Selecting the Color Space for Self-Organizing Map Based Foreground Detection in Video

详细信息    查看全文

• 作者：Francisco J. López-Rubio ; Enrique Domínguez ; Esteban J. Palomo…
• 关键词：Probabilistic self ; organising maps ; Unsupervised learning ; Video segmentation ; Foreground detection ; Color space
• 刊名：Neural Processing Letters
• 出版年：2016
• 出版时间：April 2016
• 年：2016
• 卷：43
• 期：2
• 页码：345-361
• 全文大小：2,730 KB
• 参考文献：1.Balcilar M, Karabiber F, Sonmez A (2013) Performance analysis of Lab2000HL color space for background subtraction. In: 2013 IEEE International Symposium on Innovations in Intelligent Systems and Applications (INISTA) (2013)
  2.Bishop CM, Svenson M (1998) The generative topographic mapping. Neural Comput 10(1):215–234CrossRef
  3.Brainard DH (2003) The science of color. In: Shevell SK (ed) Color appearance and color difference specification. Elsevier, Oxford, pp 191–216
  4.Doshi A, Trivedi M (2007) Satellite imagery based adaptive background models and shadow suppression. Signal Image Video Process 1(2):119–132CrossRef MATH
  5.Gao T, Liu ZG, Yue SH, Mei JQ, Zhang J (2009) Traffic video-based moving vehicle detection and tracking in the complex environment. Cybern Syst 40(7):569–588CrossRef
  6.Jlassi M, Douik A, Messaoud H (2010) Color images segmentation algorithms during a sports meeting: application to soccer video images. J Circuits Syst Comput 19(6):1307–1332CrossRef
  7.Kushner HJ, Yin GG (2003) Stochastic approximation and recursive algorithms and applications. Springer, New YorkMATH
  8.López-Rubio E (2009) Multivariate student-t self-organizing maps. Neural Netw 22(10):1432–1447CrossRef
  9.López-Rubio E (2009) Robust location and spread measures for nonparametric probability density function estimation. Int J Neural Syst 19(5):345–357CrossRef
  10.López-Rubio E, Luque-Baena RM, Dominguez E (2011) Foreground detection in video sequences with probabilistic self-organizing maps. Int J Neural Syst 21(3):225–246CrossRef
  11.López-Rubio E, Ortiz-de-Lazcano-Lobato JM, López-Rodríguez D (2009) Probabilistic PCA self-organizing maps. IEEE Trans Neural Netw 20(9):1474–1489CrossRef
  12.Maddalena L, Petrosino A (2008) A self-organizing approach to background subtraction for visual surveillance applications. IEEE Trans Image Process 17(7):1168–1177. doi:10.​1109/​TIP.​2008.​924285 MathSciNet CrossRef
  13.Ming Y, Jiang J (2008) Background modeling and moving-objects detection based on Cauchy distribution for video sequence. Acta Optica Sinica 28(3):587–592CrossRef
  14.Ning J, Yang Y, Zhu F (2013) Background modeling and fuzzy clustering for motion detection from video. J Multimed 8(5):626–631CrossRef
  15.Rodgers JL, Nicewander WA (1988) Thirteen ways to look at the correlation coefficient. Am Stat 42(1):59–66CrossRef
  16.Sheikh Y, Shah M (2005) Bayesian modeling of dynamic scenes for object detection. Pattern Anal Mach Intell IEEE Trans 27(11):1778–1792. doi:10.​1109/​TPAMI.​2005.​213 CrossRef
  17.Shimada A, Taniguchi R (2010) Hybrid background modeling for long-term and short-term illumination changes. IEEJ Trans Electron Inf Syst 130(9):1524–1529
  18.Van Hulle M (2005) Maximum likelihood topographic map formation. Neural Comput 17(3):503–513CrossRef MATH
  19.Van Hulle MM (2002) Kernel-based topographic map formation by local density modeling. Neural Comput 14(7):1561–1573CrossRef MATH
  
• 作者单位：Francisco J. López-Rubio (1)
  Enrique Domínguez (1)
  Esteban J. Palomo (1)
  Ezequiel López-Rubio (1)
  Rafael M. Luque-Baena (2)
  
  1. Department of Computer Languages and Computer Science, University of Málaga, 29071, Málaga, Spain
  2. Department of Computer Systems and Telematics Engineering, University of Extremadura, 06800, Mérida, Spain
  
• 刊物类别：Physics and Astronomy
• 刊物主题：Physics
  Complexity
  Artificial Intelligence and Robotics
  Electronic and Computer Engineering
  Operation Research and Decision Theory
  
• 出版者：Springer Netherlands
• ISSN：1573-773X

文摘

Detecting foreground objects on scenes is a fundamental task in computer vision and the used color space is an important election for this task. In many situations, especially on dynamic backgrounds, neither grayscale nor RGB color spaces represent the best solution to detect foreground objects. Other standard color spaces, such as YCbCr or HSV, have been proposed for background modeling in the literature; although the best results have been achieved using diverse color spaces according to the application, scene, algorithm, etc. In this work, a color space and a color component weighting selection process are proposed to detect foreground objects in video sequences using self-organizing maps. Experimental results are also provided using well known benchmark videos.