SAR imagery segmentation by statistical region growing and hierarchical merging

Carvalho, Elisete Mendes; Ushizima, Daniela; Medeiros, Fátima N. S. de; Martins, C. I. O.; Marques, Régis C. P.; Oliveira, Isaura Nelsivania Sombra

doi:10.1016/j.dsp.2009.10.014

Cited by 42 publications

(19 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Traditionally, there are two strategies to stop a region iterative merging process. The first one [17], [25], [26], [33], [34] sets a threshold beforehand and compares the similarity criterion (or the number of remaining regions in the image domain) with the threshold before each merging operation, which decides whether to continue the merging process. Therefore, the threshold is very crucial, as a large threshold could lead to undersegmentation results, whereas a small threshold could cause oversegmentation results.…”

Section: B Halt Conditionmentioning

confidence: 99%

“…In fact, many algorithms operating in image domain have been studied and proposed. One classical family of approaches is the region growing (or region merging) methods [17], [25]- [36], among which different kinds of information were easily utilized to judge whether to merge neighboring pixels or patches. This kind of methods is particularly suitable for segmenting the SAR images containing obvious boundaries, e.g., ice images [31], [34]- [36] and oil spill images [32].…”

Section: Introductionmentioning

confidence: 99%

“…The advantage of CRIM is that it can merge most of the superpixels inside true segments at a very high speed, and hence improve the features' accuracy and reduce the computation burden for the fine classification stage. Compared with the traditional region merging algorithms [17], [25], [26], [29], [31], [33]- [35], [48], CRIM is able to make use of much more information. Moreover, a global halt index (HI) is carefully designed in CRIM to produce the segmentation results with global consistency in homogeneity.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

CRIM-FCHO: SAR Image Two-Stage Segmentation With Multifeature Ensemble

Jiao

Liu

2016

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

This paper investigates the synthetic aperture radar (SAR) image segmentation in terms of feature analysis and fusion and develops a new algorithm based on multifeature ensemble accordingly. This paper is characterized by two aspects. First, multiple heterogeneous features are extracted to accurately describe the objects in SAR images. These features are then integrated in the feature level and the similarity level, respectively, to avoid the mutual influences between different kinds of features and maximize the discriminability of the similarity measure between objects. Second, a two-stage algorithm consisting of a coarse merging stage and a fine classification stage is proposed. In the coarse merging stage, a context-based region iterative merging algorithm is designed to merge most of the unambiguous superpixels in image domain at a high speed. In the fine classification stage, a fuzzy clustering algorithm incorporating hybrid optimization is developed to balance the efficiency and the robustness of the algorithm by simultaneously searching heuristically in the complete high-dimension feature space and searching along the direction of the gradient steepest descent in each feature subspace. The effectiveness of the proposed method has been successfully validated on synthetic and real SAR images.Index Terms-Fuzzy clustering, image segmentation, multifeature ensemble, region merging, similarity measures, synthetic aperture radar (SAR) image.

show abstract

Section: B Halt Conditionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

CRIM-FCHO: SAR Image Two-Stage Segmentation With Multifeature Ensemble

Jiao

Liu

2016

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

show abstract

“…In the Synthetic Aperture Radar (SAR) community, using a stack of images to identify SHP in space has been extensively used for various SAR applications, such as adaptive filtering [1][2][3][4][5][6] (multitemporal configurations) [7,8] (single-temporal configurations), complex coherence estimation [9][10][11][12], and image segmentation [13]. As processing images based on SHP precludes the participation of irrelevant observations, successful SHP identification plays a key role in the SAR image analysis.…”

Section: Introductionmentioning

confidence: 99%

Identification of Statistically Homogeneous Pixels Based on One-Sample Test

Lin

Perissin

2017

Remote Sensing

View full text Add to dashboard Cite

Statistically homogeneous pixels (SHP) play a crucial role in synthetic aperture radar (SAR) analysis. In past studies, various two-sample tests were applied on multitemporal SAR data stacks under the assumption of having stationary backscattering properties over time. In this letter, we propose the Robust T-test (TR) to improve the effectiveness of test operation. The TR test reduces the impact of temporal variabilities and outliers, thus helping to identify SHP with assurances of similar temporal behaviors. This method includes three steps: (1) signal suppression; (2) outlier removal; and (3) one-sample test. In the experiments, we apply the TR test on both simulated and real data. Different stack sizes, types of distributions, and hypothesis tests are compared. Results of both experiments signify that the TR test outperforms conventional approaches and provides reliable SHP for SAR image analysis.

show abstract

“…Researchers have proposed many kinds of segmentation algorithms for SAR images, which include threshold methods, 3,[17][18][19] spectral clustering (SC) algorithms, 20,21 statistic model-based methods, 1,14,15,[22][23][24][25] artificial intelligence methods, [26][27][28][29][30] support vector machine (SVM), 6,31 region growing methods, 15,[32][33][34][35] and so on. Among these algorithms, cluster-based algorithms form one popular and representative family, whose main idea is to group pixels in such a way that the pixels in the same group are more similar to each other than those in other groups.…”

Section: Introductionmentioning

confidence: 99%

Independent feature subspace iterative optimization based fuzzy clustering for synthetic aperture radar image segmentation

Feng

2015

J. Appl. Remote Sens

View full text Add to dashboard Cite

Abstract. Synthetic aperture radar (SAR) image segmentation is investigated from feature extraction to algorithm design, which is characterized by two aspects: (1) multiple heterogeneous features are extracted to describe SAR images and the corresponding similarity measures are developed independently to avoid the mutual influences between different features in order to enhance the discriminability of the final similarity between objects. (2) A method called fuzzy clustering based on independent subspace iterative optimization (FCISIO) is proposed. FCISIO integrates multiple features into an objective function which is then iteratively optimized in each feature subspace to obtain final segmentation results. This strategy can protect the distribution structures of the data points in each feature subspace, which realizes an effective way to integrate multiple features of different properties. In order to improve the computation speed and the accuracy of feature description for FCISIO, we design a region merging algorithm before FCISIO which can use many kinds of information to quickly merge regions inside the true segments. Experiments on synthetic and real SAR images show that the proposed method is effective and robust and can obtain good segmentation results with a very short running time. © The Authors. Published by SPIE under a Creative Commons Attribution 3.0 Unported License. Distribution or reproduction of this work in whole or in part requires full attribution of the original publication, including its DOI.

show abstract

SAR imagery segmentation by statistical region growing and hierarchical merging

Cited by 42 publications

References 28 publications

CRIM-FCHO: SAR Image Two-Stage Segmentation With Multifeature Ensemble

CRIM-FCHO: SAR Image Two-Stage Segmentation With Multifeature Ensemble

Identification of Statistically Homogeneous Pixels Based on One-Sample Test

Independent feature subspace iterative optimization based fuzzy clustering for synthetic aperture radar image segmentation

Contact Info

Product

Resources

About