Synthesis and Image Matching On Structural Patterns Using Affine Transformation

This paper focuses in explaining a Fourier based affine estimator which is applied to the task of Image Synthesis. An affine transformation is an important class of linear 2-D geometric transformations which maps variables into new by applying a linear combination of translation, rotation, scaling and/or shearing operations. Conventional retrieval systems are very effective when knowledge information and query information are in a uniform orientation but fails in recognition when effects such as scaling, orientation exist. But as this technique is based on texture analysis, which is termed the affine estimator, it will even match the images with non-uniform orientation.