4 January 2006 Object-layering technology of video sequence with background restoration based on neural network
Author Affiliations +
Proceedings Volume 5985, International Conference on Space Information Technology; 59853D (2006) https://doi.org/10.1117/12.657985
Event: International Conference on Space information Technology, 2005, Wuhan, China
Abstract
The object-layering technology of Video Sequence is very important in many fields. Commonly, video image can be decomposed into background and moving objects. In this paper, based on features of different objects of the scene video sequence, aimed at the background object, we study the object-layering technology with background restoration based on neural network. This technology adopts two-stage pattern recognition and estimation with weighted multiple frames, increasing the reliability of the object-layering and extremely suppressing the noise. The experiments show the technology discussed is effective.
© (2006) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Yongqing Huo, Yongqing Huo, Qicong Peng, Qicong Peng, } "Object-layering technology of video sequence with background restoration based on neural network", Proc. SPIE 5985, International Conference on Space Information Technology, 59853D (4 January 2006); doi: 10.1117/12.657985; https://doi.org/10.1117/12.657985
PROCEEDINGS
5 PAGES


SHARE
RELATED CONTENT

Robust design of dilation and erosion CNN for gray scale...
Proceedings of SPIE (December 03 2015)
Content-based downloading for digital video
Proceedings of SPIE (January 03 1996)
OVID: toward object-based video retrieval
Proceedings of SPIE (December 23 1999)
A Simplified Scheme Of Analytic Video Predictors
Proceedings of SPIE (October 13 1987)

Back to Top