Title :
Rotated Haar-Like features at generic angles for objects detection
Author :
Oualla, Mohamed ; Sadiq, Abdelalim ; Mbarki, Samir
Author_Institution :
Dept. of Inf., Ibn Tofail Univ., Kenitra, Morocco
Abstract :
This paper propose a new approch to detect rotated object at distinct angles using the Viola-Jones detector. Our idea is based on three key points: The first is to propose a method allowing finding possible angles to rotate a normal Haar-Like feature, without destroy its sides. The tangent of each angle equal to an rational number i.e. each angle equal to arctan(Y/X) where Y and X are integers. The second point is manifested by the use of this rational number to determine the pixels forming the sides of the feature. In the latter, we propose a solution that makes use of this rational number to approximate the Integral image of a rotated feature at a generic angle.
Keywords :
Haar transforms; approximation theory; object detection; Viola-Jones detector; integral image approximation; rotated Haar-like features; rotated object detection; Classification algorithms; Detectors; Face detection; Feature extraction; Integral equations; Object detection; Haar-like feature; Integral image; Object detection; Viola and Jones algorithm;
Conference_Titel :
Information Science and Technology (CIST), 2014 Third IEEE International Colloquium in
Conference_Location :
Tetouan
Print_ISBN :
978-1-4799-5978-5
DOI :
10.1109/CIST.2014.7016645