hadamard-gabor filter: a rotation invariant feature extraction algorithm

Gabor filter is a well-known feature extraction algorithm in image processing with many applications intexture segmentation, fractal dimension management, document analysis, edge detection, retina identification, imagecoding and image representation. the gabor filter has received considerable attention mainly because the human visualsystem is similar to this filter. in this paper we propose hadamard-gabor filter a rotation-invariant feature extractionalgorithm. in our new gabor filter we apply walsh-hadamard and johnson-lindenstrauss transforms on the features ofan image extracted using gabor filter. the walsh-hadamard transform makes gaber filter rotation-invariant and thejohnson-lindenstrauss transform reduces the dimension of extracted features to a lower (target) dimension. many imageprocessing algorithms use feature extraction algorithms as one of their basic building blocks. using hadamard-gaborfilter instead of gabor filter makes them potentially faster as they deal with features of lower dimensions.