(50-3) 11 * << * >> * Russian * English * Content * All Issues

Structural analysis of spectral remote sensing data based on the use of correlation coefficient components
A.V. Lapko1,2, V.A. Lapko1,2, Yu.P. Yuronen1

1 Reshetnev Siberian State University of Science and Technology, Prospekt Krasnoyarsky Rabochy 31, Krasnoyarsk, 660037, Russia;
2 Institute of Computational Modelling SB RAS, Akademgorodok 50, Krasnoyarsk, 660036, Russia

  Full text (PDF)

DOI: 10.18287/COJ1746

Article ID: 1746

Language: English

Abstract:
The article describes a method for structural analysis of remote sensing data based on the decomposition of the range of values of spectral features and the development of an algorithm for assessing the membership of control situations in the detected classes. To decompose the range of values of spectral features, the components of the correlation coefficient are used, which are determined by the product of the normalized values of a pair of spectral features. Based on the signs of the components of the correlation coefficient, four classes are determined, which correspond to the normalized values of spectral features: positive, negative, and alternating. The discovered classes are characterized by the dependence between spectral features. On this basis, using the initial statistical data, a decision rule is formed for assessing the belonging of the spectral features of the control situation to one of the four classes. Estimates of the probability densities of the correlation coefficient components in the classes are presented and analyzed. Methods for optimizing nonparametric estimates of the probability densities are considered. The effectiveness of the proposed method is confirmed by the results of decomposition of remote sensing data of a forest area according to two spectral features (Red, Near Infrared or NIR). Under these conditions, breaking down the forest area into deciduous, coniferous, damaged stands and areas devoid of a tree cover is justified. Nonparametric estimates of the probability densities of the correlation coefficient components in the detected classes are analyzed.

Keywords:
structural data analysis, automatic classification, components of the correlation coefficient, kernel probability density estimation, remote sensing data, spectral features, forest area, NDVI index.

Acknowledgements:

Citation:
Lapko AV, Lapko VA, Yuronen YuP. Structural analysis of spectral remote sensing data based on the use of correlation coefficient components. Computer Optics 2026; 50(3): 1746. DOI: 10.18287/COJ1746.

References:

  1. Borzov SM, Potaturkin OI. Selection of the informative feature system for crops classification using hyperspectral data [In Russian]. Avtometriya 2020; 56(4): 134-44. DOI: 10.18287/2412-6179-CO-779
  2. Shipko VV, Borzov SM. Analysis of the efficiency of hyperspectral data classification under constraints on the quantization bit depth, the number of spectral channels, and spatial resolution [In Russian]. Avtometriya 2022; 58(3): 79-87. DOI: 10.15372/AUT20220309
  3. Tuboltsev VP, Lapko AV, Lapko VA. Modified nonparametric algorithm for automatic classification of large-volume statistical data and its application. Scientific and Technical Information Processing 2024; 51(6): 592-7. DOI: 10.3103/S0147688224700552
  4. Degermendzhi AG, Vysotskaya GS, Somova LA, Pisman TI, Shevyrnogov AP. Long-term NDVI dynamics of vegetation in Tundra of different classes depending on temperature and precipitation [In Russian]. Doklady Earth Sciences 2020; 493(2): 103-6. DOI: 10.31857/S2686739720080046
  5. Shevyrnogov AP, Botvich IYu, Pisman TI, Volkova AI, Kononova NA, Ivanov SA. Informative value of spectral vegetation indices for the meadow and steppe vegetation monitoring of Khakassia by ground and satellite data [In Russian]. Izvestiya, Atmospheric and Oceanic Physics 2024; (1): 16-28. DOI: 10.31857/S0205961424010028
  6. Sahu G, Garanayak M, Paikaray BK. Green vegetation detection in satellite imagery using the normalised difference vegetation index method. International Journal of Bioinformatics Research and Applications 2023; 19(4): 327-42. DOI: 10.1504/ijbra.2023.135367
  7. Basit I, Faizi F, Mahmood Kh, Faizi R, Ramzan S, Parvez Sh, Mushtaq F. Assessment of vegetation dynamics under changed climate situation using geostatistical modeling. Theoretical and Applied Climatology 2024; 155(4): 3371-3386. DOI: 10.1007/s00704-024-04840-x
  8. Sazonov DS. Correlation analysis of experimental remote sensing data and models of microwave rough sea surface emission [In Russian]. Izvestiya, Atmospheric and Oceanic Physics 2017; (3): 53-64. DOI: 10.7868/S020596141703006X
  9. Lu Z, Mingsheng L, Limin Y, Hui L. Remote sensing change detection based on canonical correlation analysis and contextual bayes decision. Photogrammetric Engineering & Remote Sensing 2007; 73(3): 311-8. DOI:10.14358/PERS.73.3.311
  10. Im J, Jensen JR, Tullis JA. Object-based change detection using correlation image analysis and image segmentation. International Journal of Remote Sensing 2008; 29(2):399-423. DOI: 10.1080/01431160601075582
  11. Lapko AV, Lapko VA. Procedure for decomposing the values of two-dimensional spectral features of remote sensing based on the analysis of correlation coefficient components. Measurement Techniques 2024; 67(6): 427-432. DOI: 10.1007/s11018-024-02362-6
  12. Lapko AV, Lapko VA, Im ST. Methodology for decomposition of spectral data from remote sensing of forest fires [In Russian]. Informatika i sistemy upravleniya 2024; 81(3): 112-20. DOI: 10.22250/18142400_2024_81_3_112
  13. Parzen E. On estimation of a probability density function and mode. Ann. Math. Statistic 1962; 33(3): 1065-1076. DOI: 10.1214/aoms/1177704472.
  14. Epanechnikov VA. Non-parametric estimation of a multivariate probability density. Theory of Probability & Its Applications 1969; 14(1): 156-161. DOI: 10.1137/1114019.
  15. Rudemo M. Empirical choice of histogram and kernel density estimators. Scandinavian Journal of Statistics 1982; 9(2): 65-78. JSTOR: 4615859
  16. Lapko AV, Lapko VA. Analysis of optimization methods for nonparametric estimation of the probability density with respect to the blur factor of kernel functions. Measurement Techniques 2017; 60(6): 515-522. DOI: 10.1007/s11018-017-1228-x

151, Molodogvardeiskaya str., Samara, 443001, Russia; E-mail: journal@computeroptics.ru; Tel: +7 (846) 242-41-24 (Executive secretary), +7 (846) 332-56-22 (Issuing editor), Fax: +7 (846) 332-56-20