Image Retrieval is one of the most important topics in image processing and computer vision. Proper Image Retrieval of an image requires the selection of suitable features that discriminate different images. Besides, ideal features must be invariant to factors such as rotation, scaling, brightness, and noise. The features that are used mostly in image processing include color, shape, and texture. The texture is an important feature among many types of images. So far, many approaches have been proposed to describe the texture, including statistical, structural, filters based, and models based methods. This thesis presents a novel wave-interference based rotation-invariant feature for texture description. In local pattern-based techniques, features are extracted from small circularly symmetric neighboring points. These features require further processing to incorporate rotation invariance. The proposed method in this thesis is based on wave-interference signals in a non-homogeneous environment. The non-homogeneousity of the environment at each pixel of an image depends on its grayscale. Therefore, this method is inherently invariant to any amount of rotation and needs no more operations for achieving rotation invariance. Also, the proposed method works well against brightness changes. Brodatz and VisTex databases have been used to evaluate the performance of the proposed method in image retrieval. The results show that the average precision of retrieving all images in each align="right" Key Words: 1. Image Processing, 2. Texture Descriptor, 3. Image Retrieval, 4. Local Feature
