SAR image time-series analysis framework using morphological operators and global and local information-based linear discriminant analysis

Fusion of spectral, spatial, and temporal information is an effective method used in many satellite remotesensing applications. On the other hand, one drawback of this fusion is an increase in complexity. In this paper, wefocus on developing a fast and well-performed classification method for agricultural crops using time-series SAR data.In order to achieve this, a novel two-stage approach is proposed. In the first stage, a high-dimensional feature space isobtained using time-series dual-pol SAR data and morphological operators. Spectral, spatial, and temporal informationis combined into a single high-dimensional feature space. In the second stage, a dimension reduction technique is appliedto the feature vector in order to decrease time complexity and increase classification accuracy by considering the globaland local pattern information in the high-dimensional feature space. The contribution of the morphological profilesto the classification performance is significant; however the time complexity is increased drastically. The proposedmethod overcomes the time complexity stemming from high-dimensional feature space; it also improves the classificationperformance. The superiority of the proposed method to the comparative methods in agricultural crop classification isexperimentally shown with the improvements in both classification and time performance using time-series TerraSAR-Ximages.

PDF

___

Skriver H, Mattia F, Satalino G, Balenzano A, Pauwels VRN, Verhoest NEC, Davidson M. Crop classification using short-revisit multitemporal SAR data. IEEE J Sel Top Appl Earth Obs Remote Sens 2011; 4: 423-431.
McNairn H, Shang J, Champagne C, Jiao X. TerraSAR-X and RADARSAT-2 for crop classification and acreage estimation. In: Proceedings of IGARSS; 12–17 July 2009; Cape Town, Africa. pp. 898-901.
Mirzaee S, Motagh M, Arefi H, Nooryazdan M. Classification of agricultural fields using time series of dual polarimetry TerraSAR-X images. Int Arch Photogramm Remote Sens Spat Inf Sci 2014; 40: 191.
Blaes X, Vanhalle L, Defourny P. Efficiency of crop identification based on optical and SAR image time series. Remote Sens Environ 2005; 96: 352-365.
Lardeux C, Frison PL, Tison C, Souyris JC, Stoll B, Fruneau B, Rudant JP. Support vector machine for multifrequency SAR polarimetric data classification. IEEE T Geosci Remote Sens 2009; 47: 4143-4152.
Tan CP, Koay JY, Lim KS, Ewe HT, Chuah HT. Classification of multi-temporal SAR images for rice crops using combined entropy decomposition and support vector machine technique. Prog Electromagn Res 2007; 71: 19-39.
Fukuda S, Hirosawa H. A wavelet-based texture feature set applied to classification of multifrequency polarimetric SAR images. IEEE T Geosci Remote Sens 1999; 37: 2282-2286.
Demirkesen C, Demirpolat C, Teke M. Tarımsal alanların çok zamanlıTerraSAR-X görüntülerinde morfolojik profiller kullanılarak sınıflandırılması. In: IEEE 24. Sinyal İşleme, İletişim ve UygulamalarıKurultayı; 16–19 May 2016; Zonguldak, Turkey. pp. 1701-1704 (in Turkish).
Demirpolat C, Teke M. Crop classification with morphological profiles derived from SAR and electro-optical satellite data. In: 2017 8th International Conference on Recent Advances in Space Technologies; 19–22 July 2017; İstanbul, Turkey. pp. 119-124.
He C, Han G, Feng D, Du J, Liao M. A sparse manifold classification method based on a multi-dimensional descriptive primitive of polarimetric SAR image time series. ISPRS Int J Geo-Information 2017; 6: 97.
Duda RO, Hart PE, Stork DG. Pattern Classification. 2nd ed. New York, NY, USA: Wiley, 2001.
Zhang D, He J, Zhao Y, Luo Z, Du M. Global plus local: a complete framework for feature extraction and recognition. Pattern Recognit 2014; 47: 1433-1442.
Sakarya U. Hyperspectral dimension reduction using global and local information based linear discriminant analysis. ISPRS Ann Photogramm Remote Sens Spat Inf 2014; 2: 61.
Sakarya U, Demirpolat C. SAR image time series analysis in vegetation classification using global and local information based linear discriminant analysis. In: Signal Processing and Communications Applications Conference; 15–18 May 2017; Antalya, Turkey. pp. 1-4.
Sakarya U. Thermal infrared hyperspectral dimension reduction experiment results for global and local information based linear discriminant analysis. In: Signal Processing and Communications Applications Conference; 16–19 May 2015; Malatya, Turkey. pp. 268-271.
Dalla Mura M, Benediktsson JA, Waske B, Bruzzone L. Morphological attribute filters for the analysis of very high resolution remote sensing images. In: Proceedings of IGARSS; 12–17 July 2009; Cape Town, Africa. p. III-97.
European Space Agency. Sentinel Application Platform [Online]. http://step.esa.int/main/third-party-plugins2/sen2cor.
Yu Y, Acton ST. Speckle reducing anisotropic diffusion. IEEE T Image Process 2002; 11: 1260-1270.
Foody GM. Status of land cover classification accuracy assessment. Remote Sens Environ 2002; 80: 185-201.
R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing.
Hijmans RJ. Raster: Geographic Data Analysis and Modeling. R Package Version 2.3-24 [Online]. http://CRAN.Rproject.org/package=raster.
Beygelzimer A, Kakadet S, Langford J, Arya S, Mount D, Li S. FNN: Fast Nearest Neighbor Search Algorithms and Applications. R package version 1.1 [Online]. http://CRAN.R-project.org/package=FNN.
Venables WN, Ripley BD. Modern Applied Statistics with S-PLUS. 3rd ed. New York, NY, USA: Springer, 2013.
Sakarya U, Telatar Z, Alatan AA. Dominant sets based movie scene detection. Signal Process 2012; 92: 107-119