domingomery / balu3

balu3

Python implementation for Balu Toolbox, based on previous implementations: Balu for Matlab (implemented by Domingo Mery), pyBalu for Python (implemented by Marco Bucchi), and pyXvis (implemented by Domingo Mery and Christian Pieringer).

Authors

Domingo Mery, Christian Pieringer and Marco Bucchi

Examples

Examples implemented in Balu3 using Google Colab.

Modules

fx -- Feature Extraction

fx.geo -- Geometric features

Function	Description
basicgeo	Basic geometric features
hugeo	Hu moments
flusser	Flusser moments
gupta	Gupta moments
fourierdes	Fourier descriptors
efourierdes	Elliptic Fourier descriptors
fit_ellipse	Elliptic feaures
lbp	Local Binary Patterns

fx.chr -- Chromatic features

Function	Description
basicint	Basic intensity features
hugeo	Hu moments
lbp	Local binary patterns
hog	Histogram of gradients
haralick	Haralick texture features
gabor	Gabor features (by Balu)
sci_gabor	Gabor features (by scimage)
fourier	Fourier features (DFT)
dct	Cosine transform features (DCT)
clt	Crossing line profile
contrast	Contrast features

ft -- Feature Transformation

ft.norm -- Normalization

Function	Description
minmax	MinMax normalization
mean0	mean=0, std=1 normalization

ft.trans -- Linear transformation

Function	Description
pca	Principal component analysis

fs -- Feature Selection

fs.sel -- Selection algorithms

Function	Description
jfisher	Fisher score
sp100	specifity=1 score
clean	Cleaning
sfs	Sequential forward selection
exsearch	Exhaustive search selection

io -- Input/Output

io.misc -- Miscellaneous

Function	Description
dirfiles	Files of a drectory
num2fixstr	Number to string with fixed 0s
imageload	Load of an image with indices

• im -- Image Processing

Requirements

• Python 3.6 or higher
• Numpy
• Scipy
• Matplotlib
• OpenCV 4.0 or higher

Instalation

In the first installation use directly in the directory folder:

pip install .

Then, to upgrade:

pip install --upgrade .

Or:

git clone https://github.com/domingomery/balu3

pip install ./balu3