Signal processing What are Gabor filters

What is the Gabor filter and what is its main purpose?


I do research on the Gabor filter but when I googled it I had very long and complicated articles. Can someone help me find a simple explanation or recommend a website or article to read? I want to understand this filter in order to be able to use it in Matlab.





Reply:


Gabor filters are orientation sensitive Filters used for texture analysis.

Which usually travel in packs, one for each direction. A Gabor filter set with a given direction gives a strong answer to locations of the target images that have structures in this given direction. For example, if your target image consists of a periodic grid in a diagonal direction, you will only get a strong response with a Gabor filter set if its direction is the same as that of the grid.

I know it's used a lot for character recognition and fingerprint enhancement. I (try) use it in biomedical imaging to characterize the main orientation in fibril structures.

Here is a very good tutorial from Javier Movellan, pdf

And if you can read in French, here is a PDF on how to create filter banks by Adrien Marion







A Gabor filter is a parameterization of the idea of ​​edges. This combines two somewhat conflicting ideas: an abrupt transition AND a fuzzy idea of ​​where it is located.

It's a clever idea mathematically because it translates well in the Fourier domain: the Fourier transform of a Gabor is a Gaussian in Fourier space, and a Gaussian blob is the most neutral guess of something fuzzy you can do (remember to throw darts and to look) at the patterns of hits).

So when you use a Gabor, there is no “right” formula: it all depends on what you want to detect / filter. In visual neuroscience, a Gabor is often chosen that corresponds to a spot on the logarithm of the frequencies in Fourier space (according to Weber's law, we are sensitive to relative differences in frequencies). These are Log Gabor filters.

To understand the Gabor filters, first check which filter parameters are best suited for your specific application.


It is a Edge detector . Only the Gabor transformation is used. The Gabor filter is basically a Gaussian filter (with variances sx and sy along the x and y axes) that modulates using a complex sinusoid (with center frequencies U and V along the x and y axes) becomes. See an example here.



We use cookies and other tracking technologies to improve your browsing experience on our website, to show you personalized content and targeted ads, to analyze our website traffic, and to understand where our visitors are coming from.

By continuing, you consent to our use of cookies and other tracking technologies and affirm you're at least 16 years old or have consent from a parent or guardian.

You can read details in our Cookie policy and Privacy policy.