Hyperspectral Image Classification With Multi-Attention Transformer and Adaptive Superpixel Segmentation-Based Active Learning
The DOI matches its Crossref record; 61 of the 61 references it cites were found in a registry or an index.
Abstract
Deep learning (DL) based methods represented by convolutional neural networks (CNNs) are widely used in hyperspectral image classification (HSIC). Some of these methods have strong ability to extract local information, but the extraction of long-range features is slightly inefficient, while others are just the opposite. For example, limited by the receptive fields, CNN is difficult to capture the contextual spectral-spatial features from a long-range spectral-spatial relationship. Besides, the success of DL-based methods is greatly attributed to numerous labeled samples, whose acquisition are time-consuming and cost-consuming. To resolve these problems, a hyperspectral classification framework based on multi-attention Transformer (MAT) and adaptive superpixel segmentation-based active learning (MAT-ASSAL) is proposed, which successfully achieves excellent classification performance, especially under the condition of small-size samples. Firstly, a multi-attention Transformer network is built for HSIC. Specifically, the self-attention module of Transformer is applied to model long-range contextual dependency between spectral-spatial embedding. Moreover, in order to capture local features, an outlook-attention module which can efficiently encode fine-level features and contexts into tokens is utilized to improve the correlation between the center spectral-spatial embedding and its surroundings. Secondly, aiming to train a excellent MAT model through limited labeled samples, a novel active learning (AL) based on superpixel segmentation is proposed to select important samples for MAT. Finally, to better integrate local spatial similarity into active learning, an adaptive superpixel (SP) segmentation algorithm, which can save SPs in uninformative regions and preserve edge details in complex regions, is employed to generate better local spatial constraints for AL. Quantitative and qualitative results indicate that the MAT-ASSAL outperforms seven state-of-the-art methods on three HSI datasets.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Exploiting Generative AI to Scale up Intelligent Tutoring Systems
Jakubův, J., Chvalovský, K., Goertzel, Z., Kaliszyk, C., Olšák, M., Piotrowski, B., Schulz, S., Suda, M., & Urban, J. (2023) · DROPS (Schloss Dagstuhl – Leibniz Center for Informatics) · doi:10.4230/lipics.itp.2023.19 · Cited by 76,658
Fully convolutional networks for semantic segmentation
Long, J., Shelhamer, E., & Darrell, T. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298965 · Cited by 38,834
Classification of hyperspectral remote sensing images with support vector machines
Melgani, F., & Bruzzone, L. (2004) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2004.831865 · Cited by 4,411 · cited 7×
“Consequently, although the classification accuracy of SVM is low, SVM retains the clear edge of an HSI due to the reason that SVM can process each pixel of hyperspectral image directly.”
BackgroundMethodInfluentialHaboudane, D. (2004) · Remote Sensing of Environment · doi:10.1016/j.rse.2003.12.013 · Cited by 2,608 · cited 1×
“…[4], a hot research field which assigns each pixel with predefined labels by using some manually labeled samples [5], is widely used in agriculture [6], ecological monitoring [7], national defense [8], urban construction [9], and et al. Initially, the traditional machine learning (ML) methods that…”
BackgroundZhong, Z., Li, J., Luo, Z., & Chapman, M. (2018) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2017.2755542 · Cited by 1,959 · cited 7×
“In general, DBDA and A 2 S 2 K-ResNet can achieve good performance, but in some complex regions, such as the right region of houston, their results are not satisfactory.”
MethodInfluentialHybridSN: Exploring 3-D–2-D CNN Feature Hierarchy for Hyperspectral Image Classification
Roy, S. K., Krishna, G., Dubey, S. R., & Chaudhuri, B. B. (2020) · IEEE Geoscience and Remote Sensing Letters · doi:10.1109/lgrs.2019.2918719 · Cited by 1,898
Harsanyi, J. C., & Chang, C.-I. (1994) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/36.298007 · Cited by 1,517 · cited 1×
“In recent years, HSI classification (HSIC) [3], [4], a hot research field which assigns each pixel with predefined labels by using some manually labeled samples [5], is widely used in agriculture [6], ecological monitoring [7], national defense [8], urban construction [9], and et al. Initially, the…”
BackgroundSpectralFormer: Rethinking Hyperspectral Image Classification With Transformers
Hong, D., Han, Z., Yao, J., Gao, L., Zhang, B., Plaza, A., & Chanussot, J. (2022) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2021.3130716 · Cited by 1,382 · cited 2×
“Motivated by ViT, a SpectralFormer [33] which splits the spectral bands of HSIs into multiple groups (tokens) is proposed for HSIC.”
BackgroundGoing deeper with Image Transformers
Touvron, H., Cord, M., Sablayrolles, A., Synnaeve, G., & Jegou, H. (2021) · 2021 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00010 · Cited by 1,259 · cited 2×
“The goal of Cl_A [64] is to avoid conflicts that may arise from inconsistent learning objectives between patch embeddings and the class token, so the Cl_A freezes patch embeddings and only updates the class token.”
BackgroundMethodSpectral and Spatial Classification of Hyperspectral Data Using SVMs and Morphological Profiles
Fauvel, M., Benediktsson, J. A., Chanussot, J., & Sveinsson, J. R. (2008) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2008.922034 · Cited by 990 · cited 1×
“[3], [4], a hot research field which assigns each pixel with”
Background
Cited by
Tracing unknown tumor origins with a biological-pathway-based transformer model
Xie, J., Chen, Y., Luo, S., Yang, W., Lin, Y., Wang, L., Ding, X., Tong, M., & Yu, R. (2024) · Cell Reports Methods · doi:10.1016/j.crmeth.2024.100797 · Cited by 9
Metadata
- Authors
- Zhao, C., Qin, B., Feng, S., Zhu, W., Sun, W., Li, W., & Jia, X.
- Venue
- IEEE Transactions on Image Processing
- Published
- 2023
- Volume
- 32
- Pages
- 3606-3621
- ISSN
- 1057-7149 (ISSN-L)
- 1941-0042
- OpenAlex
- W4382203304
- Cited by
- 190
- References
- 61
- FWCI
- 20.98
- Type
- journal article
- Language
- en
- Open access
- closed
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.