Cited By
View all- Liu SHu J(2016)Visual saliency based on frequency domain analysis and spatial informationMultimedia Tools and Applications10.1007/s11042-016-3903-375:23(16699-16711)Online publication date: 1-Dec-2016
Probability density function estimation with the frequency polygon transform
Pages 136 - 158
Abstract
A probability density function estimator is proposed, which is based on frequency polygons.Its convergence to the true density is formally proved.A mode finding algorithm is also proposed, as an alternative to mean-shift.Our approach outperforms histogram and kernel based estimators in synthetic and real datasets.Our proposal is shown to be suitable to object tracking in video sequences. Most current nonparametric approaches to probability density function estimation are based on the kernel density estimator, also known as the Parzen window estimator. A usual alternative is the multivariate histogram, which features a low computational complexity. Multivariate frequency polygons have often been neglected, even though they share many of the advantages of the histograms, while they are continuous unlike the histograms. Here we build on our previous work on histograms in order to propose a new probability density estimator which is based on averaging multivariate frequency polygons. The convergence of the estimator is formally proved. Experiments are carried out with synthetic and real machine learning datasets. Finally, image denoising and object tracking applications are also considered.
References
[1]
T. Ahooyi, M. Soroush, J. Arbogast, W. Seider, U. Oktem, Maximum-likelihood maximum-entropy constrained probability density function estimation for prediction of rare events, AIChE J., 60 (2014) 1013-1026.
[2]
W. Andrzejewski, A. Gramacki, J. Gramacki, Graphics processing units in acceleration of bandwidth selection for kernel density estimation, Int. J. Appl. Math. Comput. Sci., 23 (2013) 869-885.
[3]
A. Asuncion, D. Newman, UCI Machine Learning Repository. <http://www.ics.uci.edu/%7Emlearn/MLRepository.html>.
[4]
D. Bashtannyk, R. Hyndman, Bandwidth selection for kernel conditional density estimation, Comput. Statist. Data Anal., 36 (2001) 279-298.
[5]
T. Baum, I. Izhaki, E. Rivlin, G. Katzir, Active tracking and pursuit under different levels of occlusion: a two-layer approach, Mach. Vis. Appl., 25 (2014) 173-184.
[6]
J.L. Bentley, Multidimensional binary search trees used for associative searching, Commun. ACM, 18 (1975) 509-517.
[7]
A. Bhattacharyya, On a measure of divergence between two statistical populations defined by their probability distributions, Bull. Calcutta Math. Soc., 35 (1943) 99-109.
[8]
R. Cappelli, D. Maltoni, On the spatial distribution of fingerprint singularities, IEEE Trans. Pattern Anal. Mach. Intell., 31 (2009) 742-748.
[9]
Z. Chen, L. Zheng, Y. Chen, Y. Zhang, 2D hand tracking based on flocking with obstacle avoidance, Int. J. Adv. Rob. Syst., 11 (2014).
[10]
S. Chiu, An automatic bandwidth selector for kernel density estimation, Biometrika, 79 (1992) 771-782.
[11]
D. Comaniciu, P. Meer, Mean shift: a robust approach toward feature space analysis, IEEE Trans. Pattern Anal. Mach. Intell., 24 (2002) 609-619.
[12]
D. Comaniciu, V. Ramesh, P. Meer, Kernel-based object tracking, IEEE Trans. Pattern Anal. Mach. Intell., 25 (2003) 564-577.
[13]
F. Cordelières, V. Petit, M. Kumasaka, O. Debeir, V. Letort, S. Gallagher, L. Larue, Automated cell tracking and analysis in phase-contrast videos (iTrack4U): development of Java software based on combined mean-shift processes, PLoS ONE, 8 (2013).
[14]
J. de la Rosa, J. Villa-Hernandez, E. De la Rosa, E. Gonzalez-Ramirez, O. Gutierrez, N. Escalante, R. Ivanov, G. Fleury, MAP entropy estimation: applications in robust image filtering, J. Eur. Opt. Soc., 8 (2013) 13047.
[15]
L. Devroye, G. Lugosi, Bin width selection in multivariate histograms by the combinatorial method, Test, 13 (2004) 129-145.
[16]
S. Fahmy, A. Mohan, Architecture for real-time nonparametric probability density function estimation, IEEE Trans. Very Large Scale Integr. VLSI Syst., 21 (2013) 910-920.
[17]
S. Fanello, I. Gori, G. Metta, F. Odone, Keep it simple and sparse: real-time action recognition, J. Mach. Learn. Res., 14 (2013) 2617-2640.
[18]
K. Fukunaga, L.D. Hostetler, The estimation of the gradient of a density function, with applications in pattern recognition, IEEE Trans. Inf. Theory, 21 (1975) 32-40.
[19]
M. Geissbuehler, T. Lasser, How to display data by color schemes compatible with red-green color perception deficiencies, Opt. Express, 21 (2013) 9862-9874.
[20]
P. Ghamisi, M. Couceiro, M. Fauvel, J. Benediktsson, Integration of segmentation techniques for classification of hyperspectral images, IEEE Geosci. Remote Sens. Lett., 11 (2014) 342-346.
[21]
A.G. Gray, A.W. Moore, N-body problems in statistical learning, Adv. Neural Inf. Process. Syst., 4 (2000) 521-527.
[22]
L. Greengard, J. Strain, The fast Gauss transform, SIAM J. Sci. Statist. Comput., 12 (1991) 79-94.
[23]
M. Haiman, A simple and relatively efficient triangulation of the n-cube, Discrete Comput. Geomet., 6 (1991) 287-289.
[24]
M. Hanmandlu, O. Verma, S. Susan, V. Madasu, Color segmentation by fuzzy co-clustering of chrominance color features, Neurocomputing, 120 (2013) 235-249.
[25]
T. Hastie, R. Tibshirani, J. Friedman, The Elements of Statistical Learning, Springer, 2009.
[26]
N.-B. Heidenreich, A. Schindler, S. Sperlich, Bandwidth selection for kernel density estimation: a review of fully automatic selectors, AStA Adv. Statist. Anal., 97 (2013) 403-433.
[27]
Hjort, N.L., 1986. On Frequency Polygons and Average Shifted Histograms in Higher Dimensions, Tech. Rep. 22, Stanford University.
[28]
H. Jeffreys, An invariant form for the prior probability in estimation problems, Proc. R. Soc. London, Ser. A, Math. Phys. Sci., 186 (1946) 453-461.
[29]
M. Jones, M. Samiuddin, A. Al-Harbey, T. Maatouk, The edge frequency polygon, Biometrika, 85 (1998) 235-239.
[30]
D.A. Klein, D. Schulz, S. Frintrop, A.B. Cremers, Adaptive real-time video-tracking for arbitrary objects, in: IEEE International Conference on Intelligent Robots and Systems (IROS), October 2010, pp. 772-777.
[31]
J. Klemelä, Multivariate histograms with data-dependent partitions, Statistica Sinica, 19 (2009) 159-176.
[32]
H. Kuhn, Some combinatorial lemmas in topology, IBM J. Res. Dev., 45 (1960) 518-524.
[33]
D. Lang, M. Klaas, F. Hamze, A. Lee, N-Body Methods Code and Matlab Binaries, 2006. <http://www.cs.ubc.ca/%7Eawll/nbody_methods.html>.
[34]
F. Lang, J. Yang, D. Li, L. Shi, J. Wei, Mean-shift-based speckle filtering of polarimetric SAR data, IEEE Trans. Geosci. Remote Sens., 52 (2014) 4440-4454.
[35]
J. Leiva-Murillo, A. Artés-Rodríguez, Algorithms for maximum-likelihood bandwidth selection in kernel density estimators, Pattern Recogn. Lett., 33 (2012) 1717-1724.
[36]
T. Li, S. Sun, T. Sattar, J. Corchado, Fight sample degeneracy and impoverishment in particle filters: a review of intelligent approaches, Expert Syst. Appl., 41 (2014) 3944-3954.
[37]
B. Liu, J. Huang, C. Kulikowski, L. Yang, Robust visual tracking using local sparse appearance model and k-selection, IEEE Trans. Pattern Anal. Mach. Intell., 35 (2013) 2968-2981.
[38]
D. Liu, M. Cong, Y. Du, C. De Silva, GMM-based visual attention for target selection of indoor robotic tasks, Ind. Robot, 40 (2013) 583-596.
[39]
E. López-Rubio, A histogram transform for probability density function estimation, IEEE Trans. Pattern Anal. Mach. Intell., 36 (2014) 644-656.
[40]
E. López-Rubio, A Histogram Transform for Probability Density Function Estimation: Source Code and Demos, 2014. <http://www.lcc.uma.es/%7Eezeqlr/dht/dht.html>.
[41]
F. Mezzadri, How to generate random matrices from the classical compact groups, Notices ACM, 54 (2007) 592-604.
[42]
R. Mittelman, E.L. Miller, Nonlinear filtering using a new proposal distribution and the improved fast Gauss transform with tighter performance bounds, IEEE Trans. Signal Process., 56 (2008) 5746-5757.
[43]
S. Mohammad Saberali, H. Amindavar, J. Ritcey, Blind detection in symmetric non-Gaussian noise with unknown PDF using maximum entropy method with moment generating function constraints, Signal Process., 90 (2010) 891-899.
[44]
A. Moore, The anchors hierarchy: using the triangle inequality to survive high-dimensional data, in: Proceedings of the Twelfth Conference on Uncertainty in Artificial Intelligence, AAAI Press, 2000, pp. 397-405.
[45]
V.I. Morariu, V.C. Raykar, C. Yang, FIGTree Library, 2007. <http://www.umiacs.umd.edu/%7Emorariu/figtree/>.
[46]
G. Moser, J. Zerubia, S. Serpico, SAR amplitude probability density function estimation based on a generalized gaussian model, IEEE Trans. Image Process., 15 (2006) 1429-1442.
[47]
J.A. Nelder, R. Mead, A simplex method for function minimization, Comput. J., 7 (1965) 308-313.
[48]
B. Prakasa Rao, Nonparametric Functional Estimation, Academic Press, 1983.
[49]
M. Rosenblatt, Remarks on some nonparametric estimates of a density function, Ann. Math. Statist., 27 (1956) 832-837.
[50]
J.F. Sallee, A triangulation of the n-cube, Discrete Math., 40 (1982) 81-86.
[51]
J.F. Sallee, The middle-cut triangulations of the n-cube, SIAM J. Algebraic Discrete Methods, 5 (1984) 407-419.
[52]
D. Scott, Frequency polygons: theory and application, J. Am. Statist. Assoc., 80 (1985) 348-354.
[53]
D. Scott, S. Sain, Data mining and computational statistics, Elsevier, Amsterdam, 2005.
[54]
D.W. Scott, Averaged shifted histograms: effective nonparametric density estimators in several dimensions, Ann. Statist., 13 (1985) 1024-1040.
[55]
D.W. Scott, Multivariate Density Estimation, Wiley, 1992.
[56]
B.W. Silverman, Density Estimation for Statistics and Data Analysis, Chapman and Hall, 1986.
[57]
T. Song, C. Gasparovic, N. Andreasen, J. Bockholt, M. Jamshidi, R. Lee, M. Huang, A hybrid tissue segmentation approach for brain MR images, Med. Biol. Eng. Comput., 44 (2006) 242-249.
[58]
M. Sugiyama, M. Kawanabe, P. Chui, Dimensionality reduction for density ratio estimation in high-dimensional spaces, Neural Networks, 23 (2010) 44-59.
[59]
M. Sugiyama, M. Yamada, P. von Bünau, T. Suzuki, T. Kanamori, M. Kawanabe, Direct density-ratio estimation with dimensionality reduction via least-squares hetero-distributional subspace search, Neural Networks, 24 (2011) 183-898.
[60]
B. Wang, L. Wan, T.-D. Zhang, An underwater laser image segmentation algorithm based on pulse coupled neural network and morphology, Adv. Intell. Syst. Comput., 215 (2014) 437-449.
[61]
J. Wang, Y. Yagi, Integrating color and shape-texture features for adaptive real-time object tracking, IEEE Trans. Image Process., 17 (2008) 235-240.
[62]
A. Weber, USC-SIPI Image Database, 2010. <http://sipi.usc.edu/database/>.
[63]
C. Xie, J. Tan, P. Chen, J. Zhang, L. He, Collaborative object tracking model with local sparse representation, J. Vis. Commun. Image Represent., 25 (2014) 423-434.
[64]
B. Xu, X. Shen, F. Ding, A VSS algorithm based on multiple features for object tracking, J. Software, 8 (2013) 3029-3034.
[65]
C. Yang, R. Duraiswami, N.A. Gumerov, L. Davis, Improved fast Gauss transform and efficient kernel density estimation, in: Proceedings of the IEEE International Conference on Computer Vision, 2003, pp. 464-471.
[66]
L.-J. Zhao, T.-Y. Chai, D.-C. Yuan, X.-K. Diao, Probabilistic partial least square based extreme learning machine to enhance reliability of operating conditions recognition, J. Zhejiang Univ. (Eng. Sci.), 47 (2013) 1747-1752.
[67]
Y. Zhong, B. Zhao, L. Zhang, Multiagent object-based classifier for high spatial resolution imagery, IEEE Trans. Geosci. Remote Sens., 52 (2014) 841-847.
[68]
H. Zhou, Y. Yuan, C. Shi, Object tracking using SIFT features and mean shift, Comput. Vis. Image Underst., 113 (2009) 345-352.
- Probability density function estimation with the frequency polygon transform
Recommendations
Probability of default estimation in credit risk using mixture cure models
AbstractAn estimator of the probability of default (PD) in credit risk is proposed. It is derived from a nonparametric conditional survival function estimator based on cure models. Asymptotic expressions for the bias and the variance, as well as the ...
Choice of Kernel Function for Density Estimation
Let l=f^n(x) be the kernel estimate of a density f(x) from a sample of size n. Wahba [6] has developed an upper bound to E[f(x)-l=f^n(x)]2. In the present paper, we find the kernel function of finite support [m=-T, T] that minimizes Wahba's upper bound. ...
Weak conditions for shrinking multivariate nonparametric density estimators
Nonparametric density estimators on R^K may fail to be consistent when the sample size n does not grow fast enough relative to reduction in smoothing. For example a Gaussian kernel estimator with bandwidths proportional to some sequence h"n is not ...
Comments
Please enable JavaScript to view thecomments powered by Disqus.Information & Contributors
Information
Published In
Copyright © Elsevier Inc.
Publisher
Elsevier Science Inc.
United States
Publication History
Published: 20 March 2015
Author Tags
Qualifiers
- Research-article
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- View Citations1Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 28 Dec 2024
Other Metrics
Citations
Cited By
View all- Liu SHu J(2016)Visual saliency based on frequency domain analysis and spatial informationMultimedia Tools and Applications10.1007/s11042-016-3903-375:23(16699-16711)Online publication date: 1-Dec-2016