How to interpret the result of gabor texture

Question

Meenu 2013-2-23

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/64632-how-to-interpret-the-result-of-gabor-texture

I am applying gabor filter for image. How to interpret the result? what is the meaning if i get higher value in the gabor output image? where can i get details about this?

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Youssef Khmou 2013-2-23

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/64632-how-to-interpret-the-result-of-gabor-texture#answer_76225

编辑：Youssef Khmou 2013-2-25

hi Meenu

I am not specialist in the domain, but i know that Gabor filter ca be used for both Filtering (denoising) and for edges detection, so high values then means your aplication detects well the edges ,

I think "Image analyst" can answer your question.....

2 个评论
显示无隐藏无

Meenu 2013-2-25

Thanks for the response Youssef.

I am using Gabor wavelets for texture based image classification. I understood that applying gabor filter/wavelet for a local section of image describes the sinusoidal and phase content of the signal in that section. But i am unable find documents to interpret resulting value in the section. So far I have understood that if the gabor output is high, the signal in particular section highly matches with the applying gausian signal of the gabor. But I am not sure whether it is correct.

Youssef Khmou 2013-2-26

hi , sorry i wanted to say that i am not specialist in the domain, but in the mean while , take a look at the answer below .

请先登录，再进行评论。

Answer 2

Youssef Khmou 2013-2-26

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/64632-how-to-interpret-the-result-of-gabor-texture#answer_76566

hi Mennu,

your interpretation is correct because a two-dimensional (2-D) Gabor filter is linear and local. Its convolution kernel is a product of a Gaussian and a cosine function , take a look at this paper and look at the results obtained :

Comparison of Texture Features Based on Gabor Filters

Simona E. Grigorescu, Nicolai Petkov, and Peter Kruizinga IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 11, NO. 10, OCTOBER 2002

http://iwi.eldoc.ub.rug.nl/FILES/root/2002/IEEETIPGrigorescuS/2002IEEETIPGrigorescuS.pdf

I hope that helps .