INCREASING THE EFFECTIVENESS OF THE ALGORITHMS FOR DETECTING FACES USING THE VIOLA-JONES METHOD
DOI:
https://doi.org/10.26906/SUNZ.2018.5.079Keywords:
face detection, Viola-Jones method, frame brightness automatic stabilization, face and its parts detection probabilityAbstract
There are proposed new methods for increasing the efficiency of algorithms for detecting faces on digital images and video sequences based on the Viola-Jones method and used in solving face recognition problems. This allows to eliminate theinfluence of the one main interference factors - to compensate the effect of changes in the scene illumination level on face detection quality. For this purpose, automatically stabilizing frame brightness procedure is additionally introduced into the classical structure of these algorithms. The structure of algorithms is described and software is developed using Python programming language and OpenCV library resources, that allows to conduct video data processing in real time. There is proposed and programmatically implemented an original method for estimating the algorithm efficiency based on the maximum probability criterion of correct faces and their main elements (eye, nose, mouth) detection. The results of the classical and proposed algorithms are compared. Examples of work and software testing results are given.Downloads
References
Viola P., Jones M.J. Rapid object detection using a boosted cascade of simple features // IEEE Conf. on Computer Vision and Pattern Recognition. – Kauai, Hawaii, USA, 2001. – V. 1. – P. 511–518.
Viola P., Jones M.J. Robust real_time face detection // Int. Journal of Computer Vision. – 2004. – V. 57. – № 2. – P. 137–154.
Joseph Howse, Joe Minichino, Learning OpenCV 3 Computer Vision with Python, Packt Publishing, 2015, Packt Publishing.
Saurabh Kapur, Computer Vision with Python 3, Packt Publishing, August 2017, ISBN: 978-1-78829-976-3.
Prateek Joshi, OpenCV with Python By Example, Packt Publishing, September 2015, ISBN: 978-1-78528-393-2..
Библиотека компьютерного зрения OpenCV [Электронный ресурс]. – Режим доступа: http://docs.opencv.org/trunk/doc/py_tututoria/py_objdetect/py_face_detection/py_face_detection.html.
Разработка мультимедийных приложений с использованием библиотек OpenCV и IPP [Электронный ресурс] / А. В. Бовыкин [и др.]/ – Электрон. текстовые данные. – М.: Интернет-Университет Информационных Технологий (ИНТУИТ), 2016. – 515 с. – Режим доступа: http://www.iprbooksshop.ru/39564/
F. Comasch, S. Stuijk, T. Basten and H. Corporaal "Rasw: A runtime adaptive sliding window to improve viola-jones object detection", Distributed Smart Cameras (ICDSC), 2013.
