This paper discusses the experimentation of a robust face recognition descriptor and its implementation in camera based lecture attendance system. The Speeded Up Robust Features (SURF) descriptor used for recognition produced a reliable performance with regard to change in illumination, scale and rotation of the test images used. Experiments were done on various test images with varying conditions in each image and obtained repeatable output. Finally, an interactive MATLAB GUI was created to mark the attendance and to locate a student along with his academic details.

1. Philippe Dreuw, Pascal Steingrube,SURF-Face: Face Recognition UnderViewpoint Consistency Constraints, Human Language Technology andPattern Recognition RWTH Aachen University,Aachen, Germany
2. Matthew Brown and David Lowe. 2002. Invariant feature from interest point groups. BMVC
3. Jianguo Li, Yimin Zhang. CVPR2013. Learning SURF Cascade for Fast and Accurate Object Detection,Intel Labs China.
4. http://www. mathworks. in/help/vision/ref/estimatege ometrictransform. html. MATLAB 2013 documentation.
5. Geng Du, Fei Su, Anni Cai. MIPPR 2009: Pattern Recognition and Computer Vision.
6. Jianguo Li, Tao Wang, Yimin Zhang. Face detection using SURF. Intel labs, China.
7. Viola, P. , Jones, M. : Rapid object detection using a boosted cascade of simplefeatures. In: CVPR (1). (2001).

Attendance System Speeded Up Robust Features Face Recognition