Investigation of methods of estimating geometrical misalignment in face images

Two methods of estimating the geometrical misalignment parameters for portrait images are considered in the work. The first method is based on use of the Fourier transform and application of the correlative approach and the second is based on calculating the moment characteristics of an image. Results of experimental studies are presented. The advantage of the method based on the Fourier transformation is shown. Recommendations for both methods are given. Evgenii Valer’evich Myasnikov. Born in 1981. Graduated from Samara State Aerospace University (SSAU) 2004. Completed his post-graduate studies at SSAU and received his candidate’s degree in 2007 (technical sciences). Myasnikov is currently with both the Institute of Image Processing RAS as a probationer-researcher and SSAU as a junior member of teaching at the geoinformatics chair. Scientific interests include the creation of software complexes, image processing and pattern recognition, and searching for images in databases. He is the author of 23 publications, including six articles, and a member of the Russian Association for Pattern Recognition and Image Analysis.