Deformable Dictionary Learning for SAR Image Change Detection

Lin, Li; Zhao, Yongqiang; Sun, Jinjun; Chan, Jonathan Cheung-Wai; Kong, Seong G.; Liu, Zhun-ga

doi:10.1109/tgrs.2018.2829630

Cited by 18 publications

(18 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From them we can see that there are little difference among the detection results of different methods, since the original images are relatively clean. MMK [30], LRCN [28], CA [29] and ESLM present similar results, and there are some wrongly detected pixels in the result of DDL [26] and NR [27]. Compared with the unsupervised methods, ESLM not only provides more refined edges but also less noisy pixels when compared with other methods.…”

Section: Experimental Results On the First Three Datasetsmentioning

confidence: 72%

“…The change detection results of the first three datasets are shown in Fig.10∼Fig.12 respectively. Fig.10(a)-(f) show the detected change maps of the first dataset, by MMK [30], DDL [26], NR [27], LRCN [28], CA [29] and ESLM respectively. From them we can see that there are little difference among the detection results of different methods, since the original images are relatively clean.…”

Section: Experimental Results On the First Three Datasetsmentioning

confidence: 99%

“…Two obviously changed areas are selected in original SAR images to verify the performance of the proposed algorithm. In order to verify the effectiveness of ESLM, five related and some state-of-the-art algorithms are used for a comparison, including the Mathematical Morphology and K-means clustering (MMK) [30], Deformable Dictionary Learning (DDL) [26], Neighborhood-based Ratio (NR) [27], Local Restricted Convolutional Network (LRCN) [28] and Contractive Autoencoder (CA) [29]. Both the visual results and numerical results are demonstrated, along with the consumed time of these methods.…”

Section: ) Datasetmentioning

confidence: 99%

See 2 more Smart Citations

Extreme Self-Paced Learning Machine for On-Orbit SAR Images Change Detection

et al. 2019

View full text Add to dashboard Cite

With the rapid development of earth observation satellites, on-orbit data processing is becoming more and more desirable. In this paper, a new on-orbit change detection method for Synthetic Aperture Radar (SAR) images, is proposed via an Extreme Self-paced Learning Machine (ESLM). First, a reflectivityspatial affinity is defined to measure the similarity between two segmented super-pixels, to identify the initial three groups of pixels: strictly changed, strictly unchanged and fuzzy pixels. Then a new extreme self-paced learning machine is developed, by gradually selecting the most confident changed pixels and predicting the changed pixels in an incremental pattern. Moreover, both the labeled and unlabeled samples are explored to realize semi-supervised classification. Different with the available methods, ESLM works in a selfpaced learning pattern and achieves accurate detection, for it can automatically choose the training samples and explore unlabeled samples to enhance the online prediction of changes. Therefore, ESLM has the characteristics of accurate and robust detection, parameter free, low-complexity and rapid implementation, which is very suitable for on-orbit processing. Some experiments are taken on five real benchmark datasets, and the results verify the effectiveness of ESLM. INDEX TERMS Change detection, synthetic aperture radar, extreme self-paced learning machine, affinity propagation super-pixel clustering, manifold regularizer.

show abstract

Section: Experimental Results On the First Three Datasetsmentioning

confidence: 72%

Section: Experimental Results On the First Three Datasetsmentioning

confidence: 99%

Section: ) Datasetmentioning

confidence: 99%

See 1 more Smart Citation

Extreme Self-Paced Learning Machine for On-Orbit SAR Images Change Detection

et al. 2019

View full text Add to dashboard Cite

show abstract

“…In general, change detection algorithm consists of three steps: 1) image preprocessing; 2) difference image (DI) formation of a pair of multi-temporal images; 3) DI image analysis to achieve segmentation of the changed regions [10][11][12][13].…”

Section: Introductionmentioning

confidence: 99%

SAR Image Change Detection Based on Data Optimization and Self-Supervised Learning

Wang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

In the SAR change detection algorithm based on self-supervised learning, speckle noise reduces the difference image (DI) quality. Therefore, the contrast of the DI is low, and its change area is not significant. Moreover, the preclassification algorithm with the poor robustness makes the classification results of the low-quality DI inaccurate. When the wrong labels are sent into the classification network, the accuracy of the final detection results is reduced. First, to improve the quality of the initial DI, we design an adaptive gamma correction algorithm that adjusts the contrast according to the mean value of the initial DI and the variation coefficient β. The contrast of the new DI generated by this algorithm is higher. Furthermore, to suppress the noise, we adopt a new algorithm based on popular ranking to obtain the saliency map of the new DI. Combining the initial DI with this saliency map, a high-quality DI with a low noise level is obtained. After that, we introduce the structure tensor into the fuzzy local information c-means clustering algorithm (FLICM) to classify the DI more accurately. The new clustering algorithm improves the accuracy of preclassification, especially its hierarchical version. Besides, we use the structure tensor to generate the structure maps of the original images. Finally, according to the prior information obtained from the preclassification, we use a convolution wavelet neural network (CWNN) to supervise and train the structure maps of the original images. Experimental results show that the DI generated by us is closer to the groundtruth than other methods. Our preclassification algorithm performs better. Our algorithm shows higher detection accuracy for SAR images with strong noise than some advanced change detection algorithms. INDEX TERMS Adaptive gamma correction, convolution wavelet neural network, fuzzy local information c-means clustering, saliency map, structure tensor.

show abstract

“…With the development of remote sensing techniques, a large collection of high-resolution remote sensing images is becoming available. These images have been widely applied to many fields [1][2][3], such as urban planning, natural hazard detection, and environment monitoring. For this reason, more and more research efforts have been put into developing methods for remote sensing scene classification which is a hot research topic in the remote sensing field to better interpret the images [4][5][6].…”

Section: Introductionmentioning

confidence: 99%

Remote sensing scene classification based on rotation-invariant feature learning and joint decision making

Zhou

Liu

Zhao

et al. 2019

J Image Video Proc.

View full text Add to dashboard Cite

With the popular use of high-resolution satellite images, remote sensing scene classification has always been a hot research topic in its related areas. However, limited to the issues of remote sensing datasets including the small scale of scene classes, the lack of rich label information and so on, it is quite challenging for deep learning methods to learn powerful feature representation. To overcome this problem, we propose a rotation-invariant feature learning and joint decision-making method based on Siamese convolutional neural networks with the combination of identification and verification models. Firstly, a novel data augmentation strategy is proposed specially for the Siamese model to learning rotation-invariant features. Secondly, a joint decision mechanism is introduced in our method, which is realized by the identification and verification model to better improve the classification performance. The proposed method can not only suppress problems caused by lack of rich label samples but also improve the robustness of Siamese convolutional neural networks. Experimental results demonstrate that the proposed method is effective and efficient for remote sensing scene classification.

show abstract

Deformable Dictionary Learning for SAR Image Change Detection

Cited by 18 publications

References 50 publications

Extreme Self-Paced Learning Machine for On-Orbit SAR Images Change Detection

Extreme Self-Paced Learning Machine for On-Orbit SAR Images Change Detection

SAR Image Change Detection Based on Data Optimization and Self-Supervised Learning

Remote sensing scene classification based on rotation-invariant feature learning and joint decision making

Contact Info

Product

Resources

About