Abstract
Thin plate spline (TPS) has been widely accepted as a method for smooth fitting of noisy data. However, the classical TPS always has an ill-conditioning problem when two sample points are very close. Although the modified orthogonal least squares-based TPS (TPS-M) avoids this ill-conditioning problem, it is not completely immune to over-fitting when sample points are noisy. In this paper, a regularized least squares method of thin plate spline (TPS-RLS) was developed, which adds a weight penalty term to the error criterion of orthogonal least squares (OLS). TPS-RLS combines the advantages of both regularization and OLS, which avoid the over-fitting and the ill-conditioning problems simultaneously. Numerical tests indicate that irrespective of the standard deviation of sampling errors and the number of knots, TPS-RLS is always more accurate than TPS-M for smooth fitting of noisy data, whereas TPS-M would have a serious over-fitting problem if the optimal number of knots were not determined in advance. The real-world example of fitting total station instrument data shows that among the classical interpolation methods including IDW, natural neighbor and ordinary kriging, TPS-RLS has the highest accuracy for a series of DEMs with different resolutions, especially for the coarse one. Surface modeling of DEMs with contour lines demonstrate that TPS-RLS has a better performance than the classical methods in terms of both root mean squared error and relief shaded map appearance.
Similar content being viewed by others
References
Bates DM, Lindstrom MJ, Wahba G, Yandell BS (1987) GCVPACK-routines for generalized cross validation. Commun Stat Simul Comput 16(1):263–297
Billings SD, Newsam GN, Beatson RK (2002) Smooth fitting of geophysical data using continuous global surfaces. Geophysics 67(6):1823–1834
Bjorck A (1996) Numerical methods for least squares problems. SIAM, Philadelphia
Chen CF, Li YY (2012) A robust method of thin plate spline and its application to DEM construction. Comput Geosci 48:9–16
Ermini L, Catani F, Casagli N (2005) Artificial Neural Networks applied to landslide susceptibility assessment. Geomorphology 66(1–4):327–343
Fisher P (1991) First experiments in viewshed uncertainty: the accuracy of the viewshed area. Photogramm Eng Remote Sens 57(10):1321–1327
Fisher PF (1998) Improved modeling of elevation error with geostatistics. GeoInformatica 2(3):215–233
Fisher PF, Tate NJ (2006) Causes and consequences of error in digital elevation models. Prog Phys Geogr 30(4):467–489
Garcia D (2010) Robust smoothing of gridded data in one and higher dimensions with missing values. Comput Stat Data Anal 54(4):1167–1178
Goodchild MF, Yuan M, Cova TJ (2007) Towards a general theory of geographic representation in GIS. Int J Geogr Inf Sci 21(3):239–260
Gousie MB, Franklin WR (2005) Augmenting grid-based contours to improve thin plate DEM generation. Photogramm Eng Remote Sens 71(1):69–79
Hoerl AE, Kennard RW (1970) Ridge regression: applications to nonorthogonal problems. Technometrics 12(1):69–82
Hofierka J, Cebecauer T (2007) Spatial interpolation of elevation data with variable density: a new methodology to derive quality DEMs. Geosci Remote Sens Lett IEEE 4(1):117–121
Hunter GJ, Goodchild MF (1997) Modeling the uncertainty of slope and aspect estimates derived from spatial databases. Geogr Anal 29(1):35–49
Kawabata D, Bandibas J (2009) Landslide susceptibility mapping using geological data, a DEM from ASTER images and an Artificial Neural Network (ANN). Geomorphology 113(1–2):97–109
Ledoux H, Gold C (2005) An efficient natural neighbour interpolation algorithm for geoscientific modelling. Developments in spatial data handling. Springer, Berlin, pp 97–108
Longley PA, Goodchild MF, Maguire DJ, Rhind DW (2011) Geographic information systems and science. Wiley, New York
MacKay DJ (1992) Bayesian interpolation. Neural Comput 4(3):415–447
McMahon JR, Franke R (1992) Knot selection for least squares thin plate splines. SIAM J Sci Stat Comput 13:484–498
Melchiorre C, Matteucci M, Azzoni A, Zanchi A (2008) Artificial neural networks and cluster analysis in landslide susceptibility zonation. Geomorphology 94(3–4):379–400
Moore ID, Grayson RB, Ladson AR (1991) Digital terrain modelling: a review of hydrological, geomorphological, and biological applications. Hydrol Process 5(1):3–30
Murphy PNC, Ogilvie J, Meng F-R, Arp P (2008) Stream network modelling using lidar and photogrammetric digital elevation models: a comparison and field verification. Hydrol Process 22(12):1747–1754
Orr MJL (1995) Regularization in the selection of radial basis function centers. Neural Comput 7(3): 606–623
Van Loan C (1985) On the method of weighting for equality-constrained least-squares problems. SIAM J Numer Anal 22(5):851–864
Van Niel TG, McVicar TR, Li L, Gallant JC, Yang Q (2008) The impact of misregistration on SRTM and DEM image differences. Remote Sens Environ 112(5):2430–2442
Wilson JP (2012) Digital terrain modeling. Geomorphology 137(1):107–121
Yao J, Murray AT (2013) Continuous surface representation and approximation: spatial analytical implications. Int J Geogr Inf Sci 27(5):883–897
Yue TX, Fan ZM, Chen CF, Sun XF, Li BL (2011) Surface modelling of global terrestrial ecosystems under three climate change scenarios. Ecol Model 222:2342–2361
Zandbergen PA (2011) Characterizing the error distribution of lidar elevation data for North Carolina. Int J Remote Sens 32(2):409–430
Acknowledgments
Special thanks go to the two anonymous reviewers for their assistance, comments and suggestions. This work was supported by the National Natural Science Foundation of China (Grant No. 41101433, 41371367), Young and Middle-Aged Scientists Research Awards Fund of Shandong Province (Grant No. BS2012HZ010), Qingdao Science and Technology Program of Basic Research Project (Grant No. 13-1-4-239-jch), the Key Laboratory of Marine Surveying and Mapping in Universities of Shandong (Shandong University of Science and Technology) (Grant No. 2013B03), and Special Project Fund of Taishan Scholars of Shandong Province.
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix A: Regularization Parameter Deduction Based on Bayesian Interpretation
Based on Bayesian interpretation, the error criterion Eq. (11) can be transformed into the form
where \(\alpha \) represents a regularizing constant; \(\beta =1/\sigma _\mathrm{e}^{2};\rho =\alpha /\beta ;\sigma _\mathrm{e}\) is the standard deviation of sampling errors. Based on Eq. (4.6) in the paper of MacKay (1992), the log evidence for \(\alpha \) and \(\beta \) can be expressed as
where \(\mathbf{A}\) is the second derivative of Eq. (22) with respect to \(\theta \), namely
Setting the first derivatives of Eq. (23) with respect to \(\alpha \) and \(\beta \) to zeros, we can respectively obtain the following equations
Hence
or
Letting \(\gamma =\sum _{i=1}^M {\frac{\mathbf{q}_i^\mathrm{T} \mathbf{q}_i}{\rho +\mathbf{q}_i^\mathrm{T}\mathbf{q}_i}}\), Eq. (28) can be transformed into the expression
Namely
Appendix B: Procedure of TPS-RLS
Repeat the above procedure until \(\rho \) converges to a fixed value.
Rights and permissions
About this article
Cite this article
Chen, C., Li, Y., Cao, X. et al. Smooth Surface Modeling of DEMs Based on a Regularized Least Squares Method of Thin Plate Spline. Math Geosci 46, 909–929 (2014). https://doi.org/10.1007/s11004-013-9519-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11004-013-9519-5