Current Proceedings on Technology

Yazarlar: A.M. Adeshina, R. Hashim, N.E.A. Khalid, Siti Z. Z. Abidin

Konular:-

Anahtar Kelimeler: Face detection analysis,Face detection and tracking, Human computer interaction,Computer vision and image processing

Özet: Face Detection Analysis has been an active research area in Computer Vision and Image Processing over decades and it is fundamental to many facial processing applications. Consequently, real-time Face Detection and Tracking, for the purpose of facial recognition, has remained an important research area in many human-computer interaction application scenarios. Previously, research has been performed to introduce methods to accurately and rapidly detect faces individually or in a group, i.e. differentiate a human face from any other object. One of the approaches consists of a combination of four concepts based on Haar Wavelets, which accurately detect facial features. Presently, researchers have discovered that infrared (IR) facial imaging is independent of external illumination and shadow problems; however infrared camera is very costly. This work describes an accomplishment of face detection and tracking through direct image inputs from an infrared camera, using OpenCV’s face detector, which is based on haar-cascades algorithm. Uniquely, our work uses a low cost webcam/camera, modified to function as an infrared camera. Moreover, using this method, it was also possible to achieve illumination-free face detection at a reasonably good speed of image capture.  

---