Page - 177 - in Document Image Processing
Image of the Page - 177 -
Text of the Page - 177 -
J. Imaging 2017,3, 62
Figure5.Holedegradation. (Left)original images. (Right)degradedimages.
3.4. Bleed-Through
WithDocCreator it ispossible toaddbleed-throughdefects. Thisalgorithmisdirectly inspired
from[49] that initiallyproposesanalgorithmforerasingthebleed-throughfromadocument image.
By justgivingan input recto image, an inputverso imageand theamountofwisheddegradation,
a physicalmodel is applied. Thismodelmimics the verso ink that seeps through the recto side.
Thismodel simulatesananisotropicdiffusionandeachpixelat time t+1 ismodiļ¬edaccordingto the
pixelsvaluesat time twiththe followingequation:
It+1i,j = I t
i,j+Ī»ā(ctNi,j Ā· NIti,j+ctSi,j Ā· SIti,j+ctEi,j Ā· EIti,j+ctWi,j Ā· WIti,j)
where I is the recto image, V is the verso image, lambda a constant value in [0;0.25] and N, S,
E, W are themnemonic subscripts forNorth, South, East, West. NIti,j and ctNi,j are deļ¬ned as:
NIti,j =Vtiā1,jā Iti,j and ctNi,j = 1
1+ (Vtiā1,jāI0i,j)2
Ļ2 . Theuser sets thenumberof iterationsand thus the
quantityofgeneratedbleed-through.
SeeFigure6 forableed-throughexample.
Figure6.Bleed-throughdefect. (Left)original image. (Right)degradedimage.
177
back to the
book Document Image Processing"
Document Image Processing
- Title
- Document Image Processing
- Authors
- Ergina Kavallieratou
- Laurence Likforman-Sulem
- Editor
- MDPI
- Location
- Basel
- Date
- 2018
- Language
- German
- License
- CC BY-NC-ND 4.0
- ISBN
- 978-3-03897-106-1
- Size
- 17.0 x 24.4 cm
- Pages
- 216
- Keywords
- document image processing, preprocessing, binarizationl, text-line segmentation, handwriting recognition, indic/arabic/asian script, OCR, Video OCR, word spotting, retrieval, document datasets, performance evaluation, document annotation tools
- Category
- Informatik