2015
DOI: 10.1109/lgrs.2014.2351396
|View full text |Cite
|
Sign up to set email alerts
|

SAR Image Registration Using Phase Congruency and Nonlinear Diffusion-Based SIFT

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
14
0

Year Published

2016
2016
2021
2021

Publication Types

Select...
6
2

Relationship

0
8

Authors

Journals

citations
Cited by 81 publications
(14 citation statements)
references
References 12 publications
0
14
0
Order By: Relevance
“…The GLOH descriptor has been successfully applied to many registration problems and shown to be efficient [7], [15], [27]. For SAR image registration, given speckle noise, it is unreliable to consider a single-scale patch to characterize feature points.…”
Section: A Novel Multiscale Circle Descriptor (Mcd)mentioning
confidence: 99%
See 2 more Smart Citations
“…The GLOH descriptor has been successfully applied to many registration problems and shown to be efficient [7], [15], [27]. For SAR image registration, given speckle noise, it is unreliable to consider a single-scale patch to characterize feature points.…”
Section: A Novel Multiscale Circle Descriptor (Mcd)mentioning
confidence: 99%
“…We found that different values of the response threshold SH d of the Harris function mentioned in Section III-C greatly impact the experimental results. Thus, we test the four algorithms based on the Harris corner detector; namely, SAR-SIFT [7], NDSS-SIFT [27], our method without MCD, and our method with MCD. Fig.…”
Section: Performance Of the Harris Function For Different Response Thresholdsmentioning
confidence: 99%
See 1 more Smart Citation
“…However, the data required smoothing in prior to accommodate atmospheric noise. (Fan et al, 2014) used the model in registering SAR (Synthetic Aperture Radar) images. They used nonlinear diffusion scale space along with the ratio of exponential weighted average operator to reduce noise and preserve fine details in the image.…”
Section: Overviewmentioning
confidence: 99%
“…However, it fails to provide favorable results while dealing with the SAR images. The reason is the fact that Gaussian blurring is one instance of isotropic diffusion filtering which is sensitive to speckle noise and does not respect the natural boundaries of the object 7 . As a consequence, many unstable keypoints are brought from the Gaussian scale space of SIFT and then the matching performance is degraded.…”
Section: Introductionmentioning
confidence: 99%