Convolving an image with 2D log gabor and as the result we receive two Gabor maps- amplitude map and orientation map.

Hi , I need Matlab code for - Convolving an image with 2D log gabor and as the result of we receive two Gabor maps. The map with the maximum values of amplitude M(x;y) , is estimated as the highest magnitude of amplitude among 180 values of different orientation (for 180 corresponding filter responses). The second map Q(x;y) contains the dominant texture orientation angles , i.e. angles for which the magnitude of amplitudes is maximum). Thanks and regards Richa

Answer 1

[Image Analyst]

Take a look at conv2() or imfilter() to do your convolution. Then check out imregionalmax() to find local maxima in the resulting image.

Answer 2

[Laurent]

Perhaps your 2 maps correspond to the real and imaginary maps? (and thus to symmetric or odd log-Gabor filters)

if you are interested in more info + python code, something similar is coded using the package available @ https://pythonhosted.org/LogGabor/