本篇提出了一個基於時間相干性和彩色獨特性的背景擷取法。此方法將背景擷取視為一個影像融合（image fusion）的過程，並使用無動量遮罩 (motion-less mask) 擷取以及加權引導影像濾波器這兩種技術。因為背景被假設為靜止的，基 於像素間時間相干性以及色彩顯著性，進行無動量遮罩 (motion-less mask) 擷取。一個改進的加權引導影像濾波器，稱為對比計算加權引導影像濾波器（contrastmeasure weighted guided image filter)，被使用在計算每個 frame，基於無動量遮罩以及對比靈敏度的權重圖 (weight map) 上面。引導影像濾波的過程將會平滑權重圖上面的邊緣，以及透過無動量遮罩上的移動的資訊和 frame 上面的紋理資訊來產生更細微的像素權重。透過這些權重圖產生沒有失真的背景圖片。實驗結果表示，比起多數最先進的演算法，此論文提出的方法表現更好。即使影片當中擁有複雜移動的前景物體，此論文提出的方法仍然可以移除所有的前景物體並產生乾淨的背景圖片。

In this paper, we present a novel method for robust background extraction
that exploits temporal coherence and color distinctness. The proposed method formulate the background extraction process as a image fusion process and estimate the frame weights based on two kernel techniques: motion-less mask extraction and contrast weighted guided image filter. The improved weighted guided image filter, call contrast measure guided image filter (CMGIF), which is developed for computing the weight map of each frame based on the contrast sensitivity and motion-less mask. The guide filtering process will smooth the edge and generate more detailed pixel weighting on motion-less mask by utilizing the motion information and texture information in motion-less mask and video frame, respectively. The no-artifact background extraction result is obtained based on the weighted maps. In the experiment, our proposed method has better performance when compare to the state-of-art background methods. Even though the test video sequence contain complex foreground movement, our method still can remove all the foreground object and artifacts to obtain the clear background result.

1] L. Yang, H. Cheng, J. Su, and X. Li, “Pixel-to-model distance for robust back-
ground reconstruction,” IEEE Transactions on Circuits and Systems for Video
Technology, vol. 26, pp. 903–916, May 2016.

[2] A. Colombari and A. Fusiello, “Patch-based background initialization in heavily
cluttered video,” IEEE Transactions on Image Processing, vol. 19, pp. 926–933,
April 2010.

[3] C.-C. Chen and J. K. Aggarwal, “An adaptive background model initializa-
tion algorithm with objects moving at different depths,” IEEE International
Conference on Image Processing, pp. 2664–2667, 2008.

[4] C. Stauffer and W. E. L. Grimson, “Adaptive background mixture models for
real-time tracking,” IEEE Computer Society Conference on Computer Vision
and Pattern Recognition, vol. 2, p. 252, June 1999.

[5] Y. Liu and D. A. Pados, “Compressed-sensed-domain l1-pca video surveillance,”
IEEE Transactions on Multimedia, vol. 18, pp. 351–363, January 2016.

[6] S. Aqel, A. Aarab, and M. A. Sabri, “Traffic video surveillance: background
modeling and shadow elimination,” International Conference on Information
Technology for Organizations Development (IT4OD), March 2016.

[7] O. Barnich and M. V. Droogenbroeck, “Vibe: A universal background subtrac-
tion algorithm for video sequences,” IEEE Transactions on Image Processing,
vol. 20, pp. 1709–1724, June 2011.

[8] D. K. Yadav, “Efficient method for moving object detection in cluttered
background using gaussian mixture model,” International Conference on Advances in Computing, Communications and Informatics (ICACCI), pp. 943–
948, September 2016.

[9] R. Kapoor and M. Sushma, “Fast tracking algorithm using modified potential
function,” Institution of Engineering and Technology Computer Vision, vol. 6,
pp. 111–120, March 2012.

[10] S. Kim, D. W. Yang, and H. W. Park, “A disparity-based adaptive multiho-
mography method for moving target detection based on global motion com-
pensation,” IEEE Transactions on Circuits and Systems for Video Technology,
vol. 26, pp. 1407–1420, August 2016.

[11] Z. Chao, W. Xiaopei, and Z. Jianying, “A moving object detection algorithm
based on improved gmm and short-term stability measure,” Journal Of Elec-
tronics and Information Technology, vol. 34, p. 2402, February 2012.

[12] C.-C. Chiu, M.-Y. Ku, and L.-W. Lian, “A robust object segmentation system
using a probability-based background extraction algorithm,” IEEE Transac-
tions on Circuits and Systems for Video Technology, vol. 20, pp. 518–528, April 2010.

[13] Y. Qin, S. Sun, X. Ma, S. Hu, and B. Lei, “A background extraction and shadow
removal algorithm based on clustering for vibe,” International Conference on
Machine Learning and Cybernetics, vol. 1, pp. 52–57, January 2014.

[14] W.-S. Hur, S.-T. Choi, S.-W. Kim, and S.-W. Seo, “Precise free space detection
and its application to background extraction,” International Conference on
Cybernetics and Intel ligent Systems (CIS) and IEEE Conference on Robotics,
Automation and Mechatronics (RAM), pp. 179–184, September 2015.

[15] D.-M. Tsai and S.-C. Lai, “Independent component analysis-based background
subtraction for indoor surveillance,” IEEE Transactions on Image Processing,
vol. 18, pp. 158–167, January 2009.

[16] D. Park and H. Byun, “A unified approach to background adaptation and
initialization in public scenes,” Pattern Recognition, vol. 46, pp. 1985–1997,July 2013.

[17] B. Gloyer, H. K. Aghajan, K.-Y. Siu, and T. Kailath, “Video-based freeway-
monitoring system using recursive vehicle tracking,” Image and Video Process-
ing III, pp. 173–180, 1995.

[18] C. Stauffer and W. E. L. Grimson, “Learning patterns of activity using real-time
tracking,” IEEE Transactions on Pattern Analysis and Machine Intel ligence,
vol. 22, pp. 747–757, August 2000.

[19] D. Baltieri, R. Vezzani, and R. Cucchiara, “Fast background initialization with
recursive hadamard transform,” IEEE International Conference on Advanced
Video and Signal Based Surveil lance (AVSS), pp. 165–171, August 2010.

[20] L. Li, W. Huang, I. Y.-H. Gu, and Q. Tian, “Statistical modeling of complex
backgrounds for foreground object detection,” IEEE Transactions on Image
Processing, vol. 13, pp. 1459–1472, November 2004.

[21] D. Russell and S. Gong, “A highly efficient block-based dynamic background
model,” IEEE Conference on Advanced Video and Signal Based Surveil lance,
pp. 417–422, September 2005.

[22] V. Reddy, C. Sanderson, and B. C. Lovell, “An efficient and robust sequential
algorithm for background estimation in video surveillance,” IEEE International
Conference on Image Processing, pp. 1109–1112, 2009.

[23] V. Reddy, C. Sanderson, and B. C. Lovell, “A low-complexity algorithm for
static background estimation from cluttered image sequences in surveillance
contexts,” Journal on Image and Video Processing, vol. 2011, pp. 1–14, April
2011.

[24] S.-S. Huang, L.-C. Fu, and P.-Y. Hsiao, “Region-level motion-based foreground
segmentation under a bayesian network,” IEEE Transactions on Circuits and
Systems for Video Technology, vol. 19, pp. 522–532, April 2009.

[25] J. Yang, W. Sun, X. Te, and K. Li, “Background extraction from video se-
quences via motion-assisted matrix completion,” IEEE International Confer-
ence on Image Processing, pp. 2437–2441, January 2014.

[26] S. Li, X. Kang, and J. Hu, “Image fusion with guided filtering,” IEEE Trans-
actions on Image Processing, vol. 22, pp. 2864–2875, July 2013.

[27] K. He, J. Sun, and X. Tan, “Guided image filtering,” IEEE Transactions on
Pattern Analysis and Machine Intel ligence, vol. 35, pp. 1397–1409, June 2013.

[28] Z. Li, J. Zheng, Z. Zhu, W. Yaoand, and S. Wu, “Weighted guided image filter-
ing,” IEEE Transactions on Image Processing, vol. 24, pp. 120–129, January
2015.

[29] G. Petschnigg, R. Szeliski, M. Agrawala, M. Cohen, H. Hoppe, and K. Toyama,
“Digital photography with flash and no-flash image pairs,” ACM Transactions
on Graphics (TOG), vol. 23, pp. 664–672, August 2004.

[30] C. Ttofis, C. Kyrkou, and T. Theocharides, “A low-cost real-time embedded
stereo vision system for accurate disparity estimation based on guided image
filtering,” IEEE Transactions on Computers, vol. 65, pp. 2678–2693, September
2016.

[31] X. Xu, R. Che, R. Nia, B. He, M. Chen, and A. Lendasse, “Underwater 3d
object reconstruction with multiple views in video stream via structure from
motion,” OCEANS 2016 - Shanghai, pp. 1–5, April 2016.

[32] E. I. Elhefnawy, H. S. Ali, and I. I. Mahmoud, “Effective visibility restora-
tion and enhancement of air polluted images with high information fidelity,”
National Radio Science Conference (NRSC), pp. 943–948, February 2016.

[33] Z. Li and J. Zheng, “Edge-preserving decomposition-based single image haze
removal,” IEEE Transactions on Image Processing, vol. 24, pp. 5432–5441,
December 2015.

[34] D.-W. Jang, H.-S. Kim, C.-D. Jung, and R.-H. Park, “Color fringe correction
based on image fusion,” IEEE International Conference on Image Processing
(ICIP), pp. 1817–1821, October 2014.

[35] S. Joshi, K. P. Upla, and M. V. Joshi, “Multi-resolution image fusion using
multistage guided filter,” National Conference on Computer Vision, Pattern
Recognition, Image Processing and Graphics (NCVPRIPG), pp. 1–4, December
2013.

[36] J. Duan, L. Chen, and C. P. Chen, “Region-based multi-focus image fusion
using guided filtering and greedy analysis,” IEEE International Conference on
Systems, Man, and Cybernetics (SMC), pp. 2932–2937, October 2015.

[37] “Background initialization website.”

[38] “Scene background initialization (SBI) dataset.”

[39] L. Maddalena and A. Petrosino, “Towars benchmarking scene background ini-tialization,” New Trends in Image Analysis and Processing, pp. 469–476, June 2015.

[40] M. Kumar, P. Ramuhalli, and S. Dass, “A total variation-based algorithm for
pixel-level image fusion,” Image Processing, IEEE Transactions on, vol. 18,
p. 2137–2143, September 2009.

[41] R. Marie, A. Potelle, and E. M. Mouaddib, “Dynamic background subtrac-
tion using moments,” IEEE International Conference on Image Processing,
pp. 2369–2372, September 2011.

[42] D.-M. Tsai and S.-C. Lai, “Independent component analysis-based background
subtraction for indoor surveillance,” IEEE Transactions on Image Processing,
vol. 18, pp. 158–167, January 2009.

[43] P. Soille, “Morphological image analysis: Principles and applications,”
Springer.

[44] W. Long and Y.-H. Yang, “Stationary background generation: an alternative to the difference of two images,” Pattern Recognition, vol. 23, pp. 1351–1359,1990.

[45] K.-L. Hua, H.-C. Wang, A. H. Rusdi, and S.-Y. Jiang, “A novel multi-focus
image fusion algorithm based on random walks,” J. Vis. Commun, Image Rep-
resent, 2014.

[46] K. Toyama, J. Krumm, B. Brumitt, and B. Meyers, “Wallfcolombari2010er:
Principles and practice of background maintenance,” IEEE International Con-
ference on Computer Vision, vol. 1, pp. 255–261, 1999.

[47] R. Cucchiara, C. Grana, M. Piccardi, and A. Prati, “Detecting moving objects,ghosts and shadows in video streams,” IEEE Transactions on Pattern Analysis and Machine Intel ligence, vol. 25, pp. 1337–1342, October 2003.

[48] L. Li, P. Wang, Q. Hu, and S. Cai, “Efficient background modeling based
on sparse representation and outlier iterative removal,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 26, pp. 278–289, February 2016.

[49] S. R. Sawalakhe, K. K. Kathalkar, M. P. Giri, N. B. Bhawarkar, S. R.
Sawalakhe, I. G. Deshmukh, and A. S. Joshi, “A novel background extraction
algorithm for large motion video,” Conference on Power, Control, Communi-
cation and Computational Technologies for Sustainable Growth (PCCCTSG),
pp. 254–260, December 2015.

[50] Z. Wei, P. Li, and HuangYue, “A foreground-background segmentation algo-rithm for video sequences,” International Symposium on Distributed Computingand Applications for Business Engineering and Science, pp. 340–343, March 2015.

[51] S. Mahmoudpour and M. Kim, “Robust foreground detection in sudden illumi-nation change,” Electronics Letters, vol. 52, pp. 441–443, March 2016.

[52] H.-Y. Wang, L.-H. Wang, and C.-B. Wu, “An efficient background extraction and object segmentation algorithm for realtime applications,” IEEE Asia Pacific Conference on Circuits and Systems (APCCAS), pp. 659–662, December 2012.

[53] L. Yang, H. Cheng, J. Su, and X. Chen, “Pixel-to-model background modeling in crowded scenes,” IEEE International Conference on Multimedia and Expo (ICME), pp. 1–6, July 2014.

[54] Z. Jingyun, D. Yifan, Y. Yi, and S. Jiasong, “Real-time defog model based on visible and near-infrared information,” 2016 IEEE International Conference on Multimedia Expo Workshops (ICMEW), pp. 1–6, July 2016.

[55] Z. Li, K. Wang, W. Zuo, D. Meng, and L. Zhang, “Detail-preserving and
content-aware variational multi-view stereo reconstruction,” IEEE Transactions on Image Processing, vol. 25, pp. 864–877, February 2016.

[56] E. I. Elhefnawy, H. S. Ali, and I. I. Mahmoud, “Effective visibility restoration and enhancement of air polluted images with high information fidelity,”National Radio Science Conference (NRSC), pp. 943–948, February 2016.

[57] A. Sharma, V. Bhateja, and A. K. Sinha, “Synthesis of flash and no-flash image pairs using guided image filtering,” Signal Processing and Integrated Networks (SPIN), 2015 2nd International Conference on, pp. 768–773, February 2015.

[58] H. Shi, L. Chen, Z. hong Mai, H. Chen, and F. kun Bi, “The guided image filtering for speckle reduction in sar images,” IET International Radar Conference 2015, pp. 1–4, October 2015.

[59] F. Zeev, F. Raanan, L. Dani, and S. Richard, “Edge-preserving decompositions for multi-scale tone and detail manipulation,” ACM Trans. Graph., vol. 27,pp. 67:1–67:10, October 2008.

[60] F. Durand and J. Dorsey, “Fast bilateral filtering for the display of high-dynamic-range images,” ACM Trans. Graph., vol. 21, pp. 257–266, July 2002.

[61] L. Xu, C. Lu, Y. Xu, and J. Jia, “Image smoothing via l0 gradient minimization,” ACM Trans. Graph., vol. 30, pp. 174:1–174:12, December 2011.