Abstract
This paper proposes a refined bilateral filtering algorithm based on adaptively trimmed-statistics (ATS-RBF) for speckle reduction in SAR imagery. The new de-speckling method is based on the bilateral filtering method, where the similarities of gray levels and the spatial location of the neighboring pixels are exploited. However, the traditional bilateral filter is not effective to reduce the strong speckle, which is often presented as impulse noise. The ATS-RBF designs an adaptive sample trimming method to properly select the samples in the local reference window and the trimming depth used for sample trimming is automatically derived according to the homogeneity of the local reference window. Furthermore, an alterable window size-based scheme is proposed to enhance the speckle noise smoothing strength in homogeneous backgrounds. Finally, bilateral filtering is applied using the adaptively trimmed samples. The ATS-RBF has an excellent speckle noise smoothing performance while preserving the edges and the texture information of the SAR images. The experiments validate the effectiveness of the proposed method using TerraSAR-X images.
Highlights
Synthetic aperture radar (SAR) is an active radar which has the advantages of all-time and all-weather sensing capability
The proposed ATS-RBF designs an adaptive threshold based method to trim the samples in the local reference window automatically, the trimmed samples are used for bilateral filtering, so the strong speckle noise can be greatly smoothed
The trimming depth is of great significance for the de-speckling performance, Fig. 5 illustrates the filtering performance of the trimmed statistics based bilateral filter under different trimming depths of α, where the speckle noise smoothing ability on a homogeneous region is evaluated using the Equivalent Number of Looks (ENL) [34], and the edge preservation capability on a detailed image is evaluated through the Edge Sustaining Index (ESI) [34]
Summary
Synthetic aperture radar (SAR) is an active radar which has the advantages of all-time and all-weather sensing capability. The proposed ATS-RBF designs an adaptive threshold based method to trim the samples in the local reference window automatically, the trimmed samples are used for bilateral filtering, so the strong speckle noise can be greatly smoothed. The trimming depth is of great significance for the de-speckling performance, Fig. 5 illustrates the filtering performance of the trimmed statistics based bilateral filter under different trimming depths of α, where the speckle noise smoothing ability on a homogeneous region is evaluated using the Equivalent Number of Looks (ENL) [34], and the edge preservation capability on a detailed image is evaluated through the Edge Sustaining Index (ESI) [34]. The adjusted combined similarity weights are much more precise, and the strong speckle noise can be smoothed while the details of the images are well sustained
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
