Unsupervised Classification of Crop Growth Stages with Scattering Parameters from Dual-Pol Sentinel-1 SAR Data
Global crop mapping and monitoring requires high-resolution spatio-temporal information. In this regard, dual polarimetric Synthetic Aperture Radar (SAR) sensors provide high temporal and high spatial resolutions with large swath width. Generally, crop phenological development studies utilized SAR b...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-11-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/13/21/4412 |
| _version_ | 1797511862320889856 |
|---|---|
| author | Subhadip Dey Narayanarao Bhogapurapu Saeid Homayouni Avik Bhattacharya Heather McNairn |
| author_facet | Subhadip Dey Narayanarao Bhogapurapu Saeid Homayouni Avik Bhattacharya Heather McNairn |
| author_sort | Subhadip Dey |
| collection | DOAJ |
| description | Global crop mapping and monitoring requires high-resolution spatio-temporal information. In this regard, dual polarimetric Synthetic Aperture Radar (SAR) sensors provide high temporal and high spatial resolutions with large swath width. Generally, crop phenological development studies utilized SAR backscatter intensity-based descriptors. However, these descriptors are derived either from the covariance matrix elements or from the eigendecomposition. Therefore, this approach fails to utilize the complete polarization information of the scattered wave. In this study, we propose a target characterization parameter, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>θ</mi><mi>xP</mi></msub></semantics></math></inline-formula> that utilizes the 2D Barakat degree of polarization and the elements of the covariance matrix. We also propose an unsupervised clustering scheme using <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>θ</mi><mi>xP</mi></msub></semantics></math></inline-formula> and the scattering entropy, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>H</mi><mi>xP</mi></msub></semantics></math></inline-formula>. We utilize time-series Sentinel-1 data of canola and wheat fields over a Canadian test site to show the sensitivity of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>θ</mi><mi>xP</mi></msub></semantics></math></inline-formula> to the development of crop morphology at different phenological stages. During the initial growth stages, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>θ</mi><mi>xP</mi></msub></semantics></math></inline-formula> values are low due to the low vegetation density. In contrast, at advanced phenological stages, we observe decreased values of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>θ</mi><mi>xP</mi></msub></semantics></math></inline-formula> due to the appearance of complex canopy structure. Similarly, the effectiveness of the unsupervised <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msub><mi>H</mi><mi>xP</mi></msub><mo>/</mo><msub><mi>θ</mi><mi>xP</mi></msub></mrow></semantics></math></inline-formula> clustering plane is also evident from the temporal clustering plots. This innovative clustering framework is beneficial for the operational use of Sentinel-1 SAR data for agricultural applications. |
| first_indexed | 2024-03-10T05:53:57Z |
| format | Article |
| id | doaj.art-7cca86ff896b430194d5ba681747f379 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T05:53:57Z |
| publishDate | 2021-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-7cca86ff896b430194d5ba681747f3792023-11-22T21:33:05ZengMDPI AGRemote Sensing2072-42922021-11-011321441210.3390/rs13214412Unsupervised Classification of Crop Growth Stages with Scattering Parameters from Dual-Pol Sentinel-1 SAR DataSubhadip Dey0Narayanarao Bhogapurapu1Saeid Homayouni2Avik Bhattacharya3Heather McNairn4Microwave Remote Sensing Lab, Centre of Studies in Resources Engineering, Indian Institute of Technology Bombay, Mumbai 400076, IndiaMicrowave Remote Sensing Lab, Centre of Studies in Resources Engineering, Indian Institute of Technology Bombay, Mumbai 400076, IndiaCentre Eau Terre Environnement, INRS, 490 Couronne St, Quebec City, QC G1K 9A9, CanadaMicrowave Remote Sensing Lab, Centre of Studies in Resources Engineering, Indian Institute of Technology Bombay, Mumbai 400076, IndiaOttawa Research and Development Centre, Agriculture and Agri-Food Canada, 1341 Baseline Road, Ottawa, ON K1A 0C5, CanadaGlobal crop mapping and monitoring requires high-resolution spatio-temporal information. In this regard, dual polarimetric Synthetic Aperture Radar (SAR) sensors provide high temporal and high spatial resolutions with large swath width. Generally, crop phenological development studies utilized SAR backscatter intensity-based descriptors. However, these descriptors are derived either from the covariance matrix elements or from the eigendecomposition. Therefore, this approach fails to utilize the complete polarization information of the scattered wave. In this study, we propose a target characterization parameter, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>θ</mi><mi>xP</mi></msub></semantics></math></inline-formula> that utilizes the 2D Barakat degree of polarization and the elements of the covariance matrix. We also propose an unsupervised clustering scheme using <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>θ</mi><mi>xP</mi></msub></semantics></math></inline-formula> and the scattering entropy, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>H</mi><mi>xP</mi></msub></semantics></math></inline-formula>. We utilize time-series Sentinel-1 data of canola and wheat fields over a Canadian test site to show the sensitivity of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>θ</mi><mi>xP</mi></msub></semantics></math></inline-formula> to the development of crop morphology at different phenological stages. During the initial growth stages, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>θ</mi><mi>xP</mi></msub></semantics></math></inline-formula> values are low due to the low vegetation density. In contrast, at advanced phenological stages, we observe decreased values of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>θ</mi><mi>xP</mi></msub></semantics></math></inline-formula> due to the appearance of complex canopy structure. Similarly, the effectiveness of the unsupervised <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msub><mi>H</mi><mi>xP</mi></msub><mo>/</mo><msub><mi>θ</mi><mi>xP</mi></msub></mrow></semantics></math></inline-formula> clustering plane is also evident from the temporal clustering plots. This innovative clustering framework is beneficial for the operational use of Sentinel-1 SAR data for agricultural applications.https://www.mdpi.com/2072-4292/13/21/4412Sentinel-1polarimetrydual-polcrop characterizationphenologyunsupervised classification |
| spellingShingle | Subhadip Dey Narayanarao Bhogapurapu Saeid Homayouni Avik Bhattacharya Heather McNairn Unsupervised Classification of Crop Growth Stages with Scattering Parameters from Dual-Pol Sentinel-1 SAR Data Remote Sensing Sentinel-1 polarimetry dual-pol crop characterization phenology unsupervised classification |
| title | Unsupervised Classification of Crop Growth Stages with Scattering Parameters from Dual-Pol Sentinel-1 SAR Data |
| title_full | Unsupervised Classification of Crop Growth Stages with Scattering Parameters from Dual-Pol Sentinel-1 SAR Data |
| title_fullStr | Unsupervised Classification of Crop Growth Stages with Scattering Parameters from Dual-Pol Sentinel-1 SAR Data |
| title_full_unstemmed | Unsupervised Classification of Crop Growth Stages with Scattering Parameters from Dual-Pol Sentinel-1 SAR Data |
| title_short | Unsupervised Classification of Crop Growth Stages with Scattering Parameters from Dual-Pol Sentinel-1 SAR Data |
| title_sort | unsupervised classification of crop growth stages with scattering parameters from dual pol sentinel 1 sar data |
| topic | Sentinel-1 polarimetry dual-pol crop characterization phenology unsupervised classification |
| url | https://www.mdpi.com/2072-4292/13/21/4412 |
| work_keys_str_mv | AT subhadipdey unsupervisedclassificationofcropgrowthstageswithscatteringparametersfromdualpolsentinel1sardata AT narayanaraobhogapurapu unsupervisedclassificationofcropgrowthstageswithscatteringparametersfromdualpolsentinel1sardata AT saeidhomayouni unsupervisedclassificationofcropgrowthstageswithscatteringparametersfromdualpolsentinel1sardata AT avikbhattacharya unsupervisedclassificationofcropgrowthstageswithscatteringparametersfromdualpolsentinel1sardata AT heathermcnairn unsupervisedclassificationofcropgrowthstageswithscatteringparametersfromdualpolsentinel1sardata |