Diffusion Subspace Clustering for Hyperspectral Images
Hyperspectral image (HSI) subspace clustering remains a challenging task due to the poor spatial and rich spectral resolutions of HSIs. Most of the existing HSI subspace clustering approaches just extract the spatial and spectral features, ignoring the intrinsic distribution information of data and...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10179942/ |
| _version_ | 1827895898196148224 |
|---|---|
| author | Jiaxin Chen Shujun Liu Zhongbiao Zhang Huajun Wang |
| author_facet | Jiaxin Chen Shujun Liu Zhongbiao Zhang Huajun Wang |
| author_sort | Jiaxin Chen |
| collection | DOAJ |
| description | Hyperspectral image (HSI) subspace clustering remains a challenging task due to the poor spatial and rich spectral resolutions of HSIs. Most of the existing HSI subspace clustering approaches just extract the spatial and spectral features, ignoring the intrinsic distribution information of data and leading to low accuracy of clustering generally. To solve this problem, this article presents a diffusion subspace clustering model (DiffSC) that learns distribution information of HSI data simultaneously through a diffusion module (DM). Specifically, due to the diffusion probabilistic model (DPM) learning raw object data distribution to generate data of the same distribution, which has received wide attention in generation tasks and outperforms other generative models significantly, we attempt to apply the DPM in the field of feature extraction. DiffSC performs distribution information extraction of HSIs by the DM and fuses them with spatial-spectral features extracted by deep subspace clustering for training jointly. Experiment outcomes demonstrate that intermediate activation of specific timestep in the inverse diffusion process captures latent distribution information of images effectively and improves the HSI clustering accuracy significantly. Since the DPM is simplified, it can be easily trained from scratch. We evaluate the presented DiffSC on five real HSI datasets, and the experiments indicate that DiffSC can obtain the most advanced clustering outcomes that notably outperform most existing HSI subspace clustering approaches. |
| first_indexed | 2024-03-12T22:28:13Z |
| format | Article |
| id | doaj.art-fe14a23ee5634d68989850898ed800b7 |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-03-12T22:28:13Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-fe14a23ee5634d68989850898ed800b72023-07-21T23:00:08ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352023-01-01166517653010.1109/JSTARS.2023.329462310179942Diffusion Subspace Clustering for Hyperspectral ImagesJiaxin Chen0https://orcid.org/0009-0003-3813-2068Shujun Liu1https://orcid.org/0000-0003-4657-1477Zhongbiao Zhang2Huajun Wang3https://orcid.org/0000-0001-5897-5562School of Computer Science and Cyber Security, Chengdu University of Technology, Chengdu, ChinaKey Laboratory of Earth Exploration and Information Techniques of Ministry of Education, Chengdu University of Technology, Chengdu, ChinaKey Laboratory of Earth Exploration and Information Techniques of Ministry of Education, Chengdu University of Technology, Chengdu, ChinaSchool of Mathematics and Physics, Chengdu University of Technology, Chengdu, ChinaHyperspectral image (HSI) subspace clustering remains a challenging task due to the poor spatial and rich spectral resolutions of HSIs. Most of the existing HSI subspace clustering approaches just extract the spatial and spectral features, ignoring the intrinsic distribution information of data and leading to low accuracy of clustering generally. To solve this problem, this article presents a diffusion subspace clustering model (DiffSC) that learns distribution information of HSI data simultaneously through a diffusion module (DM). Specifically, due to the diffusion probabilistic model (DPM) learning raw object data distribution to generate data of the same distribution, which has received wide attention in generation tasks and outperforms other generative models significantly, we attempt to apply the DPM in the field of feature extraction. DiffSC performs distribution information extraction of HSIs by the DM and fuses them with spatial-spectral features extracted by deep subspace clustering for training jointly. Experiment outcomes demonstrate that intermediate activation of specific timestep in the inverse diffusion process captures latent distribution information of images effectively and improves the HSI clustering accuracy significantly. Since the DPM is simplified, it can be easily trained from scratch. We evaluate the presented DiffSC on five real HSI datasets, and the experiments indicate that DiffSC can obtain the most advanced clustering outcomes that notably outperform most existing HSI subspace clustering approaches.https://ieeexplore.ieee.org/document/10179942/Diffusion probabilistic model (DPM)feature extractionhyperspectral image (HSI)representation learningsubspace clustering (SC) |
| spellingShingle | Jiaxin Chen Shujun Liu Zhongbiao Zhang Huajun Wang Diffusion Subspace Clustering for Hyperspectral Images IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Diffusion probabilistic model (DPM) feature extraction hyperspectral image (HSI) representation learning subspace clustering (SC) |
| title | Diffusion Subspace Clustering for Hyperspectral Images |
| title_full | Diffusion Subspace Clustering for Hyperspectral Images |
| title_fullStr | Diffusion Subspace Clustering for Hyperspectral Images |
| title_full_unstemmed | Diffusion Subspace Clustering for Hyperspectral Images |
| title_short | Diffusion Subspace Clustering for Hyperspectral Images |
| title_sort | diffusion subspace clustering for hyperspectral images |
| topic | Diffusion probabilistic model (DPM) feature extraction hyperspectral image (HSI) representation learning subspace clustering (SC) |
| url | https://ieeexplore.ieee.org/document/10179942/ |
| work_keys_str_mv | AT jiaxinchen diffusionsubspaceclusteringforhyperspectralimages AT shujunliu diffusionsubspaceclusteringforhyperspectralimages AT zhongbiaozhang diffusionsubspaceclusteringforhyperspectralimages AT huajunwang diffusionsubspaceclusteringforhyperspectralimages |