☆ 3.8 Proceedings Paper

OPEN-SET CLASSIFICATION IN REMOTE SENSING IMAGERY WITH ENERGY-BASED VISION TRANSFORMER

2022 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS 2022) (2022)

Related references

Note: Only part of the references are listed.

Review Environmental Sciences

Land use/land cover in view of earth observation: data sources, input dimensions, and classifiers-a review of the state of the art

Prem Chandra Pandey et al.

Summary: The concept of land use/land cover (LULC) is crucially linked to various aspects of the human and physical environment, and Earth observation (EO) technology has greatly enhanced our ability to classify land cover with high spatial and spectral resolution imagery. Increasing use of EO sensors has opened up new applications in different disciplines, emphasizing the importance of understanding the spatial-spectral-temporal characteristics of satellite data and classification approaches. Research in this field will need to focus on integrating techniques used in LULC mapping to gain a more comprehensive understanding at regional or global scales.

GEOCARTO INTERNATIONAL (2021)

Add to Collection

Article Geochemistry & Geophysics

Few-Shot Hyperspectral Image Classification With Unknown Classes Using Multitask Deep Learning

Shengjie Liu et al.

Summary: Current hyperspectral image classification methods often overlook unknown classes in the unseen data, leading to potential overestimation of known land covers. A proposed multitask deep learning method, MDL4OW, addresses this issue by conducting classification and reconstruction simultaneously in an open world scenario where unknown classes may exist. The method achieved state-of-the-art results on real-world hyperspectral images, particularly improving accuracy under few-shot conditions.

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING (2021)

Add to Collection

Article Geography, Physical

Robust deep alignment network with remote sensing knowledge graph for zero-shot and generalized zero-shot remote sensing image scene classification

Yansheng Li et al.

Summary: This paper constructs a new remote sensing knowledge graph (RSKG) to support the inference recognition of unseen RS image scenes; proposes to generate a Semantic Representation of scene categories by representation learning of RSKG (SR-RSKG); and introduces a novel deep alignment network (DAN) for zero-shot and generalized zero-shot RS image scene classification.

ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING (2021)

Add to Collection

Article Environmental Sciences

Deep Open-Set Domain Adaptation for Cross-Scene Classification based on Adversarial Learning and Pareto Ranking

Reham Adayel et al.

REMOTE SENSING (2020)

Add to Collection

Article Geochemistry & Geophysics

Scene Classification With Recurrent Attention of VHR Remote Sensing Images

Qi Wang et al.

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING (2019)

Add to Collection

Proceedings Paper Computer Science, Interdisciplinary Applications

Open Set SAR Target Classification

Anne Pavy et al.

ALGORITHMS FOR SYNTHETIC APERTURE RADAR IMAGERY XXVI (2019)

Add to Collection

Proceedings Paper Computer Science, Information Systems

Gabor-Filtering-Based Completed Local Binary Patterns for Land-Use Scene Classification

Chen Chen et al.

2015 1ST IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA BIG DATA (BIGMM) (2015)

Add to Collection

OPEN-SET CLASSIFICATION IN REMOTE SENSING IMAGERY WITH ENERGY-BASED VISION TRANSFORMER

Related references

Land use/land cover in view of earth observation: data sources, input dimensions, and classifiers-a review of the state of the art

Few-Shot Hyperspectral Image Classification With Unknown Classes Using Multitask Deep Learning

Robust deep alignment network with remote sensing knowledge graph for zero-shot and generalized zero-shot remote sensing image scene classification

Deep Open-Set Domain Adaptation for Cross-Scene Classification based on Adversarial Learning and Pareto Ranking

Scene Classification With Recurrent Attention of VHR Remote Sensing Images

Open Set SAR Target Classification

Gabor-Filtering-Based Completed Local Binary Patterns for Land-Use Scene Classification

Export Citation

Share Paper