Gabor_Feature_Extactor

KO | EN

gitlite — search

#javascript #python #hacktoberfest #react #ai #typescript #llm #go #golang #android #machine-learning #rust #deep-learning #linux

Gabor_Feature_Extactor

★ 9 Open GitHub ↗

C++ based Gabor feature extractor using OpenCV.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Gabor_Feature_Extactor

**C++ Based Gabor Feature Extractor using OpenCV**

This class extractors Gabor features of the input image using Gabor wavelets. You need OpenCV for using the class.

The Gabor wavelet used is of the form:

![wavelet_equation](https://cloud.githubusercontent.com/assets/7311045/15891645/4e714ce2-2d7e-11e6-98b0-ca6f339a2971.png)

where, μ and *v* define the orientation and scale of the Gabor filters, z=(x,y), ||.|| denotes the norm operator, and the wave vector ![wavelet_equation2](https://cloud.githubusercontent.com/assets/7311045/15892003/f44d6014-2d7f-11e6-8e6f-1bfc20f15b6b.png) where ![wavelet_equation3](https://cloud.githubusercontent.com/assets/7311045/15892074/79417706-2d80-11e6-9171-5483db04f35b.png) and ![wavelet_equation4](https://cloud.githubusercontent.com/assets/7311045/15892079/7cc48940-2d80-11e6-9e2c-45e4bcb609e5.png) is the spacing factor between filters in the frequency domain.

The Gabor representation of an image is derived by convolving the image with the Gabor filters. Let *I(x,y)* be the input image, its convolution with a Gabor filter ![wavelet_equation5](https://cloud.githubusercontent.com/assets/7311045/15892178/fb5ced74-2d80-11e6-9b88-e861476ee022.png) is defined as follows:

![wavelet_equation6](https://cloud.githubusercontent.com/assets/7311045/15892230/555b1b52-2d81-11e6-8aa5-13b186840bcc.png)

where, * denotes the convolution operator. Five scales ![wavelet_equation7](https://cloud.githubusercontent.com/assets/7311045/15892271/8853b514-2d81-11e6-86dd-83fe6bd71867.png) and eight orientations ![wavelet_equation8](https://cloud.githubusercontent.com/assets/7311045/15892319/b6786e12-2d81-11e6-82ae-ed415f351747.png) of Gabor filters are used. Convolving the image with each of the 40 Gabor filters can then generate the Gabor features.

Best wishes,

Ekberjan

Gabor_Feature_Extactor

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README