A Comparison between SIFT descriptor and Zernike moments feature extraction on geometric shapes representation

The process of image retrieval and recognition is based on image feature extraction. These features have specific structures in the image such as points, edges or objects. The feature extraction is considered as a method in order to represent the images with sufficient accuracy. Yet, one challenging problem in computer vision is how to extract ideal features that can reflect the intrinsic content of the images. So the proposed study summarizes two robustness feature extraction methods: Sift descriptors and Zernike moments. The robustness of studying approaches has been tested against different types of distortion. The processing technique and performance comparison of Zernike moments and SIFT is described based on the precision-recall criterion

Keywords:

Scale Invariant Feature Transform (SIFT) , Zernike moments (ZMs) , Featuresextraction , Geometric shapes (GS)