 2011;Retornaz e Marcotegui, 2007;Wu et al, 2008), isto é,…”

Section: Algoritmo Para Construção Dos úLtimos Levelingsunclassified

“…Em seguida, os vértices são agrupados em regiões de texto da seguinte forma: considere {R 1 , R 2 , ..., R n } o conjunto dos retângulos envolventes sobre os vértices (com resíduos não nulos) que deu origem a R Ω . Então, dois retângulos R i e R j pertencem a mesma região de texto, se seus posicionamentos, alturas e alinhamentos são similares (Alves e Hashimoto, 2010;Wu et al, 2008), ou seja:…”

Section: Algoritmo Para Construção Dos úLtimos Levelingsunclassified

Últimos levelings: conceitos, propriedades, algoritmos e aplicações em processamento e análise de imagens

Alves¹

View full text Add to dashboard Cite

Morphology-based text line extraction

Cited by 34 publications

References 13 publications

Explicit foreground and background modeling in the classification of text blocks in scene images

Explicit foreground and background modeling in the classification of text blocks in scene images

Text Line Segmentation in Images of Handwritten Historical Documents

Últimos levelings: conceitos, propriedades, algoritmos e aplicações em processamento e análise de imagens

Contact Info

Product

Resources

About