Matlab functions

The following functions have been tested on Matlab R14 sp 3 + Windows XP Home Edition 32bit sp 3

Feature extraction
Name
Description
Link
BGC81ri.m
Binary gradient contours
BGC81ri.m
COOCMatrices.m
Grey-level co-occurrence matrices
COOCMatrices.m
Gabor.m
Gabor features
Gabor.m
GLCMFeatures.m
Grey-level co-occurrence features
GLCMFeatures.m
Granulometry.m
Granulometry
GLCMFeatures.m
GraycoProps.m
Modified version of Matlab's graycoprops GraycoProps.m
ILBP81ri.m
Improved local binary patterns
ILBP81ri.m
LBP81ri.m
Local binary patterns
LBP81ri.m
lutRotInv.m
Rotation invariant patterns from BGC, ILBP and LBP
lutRotInv.m
TrasformFeatureVectorFFT.m
DFT normalization for rotation invariant Gabor features
TrasformFeatureVectorFFT.m
Variogram.m
Variogram
Variogram.m


Image thresholding
Name
Description
Link
ComputeThreshold.m
Top-level function to invoke image thresholding
ComputeThreshold.m
CumMeanVar.m
Routines for histogram analysis
CumMeanVar.m
EvaluateThresholdingAccuracy.m
Thresholding accuracy
EvaluateThresholdingAccuracy.m
Isoentropic.m
Threshold through isentropic partition
Isoentropic.m
Kapur.m
Threshold through Kapur's method
Kapur.m
KittlerIllingworth.m
Threshold through Kittler & Illingworth's method KittlerIllingworth.m
MisclassificationError.m
Misclassification error
MisclassificationError.m
Otsu.m
Threshold through Otsu's method Otsu.m
Yen.m
Threshold through Yen's method Yen.m


General routines
Name
Description
Link
ComputeFeatures.m
Compute image features
DoClassification.m
DoClassification.m
Supervised classification
DoClassification.m
EvaluateCrossAccuracy.m
Classification accuracy
EvaluateAccuracy.m
ConvertToToroidalCoordinate.m
Circular (toroidal) image scanning
ConvertToToroidalCoordinate.m
GetFileNames.m Names of the files contained in a directory
GetFileNames.m
SCRIPT.m
Sample script
SCRIPT.m

Dependencies

The following additional Matlab toolboxes need to be downloaded and installed:

  • PRTools - A Matlab toolbox for pattern recognition;
  • Simplegabor - Multiresolution Gabor Feature Toolbox;
  • STPRTOOL - Statistical Pattern Recognition Toolbox


Instructions

  1. Create a new directory (e.g.: ./mydir);
  2. Save the Matlab functions listed above in ./mydir;
  3. Install the required Matlab toolboxes;
  4. Create the following empty subdirectories:
    • ./mydir/Classification;
    • ./mydir/Datasets;
    • ./mydir/Debug;
      • ./mydir/Debug/Isoentropic;
      • ./mydir/Debug/Kapur;
      • ./mydir/Debug/Kittler-Illingworth;
      • ./mydir/Debug/Otsu;
      • ./mydir/Debug/Yen;
    • ./mydir/Features;
    • ./mydir/Labels;
    • ./mydir/LaTeX;
    • ./mydir/Thresholding;
    • ./mydir/ThresholdingResults;
  5. Download the images and extract them into ./mydir/Datasets;
  6. Download the TRAIN and TEST numbers and extract them into ./mydir;
  7. Check lines 1-15 of SCRIPT.m and modify them to fit your Matlab installation.
  8. Launch SCRIPT.m;
  9. Retrieve the results in LaTeX format in /mydir/data/LaTeX. The binarized images (i.e.: after thresholding) are stored in ./mydir/Debug.
NOTE: if you wish to use a different directory structure, please change the first part of SCRIPT.m accordingly.



DISCLAIMER
The information and content on this Web site are provided with no warranty whatsoever. Any use for scientific or any other purpose is conducted at your own risk and under your own responsibility. The authors are not liable for any damages, including any consequential damages, of any kind that may result to the user from the use of the materials on this Web site or of any of the products or services described hereon.