A neural-based crowd estimation by hybrid global learning algorithm
Authors:
- T.W.S. ChowTCT.W.S. Chow
- Chi-Tat LeungCLChi-Tat Leung
- 1 August 1999
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics)
- Vol. 29 (4) , 535-541
Abstract
A neural-based crowd estimation system for surveillance in complex scenes at underground station platform is presented. Estimation is carried out by extracting a set of significant features from sequences of images. Those feature indexes are modeled by a neural network to estimate the crowd density. The learning phase is based on our proposed hybrid of the least-squares and global search algorithms which are capable of providing the global search characteristic and fast convergence speed. Promising experimental results are obtained in terms of accuracy and real-time response capability to alert operators automatically.Keywords
This publication has 11 references indexed in Scilit:
- A Fast Heuristic Global Learning Algorithm for Multilayer Neural NetworksNeural Processing Letters, 1999
- Distributed data fusion for real-time crowding estimationSignal Processing, 1996
- Crowd monitoring using image processingElectronics & Communication Engineering Journal, 1995
- Handbook of Global Optimization1995
- DEKF system for crowding estimation by a multiple-modelapproachElectronics Letters, 1994
- Global optimization of statistical functions with simulated annealingJournal of Econometrics, 1994
- A real-time vision system for crowding monitoring1993
- A new approach for finding the global minimum of error function of neural networksNeural Networks, 1989
- Use of a Millimeter Wave Radiometer for Detecting Pedestrian and Bicycle Traffic1987
- Novel computer-based infrared pedestrian data-acquisition systemElectronics Letters, 1979