Research Article  Open Access
Blind Deconvolution for JumpPreserving Curve Estimation
Abstract
In many applications, observed signals are contaminated by both random noise and blur. This paper proposes a blind deconvolution procedure for estimating a regression function with possible jumps preserved, by removing both noise and blur when recovering the signals. Our procedure is based on three local linear kernel estimates of the regression function, constructed from observations in a leftside, a rightside, and a twoside neighborhood of a given point, respectively. The estimated function at the given point is then defined by one of the three estimates with the smallest weighted residual sum of squares. To better remove the noise and blur, this estimate can also be updated iteratively. Performance of this procedure is investigated by both simulation and real data examples, from which it can be seen that our procedure performs well in various cases.
1. Introduction
Nonparametric regression analysis provides us statistical tools for estimating regression functions from noisy data [1]. When the underlying regression function has jumps, the estimated functions by conventional nonparametric regression procedures are not statistically consistent at the jump positions. However, the problem to estimate jump regression functions is important because the true regression functions are often discontinuous in applications [2]. This paper focuses on estimation of the jump regression function when the observed data are contaminated by both random noise and blur.
In the literature, there are some existing procedures for estimating regression curves with jumps preserved in cases when only random noise is present in observed data. These procedures include the onesided kernel estimation methods (e.g., [3–9]), the local leastsquares estimation procedures (e.g., [10–12]), the wavelet transformation method [13], the spline smoothing method [14], and the direct estimation methods (e.g., [15, 16]).
In some applications, our observations are both blurred and contaminated by pointwise noise (e.g., signals of groundwater levels in geothermy). It is, therefore, important to remove both noise and blur when estimating the true regression function. In the nonparametric regression literature, we have not seen any discussion about this problem yet. In the context of image processing, which can be regarded as a twodimensional nonparametric regression problem [2], there is a related research area concerned about deblurring images. Most existing image deblurring procedures assume that the pointspread function (psf), which describes the blurring mechanism, either is known or has a parametric form, and this function is homogeneous in the entire image (e.g., [17–20]). In many applications, however, it is difficult to specify the psf completely or partially by a parametric model. Our major goal in this paper is to provide a method to estimate the true regression function properly in cases when both noise and blur are present and the psf is unspecified.
The remaining part of the paper is organized as follows. In next section, our proposed method is discussed in detail. Some comparative results are presented in Section 3 with several simulated examples. A real data application is presented in Section 4. Some concluding remarks are included in Section 5.
2. Our Proposed Method
Suppose that the regression model concerned is where are design points, are i.i.d. random noise with mean 0 and variance , and is an unknown nonparametric regression function that is continuous in except on some jump points . In (2.1), is a psf, and denotes the convolution between and , defined by
If is not zero at the origin and zero everywhere else, then there is no blurring in the observed curve. In such cases, model (2.1) is the conventional nonparametric regression model. Generally speaking, the problem described by model (2.1) is ill posed if both and are unknown, because infinite sets of and would correspond to the same response in such cases. Therefore, proper estimation of based on (2.1) is a challenging problem. As a demonstration, in Figure 1(a), the solid lines denote a true regression function with one jump point at , the dotted curve denotes the blurred regression function , and the small pluses denote a set of observations from model (2.1).
(a)
(b)
To estimate , by the conventional local linear kernel (LLK) smoothing [1], we would consider the problem where is a density kernel function with support and is a bandwidth parameter. Then the solution of (2.3) to , denoted as , is defined as the LLK estimator of . In Figure 1(b), the dotted curve denotes the blurred regression function, and the dashed curve denotes the LLK estimate of . It can be seen that the LLK estimate removes the noise well; but the blur is not removed and it is actually made more serious around the jump point.
Qiu [15] finds that the major reason why the conventional LLK estimate could not preserve the jump at the jump point is that it uses a “continuous” function (i.e., a linear function) locally to approximate the jump regression function. To overcome this limitation, Qiu suggests fitting a piecewise linear function around as follows: where is an indicator function defined by if and otherwise. The solution to and are denoted as and . It is easy to see that and are actually LLK estimates of constructed from observations in the leftsided neighborhood and the rightsided neighborhood , respectively, with kernel functions and . Then, Qiu suggests the following jumppreserving estimate of : where and are the Weighted Residual Mean Squares(WRMSs)of the leftsided and rightsided estimates, respectively, defined by
Qiu [15] proves that is a consistent estimate of when there is no blurring in the observed data.
Since only part of observations is actually used in , this estimator would be quite noisy in continuity regions of . To overcome this problem, similar to the method in [16], we propose a modification as follows: By (2.7), when is far away from any jump points, would be estimated by the conventional LLK estimate. It would still be estimated by one of the onesided estimates around the jump points. Therefore, it is expected that would be better for estimating than . The solid curve in Figure 2 denotes constructed from the data shown in either plot of Figure 1. The two dotted curves show and , respectively. It can be seen that indeed estimates well in this case.
The estimate defined in (2.7) can also be updated iteratively as follows. The estimated values can be used as observed data, and the estimate can be updated by (2.7) with all quantities on the righthand side of (2.7) computed from , and this process can continue iteratively. Numerical results in the next section suggest that a good estimate can usually be generated after about 5 iterations in all the cases considered there.
In our procedure, the bandwidth parameter should be chosen properly. To this end, we use the following crossvalidation (CV) procedure:
where is the estimate of using all of the data points except the th point .
3. Simulation Study
In this section, some simulated examples are presented concerning the numerical performance of our proposed procedure. In all numerical examples presented in this paper, the Epanechnikov kernel function is used. We consider the following two true regression functions:
The function has two jumps at 0.4, 0.8, and a roof discontinuity (i.e., jump in the slope) at 0.2. The function has a jump at 0.78, and an unbalanced cusp (i.e., a sharp angle) at 0.26. These functions are shown by the solid curves in Figure 3. Our observations are generated from model (2.1) with random noise from the distribution and the psf . One set of observations from each regression function is shown by the little pluses in Figure 3.
(a)
(b)
For the proposed procedure, its MeanSquared Error (MSE) values with several different combinations are presented in Figure 4 as functions of the number of iterations, where the bandwidth is chosen by the CV procedure in each iteration. All of the MSE values are computed based on 100 replications. From the plots in Figure 4, it can be seen that, for each combination, MSE values first decrease and then increase with the iteration number. The optimal number of iteration is around in each case, which is the number that we recommend to use in applications.
(a)
(b)
(c)
(d)
Next, we compare the proposed procedure (denoted as NEW) with the conventional local linear kernel (LLK) smoothing procedure and the jumppreserving curve estimation (JPCE) procedure by Qiu [15]. Figure 5 presents the estimated regression functions by all three methods from the observed data shown in Figure 3. For procedure NEW, 5 iterations are used. From the plots in Figure 5, it can be seen that LLK blurs the jumps, JPCE preserves the jumps well but its estimates are quite noisy in continuity regions, and our proposed procedure NEW preserves the jumps well and also removes noise efficiently.
(a)
(b)
Tables 1 and 2 present the MSE values and the corresponding standard errors of the three methods in various cases. We use 5 iterations in the proposed procedure NEW. From Tables 1 and 2, it can be seen that procedure NEW performs the best in all cases.


4. An Application
In this section, we apply our proposed method to a groundwater level data. Possible jumps in groundwater level arise from changes in subsurface fluid currents, which has become an important predictor of earthquakes. In Figure 6, little pluses denote the groundwater level observed by the Seismograph Network Stations of China Earthquake Center during January and May 2007. The solid curve denotes the estimated regression curve by our proposed procedure in which all parameters are chosen in the same way as in the simulation examples presented in Section 3. As indicated by the plot, the jumps around Mar 23, Aril 10, and Aril 17 are preserved well by our procedure. We checked the earthquake history and it is confirmed by China Earthquake Center that earthquakes with magnitude of more than 4.0 occurred in these periods in the area of observation.
5. Concluding Remarks
We have presented a blind deconvolution procedure for jumppreserving curve estimation when both noise and blur are present in the observed data. Numerical results show that it performs well in various cases. However, theoretical properties of the proposed method are not available yet, which requires much future research. We believe that the proposed method can be generalized to twodimensional cases to solve problems such as image deblurring and restoration.
Acknowledgments
This research was supported in part by an NSF grant. The authors thank the guest editor of this special issue, Professor Ming Li, for help during the paper submission process. They also thank the three anonymous referees for their careful reading of the paper.
References
 J. Fan and I. Gijbels, Local Polynomial Modelling and Its Applications, vol. 66 of Monographs on Statistics and Applied Probability, Chapman & Hall, London, UK, 1996. View at: Zentralblatt MATH  MathSciNet
 P. Qiu, Image Processing and Jump Regression Analysis, Wiley Series in Probability and Statistics, John Wiley & Sons, Hoboken, NJ, USA, 2005. View at: Publisher Site  Zentralblatt MATH  MathSciNet
 P. H. Qiu, C. Asano, and X. P. Li, “Estimation of jump regression function,” Bulletin of Informatics and Cybernetics, vol. 24, no. 34, pp. 197–212, 1991. View at: Google Scholar  Zentralblatt MATH  MathSciNet
 H.G. Müller, “Changepoints in nonparametric regression analysis,” The Annals of Statistics, vol. 20, no. 2, pp. 737–761, 1992. View at: Publisher Site  Google Scholar  Zentralblatt MATH  MathSciNet
 J. S. Wu and C. K. Chu, “Kerneltype estimators of jump points and values of a regression function,” The Annals of Statistics, vol. 21, no. 3, pp. 1545–1566, 1993. View at: Publisher Site  Google Scholar  Zentralblatt MATH  MathSciNet
 P. H. Qiu, “Estimation of the number of jumps of the jump regression functions,” Communications in Statistics—Theory and Methods, vol. 23, no. 8, pp. 2141–2155, 1994. View at: Publisher Site  Google Scholar  MathSciNet
 C. R. Loader, “Change point estimation using nonparametric regression,” The Annals of Statistics, vol. 24, no. 4, pp. 1667–1678, 1996. View at: Publisher Site  Google Scholar  Zentralblatt MATH  MathSciNet
 B. Zhang, Z. Su, and P. Qiu, “On jump detection in regression curves using local polynomial kernel estimation,” Pakistan Journal of Statistics, vol. 25, pp. 505–528, 2009. View at: Google Scholar
 J.H. Joo and P. Qiu, “Jump detection in a regression curve and its derivative,” Technometrics, vol. 51, no. 3, pp. 289–305, 2009. View at: Publisher Site  Google Scholar  MathSciNet
 J. A. McDonald and A. B. Owen, “Smoothing with split linear fits,” Technometrics, vol. 28, no. 3, pp. 195–208, 1986. View at: Publisher Site  Google Scholar  Zentralblatt MATH  MathSciNet
 P. Hall and D. M. Titterington, “Edgepreserving and peakpreserving smoothing,” Technometrics, vol. 34, no. 4, pp. 429–440, 1992. View at: Publisher Site  Google Scholar  MathSciNet
 P. Qiu and B. Yandell, “A local polynomial jumpdetection algorithm in nonparametric regression,” Technometrics, vol. 40, no. 2, pp. 141–152, 1998. View at: Publisher Site  Google Scholar  MathSciNet
 Y. Wang, “Jump and sharp cusp detection by wavelets,” Biometrika, vol. 82, no. 2, pp. 385–397, 1995. View at: Publisher Site  Google Scholar  Zentralblatt MATH  MathSciNet
 J.Y. Koo, “Spline estimation of discontinuous regression functions,” Journal of Computational and Graphical Statistics, vol. 6, no. 3, pp. 266–284, 1997. View at: Publisher Site  Google Scholar  MathSciNet
 P. Qiu, “A jumppreserving curve fitting procedure based on local piecewiselinear kernel estimation,” Journal of Nonparametric Statistics, vol. 15, no. 45, pp. 437–453, 2003. View at: Publisher Site  Google Scholar  Zentralblatt MATH  MathSciNet
 I. Gijbels, A. Lambert, and P. Qiu, “Jumppreserving regression and smoothing using local linear fitting: a compromise,” Annals of the Institute of Statistical Mathematics, vol. 59, no. 2, pp. 235–272, 2007. View at: Publisher Site  Google Scholar  MathSciNet
 A. S. Carasso, “Direct blind deconvolution,” SIAM Journal on Applied Mathematics, vol. 61, no. 6, pp. 1980–2007, 2001. View at: Publisher Site  Google Scholar  Zentralblatt MATH  MathSciNet
 P. Hall and P. Qiu, “Blind deconvolution and deblurring in image analysis,” Statistica Sinica, vol. 17, no. 4, pp. 1483–1509, 2007. View at: Google Scholar  Zentralblatt MATH  MathSciNet
 P. Hall and P. Qiu, “Nonparametric estimation of a pointspread function in multivariate problems,” The Annals of Statistics, vol. 35, no. 4, pp. 1512–1534, 2007. View at: Publisher Site  Google Scholar  MathSciNet
 P. Qiu, “A nonparametric procedure for blind image deblurring,” Computational Statistics & Data Analysis, vol. 52, no. 10, pp. 4828–4841, 2008. View at: Publisher Site  Google Scholar  MathSciNet
Copyright
Copyright © 2010 Xingfang Huang and Peihua Qiu. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.