Background Extraction and Snow Remove Form Video

Abstract:

Capturing good videos outdoors can be challenging due to harsh lighting, unpredictable scene changes, and most relevant to this work, dynamic weather. Particulate weather, if there were a heavy snow in monitor scene, it would create complex flickering effects that are irritating to people and confusing to computer vision algorithms. The research direction of video image background estimation and background estimation in heavy snow is to improve the visualization of surveillance video, especially to reduce the affection of snowflake in heavy snow. This algorithm can remove snowflakes and retain non-snowflakes moving targets we interested in. We propose a snow removal method of the pixels affected by snowflakes based on probability estimation of snowflakes detected. Firstly, we detect the snowflakes based on the optical of snowflake and time difference method. Then we estimated the coverage rate of snowflakes based on the detecting result. Finally, we build an adaptive estimation method of background pixels value based on the information feedback of snow detection and the model of pixels value. The calculation of this method is simple and the method has good processing results.

• Periodical: Advanced Materials Research (Volumes 756-759)
• Pages: 1382-1386
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.756-759.1382
• Online since: September 2013
• Authors: Yu Tao Ren, Yi Chao Zhang, Ke Zhu
• Keywords: Snow Detection, Snow Removal, Surveillance Video

References

[1] Peter Barnum, Takeo Kanade, and Srinivasa G. Narasimhan, Spatio-Temporal Frequency Analysis for Removing Rain and Snow from Videos, Workshop on Photometric Analysis For Computer Vision (PACV), in conjunction with ICCV, (2007).

[2] Jing Xu, Wei Zhao, Peng Liu, Xianglong Tang. J. Computer and Information Science. Vol. 5(2012), p.49.

[3] K. Garg and S.K. Nayar. in Proc. ICCV. vol. 2(2005), p.1067.

[4] K. Garg and S.K. Nayar. ACM Transactions on Graphics(TOG). Vol. 25(2006), pp.996-1002.

[5] X. Zhang, H. Li, Y. Qi, et al. IEEE International Conference on Multimedia and Expo, (2006), pp.461-464.

[6] Kshitiz Garg, Shree K. Nayar in International Journal of Computer Vision (2007), pp.3-27.

[7] Jeremie Bossu, Nicolas Hautiere, Jeaan-Philippe Tarel: Rain or Snow Deteetion in Image Sequences Through Use of a Histogram of Orientation of Streaks (Springer Science+Business Media 2011).

DOI: 10.1007/s11263-011-0421-7