Page - 155 - in Document Image Processing
Image of the Page - 155 -
Text of the Page - 155 -
J. Imaging 2018,4, 39
Figure5.OutputwordimagesofModifiedlog-Gaborfilter transformonasamplehandwrittenBangla
wordimage(shownonleft-side) for5different scalesand6differentorientations (thefirst rowshows
theoutput forno=00 andfivescales, thesecondrowshowstheoutput forno=300 andfivescales
andsoon).
3.ClassifierCombination
Classifiercombinationtries to improveonthe taskofpatternrecognitionperformance through
mathematicalmodels. Theoutputsofclassifierscanberepresentedasvectorsofnumberswhere the
dimensionofvectors isequal to thenumberofclasses.Asaresult, thecombinationproblemcanbe
definedasaproblemoffindingthecombinationfunctionacceptingN-dimensional scorevectors from
Mclassifiers andoutputtingNfinal classification scores (seeFigure6),where the function tries to
minimize themisclassificationcost.
Figure6.Classifiercombinationtakesasetofsji scoreforclass ibyclassifier jandproducescombination
scoresSi foreachclass i [34].
155
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