Journals / HKM Jeodezi Jeoinformasyon Arazi Yönetimi Dergisi / 2007 / Cilt: 0 - Sayı: 97

Least squares and total least squares methods in coordinate transformation

Koordinat dönüşümünde en küçük kareler ve toplam en küçük kareler yöntemleri

Pages
15–22
Publication Date
2022-10-18
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

In this study, coordinate transformation is conducted through both LS and TLS approaches using the same data set.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:In coordinate transformation, point coordinates in one coordinate system can be obtained in another using common points whose coordinates are known in both systems.
📊
Key Findings & Data:In the case of having more common points than required, this process is done by the least squares (LS) method.
💡
Conclusion & Impact:In this study, coordinate transformation is conducted through both LS and TLS approaches using the same data set.
🔬Design:Research Article
📍Scope:Comprehensive Sample
⚙️Variables:Domain Parameters

Abstract

In coordinate transformation, point coordinates in one coordinate system can be obtained in another using common points whose coordinates are known in both systems. In the case of having more common points than required, this process is done by the least squares (LS) method. In coordinate transformation, both observation vector and some elements of the design matrix have some stochastic properties. In classical LS approach this is often neglected and remains as an uncertainty in the results. Using "total least squares (TLS) ", a considerably new method, both observations and the whole or part of design matrix can be taken as stochastic components. Thus, more realistic values could be estimated for unknown parameters. In this study, coordinate transformation is conducted through both LS and TLS approaches using the same data set.

Özet

Koordinat Dönüşümünde iki koordinat sistemi arasındaki transformasyon parametreleri her iki sistemde de koordinatları bilinen ortak noktalardan hesaplanır. Gereğinden fazla ortak nokta bulunması durumunda bu işlem "En Küçük Kareler (EKK) "yöntemi ile gerçekleştirilir. Koordinat dönüşümünde, hem gözlem vektörü hem de katsayılar matrisinin bazı elemanları stokastîk özellikler taşır. Klasik EKK yaklaşımında bu genellikle göz ardı edilir ve bu durum çözüm sonuçları içinde bir belirsizlik olarak kalır. Oldukça yeni bir yöntem olan "Toplam En Küçük Kareler (TEKK) "yöntemi ile hem gözlemler hem de katsayılar matrisinin tamamı ya da bir parçası stokastik bileşen olarak alınabilir ve böylece bilinmeyen parametreler için daha gerçekçi değerler kestirilebilir. Bu çalışmada klasik EKK ile TEKK yöntemleriyle aynı veri kümesinde koordinat dönüşümü gerçekleştirilmiştir.

🔖

Cite This Article

AKYILMAZ, O., ÖZLÜDEMİR, M. T., ACAR, M. (2022). Least squares and total least squares methods in coordinate transformation. HKM Jeodezi Jeoinformasyon Arazi Yönetimi Dergisi, Cilt: 0 - Sayı: 97, 15–22. https://ajindex.com/articles/538097

References (5)

  1. ACAR M., ÖZLÜDEMİR M.T., AKYILMAZ O., ÇELİK R.N. ve AYAN T.: Deformation analysis with Total Least Squares, Natural Hazards and Earth System Sciences, n.6 (2006), p.663-670. AKYILMAZ O.: Total Least Squares solution of coordinate transformation, Survey Review, v.39, n.303 (2007), p.68-80. AYAN T.: Matematik istatistik ve hipotez testleri, Lisansüstü ders notlan, İstanbul Teknik Üniversitesi, 1981. AYAN T.: Dengeleme hesabı, Lisans ders notlan, İstanbul Teknik Üniversitesi,,2001. FELUS Y: Application of Total Least Squares for spatial point process analysis, Journal of Surveying Engineering, v. 130, n. 3 (2004), p. 126- 133. GOLUB H.G. ve LOAN F.C.: An analysis of the Total Least Squares problem, SLAM Journal of Numerical Analysis, v. 17, n. 6 (1980), p.883- 893. KOCHK.R.: Parameter estimation and hypothesis testing in linear models, 2nd Ed., Springer, ISBN-13: 978-3540652571, Berlin, 1999. LEON-S.: Linear algebra with applications, 6th Ed., Prentice Hall, ISBN- 13: 978-0130337832,
  2. New Jersey, ABD, 2002
  3. MARKOVSKY I., VAN HUFFEL S. ve KUKUSH A.: On the computation of the multivariate structured Total Least Squares estimator, Numerical Linear Algebra with Applications, n. 11 (2004), p. 591-608. VAN HUFFEL S.: The generalized Total Least Squares problem: formulation, algorithm and properties, In: Numerical Linear Algebra, Digital Signal Processing and Parallel Algorithms, G. H. Golub and P. V. Dooren (Eds.), NATO ASI Series F No. 70, Springer, ISBN-13: 978- 0387523002, Berlin, 1991. VAN HUFFEL S. ve VANDEWALLE J.: The Total Least Squares problem: computational aspects and analysis. Society for Industrial and Applied Mathematics (SLAM), ISBN-13:978-0-898712-75-9,
  4. Philadelphia, ABD, 1991. WOLF R.P. ve GHILANI D.C.: Adjustment computations: statistics and least squares in surveying and GIS, Wiley, ISBN-13:978-0471168331,
  5. New York, ABD, 1997.