| Object recognition is one of the important parts of computer vision, and has been widely applied in the fields such as military affairs and automatization. In this paper, we study the method of recognizing object in its 2-D image. For the same object, the geometry shape of image obtained is different when the camera is located different orientation and angle. The geometry distortion of arbitrary two images can be described with perspective transform. The perspective transform can be replaced approximately with affine transform when the distance between the camera and the object is larger than the dimension of the object.The background and development of the image retrieval based on content is introduced. A view of shape matching methods is provided also. Some shape representation methods are presented, mainly the shape simplification methods. Shape matching methods can be classified into two groups, matching by kinds of invariants and matching by local features, and according to types, an overview of the most reprentative methods is presented.Wavelet analysis is a new time-frequency analysis method which has raised a research up surge internationally in recent years. It is regarded as a breakthrough of Fourier analysis. Object recognition using wavelet transform is an active research domain and has wide application areas. The basis of wavelet anlysis is addressed. Then the relationship between the wavelet shrinkage and diffusion equation is studied. A t last stationary wavelet transform has been used to derive an affine invariant function. It's a key problem to search for affine invariant with respect to translation, scaling, rotation and skewing in multi-resolution analysis. Considering the advantages of the stationary wavelet transform, a new affine invariant function based on analyzing the object boundary using the stationary wavelet transform is constructed. Experiment results demonstrate the discriminating power of the proposed invariant function. |
PDF Full Text Request