Abstract
A real-time algorithm is proposed for the detection of abrupt scene changes, which makes use of a dual (one big and one small) window and single-side checking to avoid the false detection and miss detection caused by the violent motion of camera and/or large objects. In addition, a multi-layer perceptron is used to solve the problem of parameter determination in the proposed algorithm. The performance of our algorithm has been experimentally compared with that of some typical methods by using real video sequences. The recall rate is greatly improved while keeping high precision rate.
This work has been supported by NNSF (69672029) and HTP (863-317-9604-05).
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
J.S. Boreczky and L.A. Rowe, SPIE, V.2664: 170–179, 1996.
R.M. Ford et al., Proceedings of IEEE ICMCS, 610–611, 1997.
V. Kobla and D. Doermann, SPIE, V.3022: 200–211, 1997
H.J. Zhang et al., Multimedia Systems, V.1: 10–28, 1993.
B.L. Yeo and B. Liu, IEEE Trans. CSVT-5: 533–544, 1995.
K. Fukushima, Neural Network, V.1: 119–130, 1988.
A. Van Ooyen and B. Nienhuis, Neural Network, V.5: 465–471, 1992
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1999 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Lu, H.B., Zhang, Y.J. (1999). Detecting Abrupt Scene Change Using Neural Network. In: Huijsmans, D.P., Smeulders, A.W.M. (eds) Visual Information and Information Systems. VISUAL 1999. Lecture Notes in Computer Science, vol 1614. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-48762-X_37
Download citation
DOI: https://doi.org/10.1007/3-540-48762-X_37
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-66079-8
Online ISBN: 978-3-540-48762-3
eBook Packages: Springer Book Archive