An Automatic and Novel SAR Image Registration Algorithm: A Case Study of the Chinese GF-3 Satellite

Xiang, Yuming; Wang, Feng; You, Hongjian

doi:10.3390/s18020672

Cited by 25 publications

(17 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previous study presented various techniques on CAD image database development [5] using .stl data, image data acquisition for surface inspection [6], pre-processing techniques on image registration [7][8][9][10][11][12][13][14][15][16][17][18], features extraction such as Vote-based 3D shape recognition and registration [19], edges extraction [20], reflection symmetric [21], DoG-based detector presented by deriving scale-invariant mesh features for image registration [22], Local Procustes Regression (LPR) [23], Estimation-by-Completion (EbC) [24] and Customized three-dimensional template matching [25][26][27] In this paper, the study focus on object description, scaling and image registration method. The contribution of this paper two fold.…”

Section: Introductionmentioning

confidence: 99%

Partial pose estimation of 3D rigid object system using outer box method

Zulkoffli

Bakar

2020

IJEECS

View full text Add to dashboard Cite

<p><span>This article introduces a novel approach for identify partial pose estimation using template matching method. The algorithms performs 2D correlation matching on tested image to CAD database by using regional shape representation in order to get the similar object pose in CAD database. The descriptor named outer box method, it is useful for rescale or aligning object size in both different images of tested image and CAD database image and also provide interest point for segmentation in image registration stage. The proposed algorithm were experimentally shown to be robust to apply on scale changes, various complex shape, unstructured CAD database and mixed CAD model database. Last part, the identified pose and its retrieved pose angle was calculated and achieved high accuracyin range ±0.388˚ to ±1.471˚.</span></p>

show abstract

Section: Introductionmentioning

confidence: 99%

Partial pose estimation of 3D rigid object system using outer box method

Zulkoffli

Bakar

2020

IJEECS

View full text Add to dashboard Cite

show abstract

“…Wang et al [11] presented an improved SAR interferogram denoising method based on PCA to improve the accuracy of phase unwrapping. Xiang et al [12] combined an adaptive sampling method with SAR-SIFT to eliminate obvious scale difference images. However, the presence of speckle noises and lack of robust features have seriously degraded the feature robustness of the SAR image.…”

Section: Introductionmentioning

confidence: 99%

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

“…Chang et al [ 7 ] combined the SIFT algorithm with scale constraints and geometric constraints to achieve highly-accurate registration of high-resolution satellite image. Xiang et al [ 8 ] used a SIFT-liked matching algorithm to extract reliable matches for GF-3 SAR images and then provided a coarse-to-fine registration algorithm, giving a robust, efficient and precise performance of SAR image. Qi and Zhu [ 9 ] decomposed remote sensing images by discrete wavelet transform (DWT), and extracted the SIFT features on low-frequency sub-images, then used a coarse-to-fine strategy to improve the precision and efficiency of SIFT.…”

Section: Introductionmentioning

confidence: 99%

Improved WαSH Feature Matching Based on 2D-DWT for Stereo Remote Sensing Images

Deng

Yang

et al. 2018

Sensors

View full text Add to dashboard Cite

Image matching is an outstanding issue because of the existing of geometric and radiometric distortion in stereo remote sensing images. Weighted α-shape (WαSH) local invariant features are tolerant to image rotation, scale change, affine deformation, illumination change, and blurring. However, since the number of WαSH features is small, it is difficult to get enough matches to estimate the satisfactory homography matrix or fundamental matrix. In addition, the WαSH detector is extremely sensitive to image noise because it is built on sampled edges. Considering the shortcomings of the WαSH detector, this paper improves the WαSH feature matching method based on the 2D discrete wavelet transform (2D-DWT). The method firstly performs 2D-DWT on the image, and then detects WαSH features on the transformed images. According to the methods of descriptor construction for WαSH features, three matching methods on the basis of wavelet transform WαSH features (WWF), improved wavelet transform WαSH features (IWWF), and layered IWWF (LIWWF) are distinguished with respect to the character of the sub-images. The experimental results on the dataset containing affine distortion, scale distortion, illumination change, and noise images, showed that the proposed methods acquired more matches and better stableness than WαSH. Experimentation on remote sensing images with less affine distortion and slight noise showed that the proposed methods obtained the correct matching rate greater than 90%. For images containing severe distortion, KAZE obtained a 35.71% correct matching rate, which is unacceptable for calculating the homography matrix, while IWWF achieved a 71.42% correct matching rate. IWWF was the only method that achieved the correct matching rate of no less than 50% for all four test stereo remote sensing image pairs and was the most stable compared to MSER, DWT-MSER, WαSH, DWT-WαSH, KAZE, WWF, and LIWWF.

show abstract

An Automatic and Novel SAR Image Registration Algorithm: A Case Study of the Chinese GF-3 Satellite

Cited by 25 publications

References 34 publications

Partial pose estimation of 3D rigid object system using outer box method

Partial pose estimation of 3D rigid object system using outer box method

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

Improved WαSH Feature Matching Based on 2D-DWT for Stereo Remote Sensing Images

Contact Info

Product

Resources

About

An Automatic and Novel SAR Image Registration Algorithm: A Case Study of the Chinese GF-3 Satellite

Cited by 25 publications

References 34 publications

Partial pose estimation of 3D rigid object system using outer box method

Partial pose estimation of 3D rigid object system using outer box method

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

Improved WαSH Feature Matching Based on 2D-DWT for Stereo Remote Sensing Images

Contact Info

Product

Resources

About

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