Land Cover Classification for Synthetic Aperture Radar Imagery by Using Unsupervised Methods

Yumus, Duygu; Ozkazanc, Yakup

doi:10.1109/rast.2019.8767877

Cited by 5 publications

(4 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In urban settings, where mixed pixel problems are often encountered, the strength of mathematical morphological method as being both pre-and post-processing methods provide the ideal spatialspectral context of urban neighbourhoods. Central to mathematical morphology method are the opening and closing operations [13]. The standard opening and closing operations are repetitive and uses a customised pixelated structure defined by a radial distance to remove unwanted structures in the case of opening or to fill structural gaps in the closing operation.…”

Section: Morphological Feature Extractionmentioning

confidence: 99%

“…However, due to susceptibility of optical imagery to adverse weather conditions, recent studies have tested and confirmed the utility of only microwave RS imagery for urban LULC. For example; [13] tested several machine learning algorithms for urban LULC using Sentinel-1 Synthetic Aperture Radar (S1-SAR) datasets over Istanbul and achieved an overall accuracy of 85.17%. Similarity, [14], reported overall classification accuracies of 88.8% when multi-season S1-SAR imagery were used to delineate vegetative areas.…”

Section: Introductionmentioning

confidence: 99%

“…Although, S1-SAR was recently (first mission in 2014) launched by the European Space Agency (ESA), its Interferometric Wide (IW) swath and dual-polarised channels, provides single source sensor data where significant urban features can be extracted for land-based targets [13], [15]. S1-SAR imagery are also acquired with limited weather interference and independent of solar illumination, with a near-global coverage and operating at centre wavelength of 5.067cm larger than most atmospheric aerosols.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Reviewing the Pertinence of Sentinel-1 SAR for Urban Land Use Land Cover Classification

Abudu¹,

Parvin²,

Andogah³

2020

IJSER

View full text Add to dashboard Cite

Conventional approaches for urban land use land cover classification and quantification of land use changes have often relied on the ground surveys and urban censuses of urban surface properties. Advent of Remote Sensing technology supporting metric to centimetric spatial resolutions with simultaneous wide coverage, significantly reduced huge operational costs previously encountered using ground surveys. Weather, sensor’s spatial resolution and the complex compositions of urban areas comprising concrete, metallic, water, bare- and vegetation-covers, limits Remote Sensing ability to accurately discriminate urban features. The launch of Sentinel-1 Synthetic Aperture Radar, which operates at metric resolution and microwave frequencies evades the weather limitations and has been reported to accurately quantify urban compositions. This paper assessed the feasibility of Sentinel-1 SAR data for urban land use land cover classification by reviewing research papers that utilised these data. The review found that since 2014, 11 studies have specifically utilised the datasets.

show abstract

Section: Morphological Feature Extractionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Reviewing the Pertinence of Sentinel-1 SAR for Urban Land Use Land Cover Classification

Abudu¹,

Parvin²,

Andogah³

2020

IJSER

View full text Add to dashboard Cite

show abstract

“…To classify such high-dimensional complex data with a large number of classes, in recent years researchers have proposed several techniques. Some of these are pure classification techniques [10][11][12], while others use clustering algorithms to classify data [30][31][32][33][34]. The major issue with these techniques is the poor performance of classifying high-dimensional data with a large number of classes in terms of classification accuracy and computation cost.…”

Section: Introductionmentioning

confidence: 99%

A Hierarchical Gamma Mixture Model-Based Method for Classification of High-Dimensional Data

Azhar

Huang

2019

Entropy

View full text Add to dashboard Cite

Data classification is an important research topic in the field of data mining. With the rapid development in social media sites and IoT devices, data have grown tremendously in volume and complexity, which has resulted in a lot of large and complex high-dimensional data. Classifying such high-dimensional complex data with a large number of classes has been a great challenge for current state-of-the-art methods. This paper presents a novel, hierarchical, gamma mixture model-based unsupervised method for classifying high-dimensional data with a large number of classes. In this method, we first partition the features of the dataset into feature strata by using k-means. Then, a set of subspace data sets is generated from the feature strata by using the stratified subspace sampling method. After that, the GMM Tree algorithm is used to identify the number of clusters and initial clusters in each subspace dataset and passing these initial cluster centers to k-means to generate base subspace clustering results. Then, the subspace clustering result is integrated into an object cluster association (OCA) matrix by using the link-based method. The ensemble clustering result is generated from the OCA matrix by the k-means algorithm with the number of clusters identified by the GMM Tree algorithm. After producing the ensemble clustering result, the dominant class label is assigned to each cluster after computing the purity. A classification is made on the object by computing the distance between the new object and the center of each cluster in the classifier, and the class label of the cluster is assigned to the new object which has the shortest distance. A series of experiments were conducted on twelve synthetic and eight real-world data sets, with different numbers of classes, features, and objects. The experimental results have shown that the new method outperforms other state-of-the-art techniques to classify data in most of the data sets.

show abstract

RADARSAT-2 Synthetic-Aperture Radar Land Cover Segmentation Using Deep Convolutional Neural Networks

Saadati

Pedersoli

Cardinal

et al. 2021

Pattern Recognition. ICPR International Workshops and Challenges

View full text Add to dashboard Cite

Land Cover Classification for Synthetic Aperture Radar Imagery by Using Unsupervised Methods

Cited by 5 publications

References 17 publications

Reviewing the Pertinence of Sentinel-1 SAR for Urban Land Use Land Cover Classification

Reviewing the Pertinence of Sentinel-1 SAR for Urban Land Use Land Cover Classification

A Hierarchical Gamma Mixture Model-Based Method for Classification of High-Dimensional Data

RADARSAT-2 Synthetic-Aperture Radar Land Cover Segmentation Using Deep Convolutional Neural Networks

Contact Info

Product

Resources

About