A Modified Local Binary Pattern Descriptor for SAR Image Matching

Ghannadi, Mohammad Amin; Saadaseresht, Mohammad

doi:10.1109/lgrs.2018.2876661

Cited by 13 publications

(9 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…en, transfer the shallow layer's parameters by freezing the learned layer, and the hyperparameters of MS-CNN are retrained on D T until the model converges to the optimal solution. Specifically, for example, as transferred the parameters of conv4, the parameter update of the layer before conv4 would be the pretrained parameters, and the parameters of the rest layers 6 Computational Intelligence and Neuroscience would be trained from scratch. Finally, the small learning rate are considered to further fine tune the model slightly, to make the model more suitable for the SAR ATR task.…”

Section: Transfer Learningmentioning

confidence: 99%

“…Generally, traditional methods aim to extract discriminative and represented features from the training samples. Traditional feature extraction methods, such as Histogram of Oriented Gradients (HOG) [5], Local Binary Pattern (LBP) [6], Principal Component Analysis (PCA) [7], and Scale Invariant Feature Transform (SIFT) [8], were applied to SAR target classification task. Song et al [9] designed a novel gradient HOG-like feature-based SAR ATR method to tackle a complex application environment.…”

Section: Introductionmentioning

confidence: 99%

“…Li et al [10] proposed a HOG descriptor-based method to match features between SAR images and optical images. Ghannadi and Saadatseresht [6] proposed a modified LBP descriptor to obtain robust features for SAR ATR. Wang et al [11] presented an improved SAR interferogram denoising method based on PCA to improve the accuracy of phase unwrapping.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Robust SAR Automatic Target Recognition Based on Transferred MS-CNN with L²-Regularization

Zhai

Deng

et al. 2019

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Though Synthetic Aperture Radar (SAR) Automatic Target Recognition (ATR) via Convolutional Neural Networks (CNNs) has made huge progress toward deep learning, some key issues still remain unsolved due to the lack of sufficient samples and robust model. In this paper, we proposed an efficient transferred Max-Slice CNN (MS-CNN) with L2-Regularization for SAR ATR, which could enrich the features and recognize the targets with superior performance. Firstly, the data amplification method is presented to reduce the computational time and enrich the raw features of SAR targets. Secondly, the proposed MS-CNN framework with L2-Regularization is trained to extract robust features, in which the L2-Regularization is incorporated to avoid the overfitting phenomenon and further optimizing our proposed model. Thirdly, transfer learning is introduced to enhance the feature representation and discrimination, which could boost the performance and robustness of the proposed model on small samples. Finally, various activation functions and dropout strategies are evaluated for further improving recognition performance. Extensive experiments demonstrated that our proposed method could not only outperform other state-of-the-art methods on the public and extended MSTAR dataset but also obtain good performance on the random small datasets.

show abstract

Section: Transfer Learningmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Robust SAR Automatic Target Recognition Based on Transferred MS-CNN with L²-Regularization

Zhai

Deng

et al. 2019

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…One of the challenges in image matching is the existence of some poorly textured regions in images. Some previous researches in SAR image matching have used image texture information in order to improve the matching process [9, 10]. In this study, a method is proposed for texture image reconstruction to improve SAR image matching whereby image texture analysis is used for optimal reconstruction.…”

Section: Introductionmentioning

confidence: 99%

Optimal texture image reconstruction method for improvement of SAR image matching

Ghannadi¹,

Saadatseresht²,

Izadi

et al. 2020

IET Radar, Sonar & Navigation

Self Cite

View full text Add to dashboard Cite

“…The methods of image matching can be generally classified into two types: feature-based methods and area-based methods [17]. Feature-based methods like scale invariant feature transform (SIFT) [18] and speeded up robust feature (SURF) [19] have been widely applied in image matching [20].…”

Section: Introductionmentioning

confidence: 99%

An Image Matching Method for SAR Orthophotos from Adjacent Orbits in Large Area Based on SAR-Moravec

et al. 2020

View full text Add to dashboard Cite

In producing orthophoto mosaic in a large area from spaceborne synthetic aperture radar (SAR) images, SAR image matching from adjacent orbits is a technical difficulty due to the speckle noise and different imaging mechanism between azimuth and range direction. In this paper, an area-based method, SAR-Moravec, is proposed for SAR orthophoto matching from adjacent orbits in a large area. Compared with the classical area-based Moravec, the template of SAR-Moravec is characterized by more directions for speckle noise restraint and a main direction consistent with the azimuth. In order to get evenly distributed matching points with high accuracy, the grid control mechanism and Gaussian pyramid from coarse to fine are introduced in matching. The whole process contains three steps. First, the pyramid images are constructed by the down-sampling process. Second, under grid control, the feature points are evenly extracted by the modified template. Third, the transformation model is iteratively calculated from the first to the last layer of the pyramid. After the matching process layer-by-layer, the final matching points and transformation model can be obtained. In the experiments, we compare the SAR-Moravec with three widely used methods, including the Moravec, the SAR-scale invariant feature transform (SAR-SIFT), and the SAR-features from an accelerated segment test (SAR-FAST). The results indicate that the proposed method has the best global matching accuracy among these methods and the matching efficiency is better than SAR-SIFT and SAR-FAST methods in large area.

show abstract

A Modified Local Binary Pattern Descriptor for SAR Image Matching

Cited by 13 publications

References 20 publications

Robust SAR Automatic Target Recognition Based on Transferred MS-CNN with L²-Regularization

Robust SAR Automatic Target Recognition Based on Transferred MS-CNN with L²-Regularization

Optimal texture image reconstruction method for improvement of SAR image matching

An Image Matching Method for SAR Orthophotos from Adjacent Orbits in Large Area Based on SAR-Moravec

Contact Info

Product

Resources

About

A Modified Local Binary Pattern Descriptor for SAR Image Matching

Cited by 13 publications

References 20 publications

Robust SAR Automatic Target Recognition Based on Transferred MS-CNN with L2-Regularization

Robust SAR Automatic Target Recognition Based on Transferred MS-CNN with L2-Regularization

Optimal texture image reconstruction method for improvement of SAR image matching

An Image Matching Method for SAR Orthophotos from Adjacent Orbits in Large Area Based on SAR-Moravec

Contact Info

Product

Resources

About

Robust SAR Automatic Target Recognition Based on Transferred MS-CNN with L²-Regularization

Robust SAR Automatic Target Recognition Based on Transferred MS-CNN with L²-Regularization