/ METHOD AND APPARATUS FOR REPRESENTING AND RETRIEVING 3D IMAGE DATA USING 3D ZERNIKE MOMENTS AND FREQUENCY PLANE DIVIDED FEATURES

摘要

PURPOSE: An apparatus and a method for extracting image feature and an apparatus and method for searching image are provided to extract image feature irrelevant to movement, scale and rotation of an image, and search images fast. CONSTITUTION: Image feature is extracted by extracting a zenike feature value from an input image(S1), extracting a frequency plane division feature value from the image(S2), and extracting n-th order statistical feature value from the image(S3). The zenike feature value is extracted by performing DFT(discrete Fourier transformation) on the input image, normalizing the DFT value, and obtaining a zenike moment absolute value. The frequency plane division feature value is extracted by Gabor filters, Radon transformation or wavelet transformation. The n-th order statistical feature value is a first order moment such as a mean, an L1 norm and a variance of energy, or a second order moment such as energy.