Publication Type : Conference Paper
Thematic Areas : Amrita e-Learning Research Lab
Publisher : Proceedings - 2016 International Conference on Information Science, ICIS 2016
Source : Proceedings - 2016 International Conference on Information Science, ICIS 2016, Institute of Electrical and Electronics Engineers Inc. (2017)
ISBN : 9781509019861
Keywords : Face detection methods, Face detection technique, Face detector, Face recognition, False detections, Haar-like features, Image processing, Low illuminations, NPD Features, Number of peoples, Product development
Campus : Amritapuri
School : School of Engineering
Center : E-Learning
Department : E-Learning, Electrical and Electronics
Verified : Yes
Year : 2017
Abstract : pPeople counting systems have wide range of applications and it is a difficult task in image processing. Face detection techniques are widely used to count the presence of people in an image. In this paper we present a hybrid face detection method for people count by combining three efficient face detection methods - Normalized Pixel Difference (NPD) , haar classifier and haar classifier for profile face. NPD is a scale invariant method and its performance is good in difficult backgrounds and low illuminations. A combination of these three face detectors are used in order to improve the accuracy and to reduce false detections. Experimental results shows that our proposed hybrid method effectively count the number of people and reduced the false detection rates. The accuracy of the proposed method is up to 90.20%. © 2016 IEEE./p
Cite this Research Publication : A. Neethu, S. Narayanan, A., and Kamal Bijlani, “People Count Estimation using Hybrid Face Detection Method”, in Proceedings - 2016 International Conference on Information Science, ICIS 2016, 2017.