Title :
A new method for pedestrian counting
Author :
Zhang, Xiaowei ; Sexton, Graham
Author_Institution :
Univ. of Northumbria, Newcastle, UK
Abstract :
This paper presents a new method for accurate real-time counting of pedestrians using model-based image processing techniques. In order to minimise computational costs, a specially designed model-specific filter (MSF) is used during the image search phase to obtain the best initial estimate position for the model matching process. It is planned to port the algorithm to a TMS320C50 image processing platform to facilitate a real time working system
Keywords :
feature extraction; filtering theory; image matching; image sequences; TMS320C50 image processing platform; computational costs minimisation; feature extraction; image search; image sequences; initial estimate position; model matching process; model-based image processing; model-specific filter; pedestrian counting; real time working system; real-time counting;
Conference_Titel :
Image Processing and its Applications, 1995., Fifth International Conference on
Conference_Location :
Edinburgh
Print_ISBN :
0-85296-642-3
DOI :
10.1049/cp:19950650
