Segmentation of heterogeneous document images : an ... - Tel
Segmentation of heterogeneous document images : an ... - Tel
Segmentation of heterogeneous document images : an ... - Tel
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
tel-00912566, version 1 - 2 Dec 2013<br />
Match counting, 33<br />
Maximal empty rect<strong>an</strong>gles, 18, 19<br />
Maximum likelihood, 68<br />
Message passing algorithm, 70<br />
Minimum sp<strong>an</strong>ning tree, 23, 32, 100,<br />
101<br />
Monte Carlo method, 68<br />
Morphological closing, 16<br />
Morphological hole-filling, 16<br />
Morphological opening, 16, 20<br />
Morphological operators, 32<br />
Multi-layer perceptron, 16<br />
Multi-resolution morphology, 16<br />
Mumford-Shah functional, 32<br />
Neuro-fuzzy approach, 21<br />
Newton’s method, 71<br />
Niblack binarization, 36<br />
Noise removal, 5<br />
Non-stationary Gabor filters, 111<br />
Normalization factor, 55<br />
Observations, 57<br />
OCRopus, 2<br />
OpenCV, 46, 116<br />
Optical character recognition, 2<br />
Orientation <strong>of</strong> auto-correlation, 17<br />
Otsu’s binarization, 36<br />
Paragraph detection, 5<br />
Part-<strong>of</strong>-speech tagging, 53, 71<br />
Partition function, 55, 71<br />
Piecewise projections, 27<br />
Pixel density features, 18<br />
Pixel hit rate, 34<br />
Polynomial spline wavelets, 20<br />
Prima Layout Evaluation, 11, 98, 105<br />
Projection based line detection, 27<br />
Projection based methods, 25<br />
QT, 116<br />
Quad-tree, 17<br />
Quasi-Newton methods, 71<br />
Radon tr<strong>an</strong>sform, 17<br />
RapidMiner, 44<br />
Run-length features, 18<br />
Run-length smearing, 18, 19<br />
Run-lengths, 59<br />
Sauvola binarization, 36<br />
Skewed text lines, 24<br />
Sl<strong>an</strong>ted text lines, 22<br />
Smoothed projection pr<strong>of</strong>ile, 87<br />
Spline wavelets, 18<br />
Steepest ascent algorithm, 71<br />
Steerable directional filter, 32<br />
Stroke level properties, 14<br />
Sum-product algorithm, 70<br />
Support vector machine, 16, 17, 45<br />
Tesseract-OCR, 2, 98<br />
Text line detection, 5, 21<br />
Text region detection, 5, 18<br />
Text/Graphics separation, 5<br />
Texture-based region detection, 18, 20<br />
Top-down approach, 18<br />
Touching text lines, 22<br />
Tr<strong>an</strong>sition probability, 92<br />
Turbo decoding, 68<br />
Variational methods, 68<br />
Vehicle’s plates registration, 21<br />
Viterbi algorithm, 29, 56, 68, 88<br />
Voronoi diagram, 18, 19<br />
Voronoi++, 19<br />
Voted perceptron, 69, 104<br />
Wavelet <strong>an</strong>alysis, 17<br />
Wavelet filters, 20<br />
Wavelet packets, 18<br />
Whitespace <strong>an</strong>alysis, 18, 19<br />
X-Y cut, 18, 27<br />
128