Skip to content

Image quantization as a dimensionality reduction procedure in color and texture feature extraction

License

Notifications You must be signed in to change notification settings

GabiThume/paper-image-quantization

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Image quantization as a dimensionality reduction procedure in color and texture feature extraction

Description

  1. Image feature extraction, using quantization techniques and image descriptors.

Quantization Techniques:

Gleam
Intensity
Luminance
MSB

Descriptors:

BIC
GCH
CCV
Haralick
ACC
  1. Dimensionality reduction of feature vectors.

Dimensionality Reduction Techniques:

PCA
Entropia
  1. Image classification.

Classifier:

Naive Bayes
KNN

Use

Before running the code, create a symbolic link to the images directory:

ln -s <IMAGES DIRECTORY> BaseImagens

Makefile will compile the code for you:

make

To run all descriptors and generate the feature vectors:

./runAllDescriptors.sh

After the previous command, to reduce the vectors dimension and apply the classification run:

./dimensionReduction <VECTORS DIRECTORY> <TECHNIQUE> <PARAMETERS LIST>

Options for techniques and parameters:

[0] None:
    Just the classifier is going to be used, with the extracted vectors without dimensionality reduction.
[1] PCA: 
    - <nAttributes>: number of attributes to keep on PCA
[2] Entropy:
    - <tWindow>: window size
[3] All:
    - <nAttributes>: number of attributes to keep on PCA
    - <tWindow>: window size

The classification analysis (using Naive Bayes classifier and Repeated subsampling as a cross validation method) is going to be printed on the terminal. So, to write on a file the results of the analysis:

./dimensionReduction <VECTORS DIRECTORY> <TECHNIQUE> <PARAMETERS LIST>  >  analysis/<TECHNIQUE>_<PARAMETERS>.txt

Examples:

./dimensionReduction caracteristicas_corel/256/ 1 35 > analysis/Corel/PCA_50.txt
./dimensionReduction caracteristicas_corel/256/ 2 4 > analysis/Corel/ENTROPY_4.txt

About

Image quantization as a dimensionality reduction procedure in color and texture feature extraction

Resources

License

Stars

Watchers

Forks

Packages

No packages published