Medical Image Texture Features Classification Based on Gabor Wavelet Transform
Jun Xie · Jisuanji gongcheng · 2010
Gabor wavelet transform lacks in its ability to classify the medical CT image if it's rotation invariant image. Aiming at the problem,an approach is presented for rotation invariant medical texture classification based on Gabor wavelet transform. Rotation normalization is achieved by circular shift of the feature elements,so that all images have the same dominant direction. Experimental result shows that Gabor wavelet transform with circular operator of rotation normalization has well precision to classify the medical CT image.