A new constrained spectral clustering for SAR image segmentation

Zou, Haishuang; Zhou, Wenjing; Zhang, Li; Wu, Caili; Liu, Ruochen; Jiao, Licheng

doi:10.1109/apsar.2009.5374114

Cited by 7 publications

(10 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…4b. As can be seen, in this segmentation, the creek marked with [24] c Segmentation obtained by SCE [25] d Segmentation obtained by KSFCM [26] e Segmentation obtained by FMFSSC [27] f Segmentation obtained by CSC [28] g Segmentation obtained by QCSC [29] h Segmentation obtained by the proposed method a red box below the crop has received a grey label and vegetation has been recognised. In addition, the river water advances to the crop which is marked with a blue circle that has not been properly labelled, and the place where the water enters has been labelled grey and white.…”

Section: Segmentation Of Real Sar Imagementioning

confidence: 89%

“…The proposed method has been tested on simulated and real SAR images, and compared with the methods proposed in [25][26][27][28][29][30], which will be referred to as sparse spectral clustering (SSC), spectral clustering ensemble (SCE), Kernel-based spatial FCM (KSFCM), fussing multi-feature spatial spectral clustering (FMFSSC), constrained spectral clustering (CSC) and quantum CSC (QCSC), respectively. These methods used wavelet and GLCM features for SAR image segmentation.…”

Section: Special Issue On Unsupervised Feature Learning For Visionmentioning

confidence: 99%

See 1 more Smart Citation

Unsupervised feature learning based on sparse coding and spectral clustering for segmentation of synthetic aperture radar images

Rahmani

Akbarizadeh

2015

IET Computer Vision

View full text Add to dashboard Cite

Synthetic aperture radar (SAR) image segmentation is fundamental for the interpretation and understanding of these images. In this process, the representation of SAR image features plays an important role. Spectral clustering is an image segmentation method making it possible to combine features and cues. This study presents a new spectral clustering method using unsupervised feature learning (UFL). In this method, the SAR image is primarily processed by the non‐negative matrix factorisation (NMF) algorithm and then non‐negative features containing spatial structure information are extracted. Afterwards, the extracted features are learned using a sparse coding algorithm to increase the discrimination power of the features. Sparse coding is an unsupervised learning algorithm which finds the patterns or high‐level semantics of the data. Ultimately, the SAR image segmentation operation is performed by applying spectral clustering on learned features. In this method, sparse coding learns features and simultaneously creates the similarity function required in spectral clustering through the production of sparse coefficients. Therefore this method avoids the Gaussian similarity function, which has a problem with scale parameter adjustment that is one of the drawbacks of spectral clustering methods. The results demonstrate that, compared with wavelet and GLCM features, NMF features manage to obtain more meaningful information and provide a better SAR image segmentation result. The results have also demonstrated that SAR image segmentation using learned features is significantly improved compared with segmentation by unlearned features. The experimental results indicate the effect of UFL on SAR image segmentation.

show abstract

Section: Segmentation Of Real Sar Imagementioning

confidence: 89%

Section: Special Issue On Unsupervised Feature Learning For Visionmentioning

confidence: 99%

Unsupervised feature learning based on sparse coding and spectral clustering for segmentation of synthetic aperture radar images

Rahmani

Akbarizadeh

2015

IET Computer Vision

View full text Add to dashboard Cite

show abstract

“…To compute relevant similarity, the Mahalanobis distance is modified so that it minimizes the distance between the must-link pixels and maximizes the distance between the cannot-link pixels. 23,24 Furthermore, the constraints can be propagated to the nearest neighborhood pixels that are found in the discriminating feature sub-spaces, rather than those that are identified in the original feature space. 15 However, the random forest procedure used for this purpose, requires a time consuming training step.…”

Section: Introductionmentioning

confidence: 99%

Constrained feature selection for semisupervised color-texture image segmentation using spectral clustering

2021

View full text Add to dashboard Cite

HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

show abstract

“…Normally, texture information of SAR images can be used for mosaic image [30]- [32]. Methods to extract texture information include kurtosis wavelet energy (KWE) [30], kurtosis curvelet energy (KCE) [31], spectral clustering algorithm [33] and convolutional neural network (CNN) [34], [35]. At present, machine vision has greatly support for development of electronic engineering which is important for imaging-based automatic inspection.…”

Section: Introductionmentioning

confidence: 99%

Extended Polar Format Algorithm for Large-Scene High-Resolution WAS-SAR imaging

Chen

Wang

et al. 2021

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

The polar format algorithm (PFA) is a frequencydomain algorithm with high computational efficiency especially for the high resolution imaging applications in wide angle staring synthetic aperture radar (WAS-SAR). Because of the image distortion and defocus caused by plane wave assumption, the effective imaging scene is limited to a small region which is defined as defocus-negligible region. In this paper, a novel sub-block imaging algorithm which firstly applied to WAS-SAR is proposed. This method combines sub-block imaging and spatially variant post filtering (SVPF) which can not only effectively extend depth of focus (DOF) of PFA but also greatly reduces computational burden. What's more, the method of mosaic image is proposed in this paper which can eliminate discontinuity between adjacent sub-blocks. Firstly, the large scene is segmented into multiple regions along range direction and the size of each region is still larger than that of defocus-negligible region. Secondly, refocusing is implemented by a phase compensation of the already de-chirped signal with the center of sub-blocks respectively and each sub-block image is formed by PFA. Thirdly, the region in each sub-block that exceeds the defocus-negligible region is corrected by SVPF and the image distortion can be corrected by 2-D interpolation. Finally, mosaic image method is applied to get the large-scene image with high quality. The algorithm proposed in this paper is efficient and solve the problems in the process of SAR image mosaic such as dislocation and grey scale differences. The simulated and experimental data results demonstrate the effectiveness and practicability of the proposed algorithm.Index Terms-wide angle staring synthetic aperture radar (WAS-SAR), polar format algorithm (PFA), large scene, high resolution, sub-block imaging, SVPF

show abstract

A new constrained spectral clustering for SAR image segmentation

Cited by 7 publications

References 9 publications

Unsupervised feature learning based on sparse coding and spectral clustering for segmentation of synthetic aperture radar images

Unsupervised feature learning based on sparse coding and spectral clustering for segmentation of synthetic aperture radar images

Constrained feature selection for semisupervised color-texture image segmentation using spectral clustering

Extended Polar Format Algorithm for Large-Scene High-Resolution WAS-SAR imaging

Contact Info

Product

Resources

About