Guided Nonlocal Means Estimation of Polarimetric Covariance for Canopy State Classification

Agersborg, Jørgen A.; Anfinsen, Stian Normann; Jepsen, Jane Uhd

doi:10.1109/tgrs.2021.3090831

Cited by 5 publications

(7 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The threshold parameter hα in the similarity test processing is chosen to control the probability of the null hypothesis. ln H ≥ h α → Similar patches, H 0 holds (10) In our experiments, the threshold parameter hα is assigned as h α = 2/ √ l × patch size. The corresponding weight function can then be ob-tained based on GLRT:…”

Section: Patch-based Nonlocal Estimationmentioning

confidence: 99%

See 1 more Smart Citation

Patch‐based nonlocal estimation of scattering mechanism vector in PolInSAR

Xing

Wang

et al. 2023

Electronics Letters

View full text Add to dashboard Cite

Nowadays, polarimetric interferometric synthetic aperture radar (PolIn-SAR) has attracted increasing attention for the simultaneous acquisition of scattering and terrain information. The scattering mechanism vectors corresponding to the optimal coherences have shown great effect in land cover classification and forest mapping. To extract scattering mechanism vectors, different optimization algorithms are proposed but always accompanied with spatial discontinuities. This letter proposes a method to optimize scattering mechanism vectors based on nonlocal estimations. A pixel-by-pixel optimization model is established by calculating patch-based nonlocal similarity, which effectively reduces the uncertainty of scattering mechanism vector estimation while maintaining high resolution. The phase recovery experiment is conducted on simulated data and the root mean square errors (RMSEs) of three algorithms are compared. The proposed method performs better with a RMSE of 0.22 in the whole image, against 0.25 and 1.27 for other algorithms. The recovery results in the target region (RMSE 0.11 against 0.21 and 0.99) and linear edge region (RMSE 0.26 against 0.49 and 1.52) also indicate strong ability in detail preservation. Then the observed ESAR data is utilized for further assessment. It demonstrates that the proposed method has a significant improvement to reduce singularities and distinguish artificial buildings.these data, which were used under license for this study. Data are available from the authors with the permission of DLR.

show abstract

Section: Patch-based Nonlocal Estimationmentioning

confidence: 99%

“…Buade et al. [7] proposed a nonlocal mean method to select similarity regions by patch matching, which was further developed by Sharma [8], Shen [9], and Agershorg [10]. Shen et al.…”

Section: Introductionmentioning

confidence: 99%

Patch‐based nonlocal estimation of scattering mechanism vector in PolInSAR

Xing

Wang

et al. 2023

Electronics Letters

View full text Add to dashboard Cite

show abstract

“…A study of insect induced defoliation using C-band SAR was presented in [6], which calculated the correlation between defoliation risk and smoothed time series of backscatter values averaged over five hectare (ha) plots. In a precursor to this work, we discriminated between live and dead canopy based on an accurate estimation of polarimetric covariance from a single, full-polarimetric C-band image [16].…”

Section: A Remote Sensing Of Insect Induced Canopy Defoliationmentioning

confidence: 99%

“…Radiometric calibration and terrain correction with the GETASSE30 digital elevation model (DEM) was performed in the Sentinel Application Platform (SNAP) platform, outputting data on single-look complex (SLC) format with 10.0 m × 10.0 m spatial resolution. Polarimetric covariance matrices were estimated using the guided nonlocal means method presented in [16]. This method preserves SLC resolution and was shown to give estimates of the polarimetric features that better separate between live and dead canopy than alternative methods [16].…”

Section: Creating the Forest Mortality Mapmentioning

confidence: 99%

“…Polarimetric covariance matrices were estimated using the guided nonlocal means method presented in [16]. This method preserves SLC resolution and was shown to give estimates of the polarimetric features that better separate between live and dead canopy than alternative methods [16]. The features relevant for canopy state classification were extracted from the covariance matrix C. These are the intensities in the HH, HV, and VV channels, C 11 , C 22 , C 33 respectively, and the cross-correlation between the complex scattering coefficients for HH and VV, C 13 = |C 13 |e j∠C13 [16].…”

Section: Creating the Forest Mortality Mapmentioning

confidence: 99%

See 1 more Smart Citation

Towards Targeted Change Detection with Heterogeneous Remote Sensing Images for Forest Mortality Mapping

Agersborg¹,

Luppino²,

Anfinsen³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

In this paper we develop a method for mapping forest mortality in the forest-tundra ecotone using satellite data from heterogeneous sensors. We use medium resolution imagery in order to provide the complex pattern of forest mortality in this sparsely forested area, which has been induced by an outbreak of geometrid moths. Specifically, Landsat-5 Thematic Mapper images from before the event are used, with RADARSAT-2 providing the post-event images. We obtain the difference images for both multispectral optical and synthetic aperture radar (SAR) by using a recently developed deep learning method for translating between the two domains. These differences are stacked with the original pre-and post-event images in order to let our algorithm also learn how the areas appear before and after the change event. By doing this, and focusing on learning only the changes of interest with one-class classification (OCC), we obtain good results with very little training data.

show abstract