<inline-formula><tex-math notation="LaTeX">$\mathrm{D}^{3}$</tex-math></inline-formula>T: Deep Denoising Dictionary Tensor for Hyperspectral Anomaly Detection
Hyperspectral anomaly detection (HAD) approaches with tensor low-rank representation (TLRR) have shown engaging performance, which are capable of capturing abundant spectral and spatial information and, hence, extracting sparse anomalies from the low-rank background without ruining the intrinsic geo...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10818590/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832590385108484096 |
|---|---|
| author | Qiangqiang Shen Zonglin Liu Hanzhang Wang Yanhui Xu Yongyong Chen Yongsheng Liang |
| author_facet | Qiangqiang Shen Zonglin Liu Hanzhang Wang Yanhui Xu Yongyong Chen Yongsheng Liang |
| author_sort | Qiangqiang Shen |
| collection | DOAJ |
| description | Hyperspectral anomaly detection (HAD) approaches with tensor low-rank representation (TLRR) have shown engaging performance, which are capable of capturing abundant spectral and spatial information and, hence, extracting sparse anomalies from the low-rank background without ruining the intrinsic geometric structures of tensor data. However, existing HAD approaches construct the low-rank representation by a handcrafted dictionary that still contains the anomalies, resulting in an inferior representation. To this end, we propose a deep denoising dictionary tensor for hyperspectral anomaly detection (<inline-formula><tex-math notation="LaTeX">$\mathrm{D}^{3}$</tex-math></inline-formula>T), which can balance performance and interpretability by fusing two intuitive priors, i.e., low-rankness and deep denoising, into the dictionary tensor and coefficient tensor. In particular, the proposed <inline-formula><tex-math notation="LaTeX">$\mathrm{D}^{3}$</tex-math></inline-formula>T first designs a tensor recovery module for dividing the low-rank background and sparse anomalies, where the background is optimized as a denoising dictionary tensor by the joint low-rankness and deep denoising. Following that, we build the TLRR model based on the denoising dictionary tensor to explore the latent representation of the input hyperspectral images, and hence, each background pixel is represented by the other clear background pixels in place of the anomaly pixels. Meanwhile, the coefficient tensor in TLRR is also optimized by the two priors to fully explore the spatial and spectral correlations in the background, and hence, the anomalies can be accurately detected. Numerous experimental results from various real-world datasets validate the effectiveness of our <inline-formula><tex-math notation="LaTeX">$\mathrm{D}^{3}$</tex-math></inline-formula>T. |
| format | Article |
| id | doaj-art-348b5ac2603f477ea24f2967cb221177 |
| institution | Kabale University |
| issn | 1939-1404 2151-1535 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj-art-348b5ac2603f477ea24f2967cb2211772025-01-24T00:00:58ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-01183713372710.1109/JSTARS.2024.352313310818590<inline-formula><tex-math notation="LaTeX">$\mathrm{D}^{3}$</tex-math></inline-formula>T: Deep Denoising Dictionary Tensor for Hyperspectral Anomaly DetectionQiangqiang Shen0https://orcid.org/0000-0002-3564-6042Zonglin Liu1Hanzhang Wang2https://orcid.org/0009-0006-0650-173XYanhui Xu3https://orcid.org/0000-0002-8615-2122Yongyong Chen4https://orcid.org/0000-0003-1970-1993Yongsheng Liang5https://orcid.org/0000-0002-0891-5577School of Electronics and Information Engineering, Harbin Institute of Technology, Shenzhen, ChinaSchool of Science, Harbin Institute of Technology, Shenzhen, ChinaSchool of Science, Harbin Institute of Technology, Shenzhen, ChinaCollege of Electronic and Optical Engineering, Nanjing University of Posts and Telecommunications, Nanjing, ChinaSchool of Computer Science and Technology, Harbin Institute of Technology, Shenzhen, ChinaCollege of Big Data and Internet, Shenzhen Technology University, Shenzhen, Guangdong, ChinaHyperspectral anomaly detection (HAD) approaches with tensor low-rank representation (TLRR) have shown engaging performance, which are capable of capturing abundant spectral and spatial information and, hence, extracting sparse anomalies from the low-rank background without ruining the intrinsic geometric structures of tensor data. However, existing HAD approaches construct the low-rank representation by a handcrafted dictionary that still contains the anomalies, resulting in an inferior representation. To this end, we propose a deep denoising dictionary tensor for hyperspectral anomaly detection (<inline-formula><tex-math notation="LaTeX">$\mathrm{D}^{3}$</tex-math></inline-formula>T), which can balance performance and interpretability by fusing two intuitive priors, i.e., low-rankness and deep denoising, into the dictionary tensor and coefficient tensor. In particular, the proposed <inline-formula><tex-math notation="LaTeX">$\mathrm{D}^{3}$</tex-math></inline-formula>T first designs a tensor recovery module for dividing the low-rank background and sparse anomalies, where the background is optimized as a denoising dictionary tensor by the joint low-rankness and deep denoising. Following that, we build the TLRR model based on the denoising dictionary tensor to explore the latent representation of the input hyperspectral images, and hence, each background pixel is represented by the other clear background pixels in place of the anomaly pixels. Meanwhile, the coefficient tensor in TLRR is also optimized by the two priors to fully explore the spatial and spectral correlations in the background, and hence, the anomalies can be accurately detected. Numerous experimental results from various real-world datasets validate the effectiveness of our <inline-formula><tex-math notation="LaTeX">$\mathrm{D}^{3}$</tex-math></inline-formula>T.https://ieeexplore.ieee.org/document/10818590/Deep denoisinghyperspectral anomaly detection (HAD)low-rank representation (LRR)low-rank tensor |
| spellingShingle | Qiangqiang Shen Zonglin Liu Hanzhang Wang Yanhui Xu Yongyong Chen Yongsheng Liang <inline-formula><tex-math notation="LaTeX">$\mathrm{D}^{3}$</tex-math></inline-formula>T: Deep Denoising Dictionary Tensor for Hyperspectral Anomaly Detection IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Deep denoising hyperspectral anomaly detection (HAD) low-rank representation (LRR) low-rank tensor |
| title | <inline-formula><tex-math notation="LaTeX">$\mathrm{D}^{3}$</tex-math></inline-formula>T: Deep Denoising Dictionary Tensor for Hyperspectral Anomaly Detection |
| title_full | <inline-formula><tex-math notation="LaTeX">$\mathrm{D}^{3}$</tex-math></inline-formula>T: Deep Denoising Dictionary Tensor for Hyperspectral Anomaly Detection |
| title_fullStr | <inline-formula><tex-math notation="LaTeX">$\mathrm{D}^{3}$</tex-math></inline-formula>T: Deep Denoising Dictionary Tensor for Hyperspectral Anomaly Detection |
| title_full_unstemmed | <inline-formula><tex-math notation="LaTeX">$\mathrm{D}^{3}$</tex-math></inline-formula>T: Deep Denoising Dictionary Tensor for Hyperspectral Anomaly Detection |
| title_short | <inline-formula><tex-math notation="LaTeX">$\mathrm{D}^{3}$</tex-math></inline-formula>T: Deep Denoising Dictionary Tensor for Hyperspectral Anomaly Detection |
| title_sort | inline formula tex math notation latex mathrm d 3 tex math inline formula t deep denoising dictionary tensor for hyperspectral anomaly detection |
| topic | Deep denoising hyperspectral anomaly detection (HAD) low-rank representation (LRR) low-rank tensor |
| url | https://ieeexplore.ieee.org/document/10818590/ |
| work_keys_str_mv | AT qiangqiangshen inlineformulatexmathnotationlatexmathrmd3texmathinlineformulatdeepdenoisingdictionarytensorforhyperspectralanomalydetection AT zonglinliu inlineformulatexmathnotationlatexmathrmd3texmathinlineformulatdeepdenoisingdictionarytensorforhyperspectralanomalydetection AT hanzhangwang inlineformulatexmathnotationlatexmathrmd3texmathinlineformulatdeepdenoisingdictionarytensorforhyperspectralanomalydetection AT yanhuixu inlineformulatexmathnotationlatexmathrmd3texmathinlineformulatdeepdenoisingdictionarytensorforhyperspectralanomalydetection AT yongyongchen inlineformulatexmathnotationlatexmathrmd3texmathinlineformulatdeepdenoisingdictionarytensorforhyperspectralanomalydetection AT yongshengliang inlineformulatexmathnotationlatexmathrmd3texmathinlineformulatdeepdenoisingdictionarytensorforhyperspectralanomalydetection |