Earth Observation Semantic Data Mining: Latent Dirichlet Allocation-Based Approach

Asiyabi, Reza Mohammadi; Datcu, Mihai

doi:10.1109/jstars.2022.3159277

Cited by 7 publications

(6 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the developed dataset, four constructed semantic classes have very similar features, which makes the classification of those classes very challenging. The similarities are also apparent in the LDA-based data mining results in [55]. Many of the constructed patches consist of also vegetation cover but the majority of the covered area is considered as the semantic label of the patch.…”

Section: S1slc_cvdl Annotated Datasetmentioning

confidence: 85%

See 1 more Smart Citation

Complex-Valued End-to-End Deep Network With Coherency Preservation for Complex-Valued SAR Data Reconstruction and Classification

Asiyabi

Datcu

Anghel

et al. 2023

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

Deep learning models have achieved remarkable success in many different fields and attracted many interests. Several researchers attempted to apply deep learning models to Synthetic Aperture Radar (SAR) data processing, but it did not have the same breakthrough as the other fields, including optical remote sensing. SAR data are in complex domain by nature and processing them with Real-Valued (RV) networks neglects the phase component which conveys important and distinctive information. A Complex-Valued (CV) end-to-end deep network is developed in this study for the reconstruction and classification of CV-SAR data. Azimuth subaperture decomposition is utilized to incorporate physics-aware attributes of the CV-SAR into the deep model. Moreover, the correlation coefficient amplitude (Coherence) of the CV-SAR images depends on the SAR system characteristics and physical properties of the target. This coherency should be considered and preserved in the processing chain of the CV-SAR data. The coherency preservation of the CV deep networks for CV-SAR images, which is mostly neglected in the literature, is evaluated in this study. Furthermore, a large-scale CV-SAR annotated dataset for the evaluation of the CV deep networks is lacking. A semantically annotated CV-SAR dataset from Sentinel-1 Single Look Complex StripMap mode data (S1SLC_CVDL dataset) is developed and introduced in this study. The experimental analysis demonstrated the better performance of the developed CV deep network for CV-SAR data classification and reconstruction in comparison to the equivalent RV model and more complicated RV architectures, as well as its coherency preservation and physics-aware capability.

show abstract

Section: S1slc_cvdl Annotated Datasetmentioning

confidence: 85%

“…In order to remove the misclassified patches and improve the quality of the annotation in the dataset, an LDA-based semantic data mining technique is applied to the initially classified maps. For more details on the LDA-based semantic data mining, refer to [55].…”

Section: S1slc_cvdl Annotated Datasetmentioning

confidence: 99%

Complex-Valued End-to-End Deep Network With Coherency Preservation for Complex-Valued SAR Data Reconstruction and Classification

Asiyabi

Datcu

Anghel

et al. 2023

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

show abstract

“…At work [13] the authors used three different scenarios to evaluate the discovery of semantic information in various remote sensing applications, including optical and synthetic aperture radar data at different spatial resolutions. Recently, in [14] explored the structure of social discourse on aging in Korea by analyzing newspaper articles on aging.…”

Section: Related Literaturementioning

confidence: 99%

Search and classify topics in a corpus of text using the latent dirichlet allocation model

Iparraguirre-Villanueva¹,

Sierra-Liñan

Salazar³

et al. 2023

IJEECS

View full text Add to dashboard Cite

This work aims at discovering topics in a text corpus and classifying the most relevant terms for each of the discovered topics. The process was performed in four steps: first, document extraction and data processing; second, labeling and training of the data; third, labeling of the unseen data; and fourth, evaluation of the model performance. For processing, a total of 10,322 "curriculum" documents related to data science were collected from the web during 2018-2022. The latent dirichlet allocation (LDA) model was used for the analysis and structure of the subjects. After processing, 12 themes were generated, which allowed ranking the most relevant terms to identify the skills of each of the candidates. This work concludes that candidates interested in data science must have skills in the following topics: first, they must be technical, they must have mastery of structured query language, mastery of programming languages such as R, Python, java, and data management, among other tools associated with the technology.

show abstract

“…Three StripMap (SM) SLC dual polarization (HH/HV) Sentinel-1 SAR scenes, acquired over Chicago and Huston, USA, and Sao Paulo, Brazil, are selected to consider different landcovers (e.g., various constructed areas, vegetation, agriculture, and water bodies) characterized by diverse data dynamic ranges [18]. No further preprocessing is applied on the SAR data.…”

Section: Datasetmentioning

confidence: 99%

Complex-Valued Autoencoder for Multi-Polarization SLC SAR Data Compression with Side Information

Asiyabi,

Anghel,

Rizzoli

et al. 2023

IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium

Self Cite

View full text Add to dashboard Cite

Recent advances in Synthetic Aperture Radar (SAR) sensors have enabled the acquisition of very high-resolution images with wide swaths, large bandwidth and in multiple polarization channels. As a result of the significant increase of SAR data size, an effective compression of the acquired data is of paramount importance. However, conventional data compression methods demonstrate limited effectiveness when applied to SAR data. In order to tackle this problem, in this study, a Complex-Valued (CV) end-to-end deep learning-based architecture based on convolutional autoencoders is proposed to compress Single Look Complex (SLC) SAR data. By relying on dual polarization SAR data, one of the polarization channels of the data is used as the side information to assist the reconstruction of the compressed channel with lower data loss. The obtained results demonstrate the remarkable potential and capability of CV deep learning-based methods for SAR data compression.

show abstract

Earth Observation Semantic Data Mining: Latent Dirichlet Allocation-Based Approach

Cited by 7 publications

References 41 publications

Complex-Valued End-to-End Deep Network With Coherency Preservation for Complex-Valued SAR Data Reconstruction and Classification

Complex-Valued End-to-End Deep Network With Coherency Preservation for Complex-Valued SAR Data Reconstruction and Classification

Search and classify topics in a corpus of text using the latent dirichlet allocation model

Complex-Valued Autoencoder for Multi-Polarization SLC SAR Data Compression with Side Information

Contact Info

Product

Resources

About