Entropy-Based Non-Local Means Filter for Single-Look SAR Speckle Reduction
Speckle is an interference phenomenon that contaminates images captured by coherent illumination systems. Due to its multiplicative and non-Gaussian nature, it is challenging to eliminate. The non-local means approach to noise reduction has proven flexible and provided good results. We propose in th...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-01-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/3/509 |
| _version_ | 1797485065223012352 |
|---|---|
| author | Debora Chan Juliana Gambini Alejandro C. Frery |
| author_facet | Debora Chan Juliana Gambini Alejandro C. Frery |
| author_sort | Debora Chan |
| collection | DOAJ |
| description | Speckle is an interference phenomenon that contaminates images captured by coherent illumination systems. Due to its multiplicative and non-Gaussian nature, it is challenging to eliminate. The non-local means approach to noise reduction has proven flexible and provided good results. We propose in this work a new non-local means filter for single-look speckled data using the Shannon and Rényi entropies under the <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi mathvariant="script">G</mi><mn>0</mn></msup></semantics></math></inline-formula> model. We obtain the necessary mathematical apparatus (the Fisher information matrix and asymptotic variance of maximum likelihood estimators). The similarity between samples of the patches relies on a parametric statistical test that verifies the evidence whether two samples have the same entropy or not. Then, we build the convolution mask by transforming the <i>p</i>-value into weights with a smooth activation function. The results are encouraging, as the filtered images have a better signal-to-noise ratio, they preserve the mean, and the edges are not severely blurred. The proposed algorithm is compared with three successful filters: SRAD (Speckle Reducing Anisotropic Diffusion), Lee, and FANS (Fast Adaptive Nonlocal SAR Despeckling), showing the new method’s competitiveness. |
| first_indexed | 2024-03-09T23:14:29Z |
| format | Article |
| id | doaj.art-3dbd9a0173ce46a4a5c85a71024a7d77 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T23:14:29Z |
| publishDate | 2022-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-3dbd9a0173ce46a4a5c85a71024a7d772023-11-23T17:38:47ZengMDPI AGRemote Sensing2072-42922022-01-0114350910.3390/rs14030509Entropy-Based Non-Local Means Filter for Single-Look SAR Speckle ReductionDebora Chan0Juliana Gambini1Alejandro C. Frery2Facultad Regional Buenos Aires, Universidad Tecnológica Nacional, Ciudad Autonoma de Buenos Aires C1179AAQ, ArgentinaDepartamento de Ingeniería Informática, Instituto Tecnológico de Buenos Aires, Av. Madero 399, Buenos Aires C1106ACD, ArgentinaSchool of Mathematics and Statistics, Victoria University of Wellington, Wellington 6140, New ZealandSpeckle is an interference phenomenon that contaminates images captured by coherent illumination systems. Due to its multiplicative and non-Gaussian nature, it is challenging to eliminate. The non-local means approach to noise reduction has proven flexible and provided good results. We propose in this work a new non-local means filter for single-look speckled data using the Shannon and Rényi entropies under the <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi mathvariant="script">G</mi><mn>0</mn></msup></semantics></math></inline-formula> model. We obtain the necessary mathematical apparatus (the Fisher information matrix and asymptotic variance of maximum likelihood estimators). The similarity between samples of the patches relies on a parametric statistical test that verifies the evidence whether two samples have the same entropy or not. Then, we build the convolution mask by transforming the <i>p</i>-value into weights with a smooth activation function. The results are encouraging, as the filtered images have a better signal-to-noise ratio, they preserve the mean, and the edges are not severely blurred. The proposed algorithm is compared with three successful filters: SRAD (Speckle Reducing Anisotropic Diffusion), Lee, and FANS (Fast Adaptive Nonlocal SAR Despeckling), showing the new method’s competitiveness.https://www.mdpi.com/2072-4292/14/3/509non-local meansspeckle filter<i>h</i>-<i>ϕ</i> entropiesasymptotic variance |
| spellingShingle | Debora Chan Juliana Gambini Alejandro C. Frery Entropy-Based Non-Local Means Filter for Single-Look SAR Speckle Reduction Remote Sensing non-local means speckle filter <i>h</i>-<i>ϕ</i> entropies asymptotic variance |
| title | Entropy-Based Non-Local Means Filter for Single-Look SAR Speckle Reduction |
| title_full | Entropy-Based Non-Local Means Filter for Single-Look SAR Speckle Reduction |
| title_fullStr | Entropy-Based Non-Local Means Filter for Single-Look SAR Speckle Reduction |
| title_full_unstemmed | Entropy-Based Non-Local Means Filter for Single-Look SAR Speckle Reduction |
| title_short | Entropy-Based Non-Local Means Filter for Single-Look SAR Speckle Reduction |
| title_sort | entropy based non local means filter for single look sar speckle reduction |
| topic | non-local means speckle filter <i>h</i>-<i>ϕ</i> entropies asymptotic variance |
| url | https://www.mdpi.com/2072-4292/14/3/509 |
| work_keys_str_mv | AT deborachan entropybasednonlocalmeansfilterforsinglelooksarspecklereduction AT julianagambini entropybasednonlocalmeansfilterforsinglelooksarspecklereduction AT alejandrocfrery entropybasednonlocalmeansfilterforsinglelooksarspecklereduction |