Perspective Correction For Image Processing
Romana Oancea, Vlad-Andrei Bârsan, Cristian Constantin Molder · International conference KNOWLEDGE-BASED ORGANIZATION · 2015
Abstract The paper proposes an algorithm for applying the perspective correction of images containing human faces. Based on the coordinates of the two eyes, of the center of the mouth and of a point on the line of the mouth, one can estimate the angle of inclination of the face. After applying the perspective correction, the area corresponding to the face can be used by a face recognition algorithm based on aspect in order to improve the good recognition rate.