e-PCP: A robust skew detection method for scanned document images

Dey, Prasenjit; Noushath, S.

doi:10.1016/j.patcog.2009.06.019

Cited by 17 publications

(18 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although the accuracy of this method is not high compared to conventional methods such as [23], our method is able to handle challenging cases. To be specific, experimental results on 30 vertically flowing text in [23] show that our method achieves an average error of 0.19 • with a maximum error of 0.5 • , while the method in [23] fails for 3 inputs (See.…”

Section: Application To Skew Estimationmentioning

confidence: 97%

State Estimation in a Document Image and Its Application in Text Block Identification and Text Line Extraction

Koo

Cho

2010

Computer Vision – ECCV 2010

View full text Add to dashboard Cite

This paper proposes a new approach to the estimation of document states such as interline spacing and text line orientation, which facilitates a number of tasks in document image processing. The proposed method can be applied to spatially varying states as well as invariant ones, so that general cases including images of complex layout, cameracaptured images, and handwritten ones can also be handled. Specifically, we find CCs (Connected Components) in a document image and assign a state to each of them. Then the states of CCs are estimated using an energy minimization framework, where the cost function is designed based on frequency domain analysis and minimized via graph-cuts. Using the estimated states, we also develop a new algorithm that performs text block identification and text line extraction. Roughly speaking, we can segment an image into text blocks by cutting the distant connections among the CCs (compared to the estimated interline spacing), and we can group the CCs into text lines using a bottom-up grouping along the estimated text line orientation. Experimental results on a variety of document images show that our method is efficient and provides promising results in several document image processing tasks.

show abstract

Section: Application To Skew Estimationmentioning

confidence: 97%

State Estimation in a Document Image and Its Application in Text Block Identification and Text Line Extraction

Koo

Cho

2010

Computer Vision – ECCV 2010

View full text Add to dashboard Cite

show abstract

“…For example, PJbased methods 9,10,13,16,17 use the PJ of textblocks, and CCbased methods 6,18 exploit correlations between text lines. 12 However, at times, both these methods fail to handle complex layouts even though their assumptions are not violated (an example in which both these methods fail can be found in Ref. However, they heavily depend on text regions, and the performance degrades with an increase in the number of nontextual objects.…”

Section: Related Workmentioning

confidence: 99%

“…[1][2][3][4][5][6][7][8][9][10][11][12] However, most of them were developed for the processing of constrained inputs (usually bilevel, text abundant, and simple layout), and a more robust algorithm needs to be developed to manage the ever-increasing complexity of page layout in recent print media. [1][2][3][4][5][6][7][8][9][10][11][12] However, most of them were developed for the processing of constrained inputs (usually bilevel, text abundant, and simple layout), and a more robust algorithm needs to be developed to manage the ever-increasing complexity of page layout in recent print media.…”

Section: Introductionmentioning

confidence: 99%

Robust skew estimation using straight lines in document images

Koo

Cho

2016

J. Electron. Imaging

View full text Add to dashboard Cite

A skew-estimation method using straight lines in document images is presented. Unlike conventional approaches exploiting the properties of text, we formulate the skew-estimation problem as an estimation task using straight lines in images and focus on robust and accurate line detection. To be precise, we adopt a blockbased edge detector followed by a progressive line detector to take clues from a variety of sources such as text lines, boundaries of figures/tables, vertical/horizontal separators, and boundaries of textblocks. Extensive experiments on the datasets of skewed images and competition results reveal that the proposed method works robustly and yields accurate skew-estimation results.

show abstract

“…Dey et al [12] detail the computation-accuracy tradeoffs for f 2 in their Enhanced Piecewise Covering of Parallelograms (ePCP) method. The traditional PCP method iterates over all angles and determines the skew angle to be the angle where the white sections in the document are maximized.…”

Section: Skew Detection As a Linear Ensemblementioning

confidence: 99%