BCDDM: Branch Correction Denoising Diffusion Model for Black Hole Image Generation
The properties of black holes and accretion flows can be inferred by fitting Event Horizon Telescope data to simulated images generated through general relativistic ray tracing (GRRT). However, due to the computationally intensive nature of GRRT, the efficiency of generating specific radiation flux...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IOP Publishing
2025-01-01
|
| Series: | The Astrophysical Journal Supplement Series |
| Subjects: | |
| Online Access: | https://doi.org/10.3847/1538-4365/add896 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849422834038734848 |
|---|---|
| author | Ao Liu Zelin Zhang Songbai Chen Cuihong Wen Jieci Wang |
| author_facet | Ao Liu Zelin Zhang Songbai Chen Cuihong Wen Jieci Wang |
| author_sort | Ao Liu |
| collection | DOAJ |
| description | The properties of black holes and accretion flows can be inferred by fitting Event Horizon Telescope data to simulated images generated through general relativistic ray tracing (GRRT). However, due to the computationally intensive nature of GRRT, the efficiency of generating specific radiation flux images needs to be improved. This paper introduces the Branch Correction Denoising Diffusion Model (BCDDM), a deep learning framework that synthesizes black hole images directly from physical parameters. The model incorporates a branch correction mechanism and a weighted mixed-loss function to enhance accuracy and stability. We have constructed a data set of 2157 GRRT-simulated images for training the BCDDM, which spans seven key physical parameters of the radiatively inefficient accretion flow model. Our experiments show a strong correlation between the generated images and their physical parameters. By enhancing the GRRT data set with BCDDM-generated images and using ResNet50 for parameter regression, we achieve significant improvements in parameter prediction performance. BCDDM offers a novel approach to reducing the computational costs of black hole image generation, providing a faster and more efficient pathway for data set augmentation, parameter estimation, and model fitting. |
| format | Article |
| id | doaj-art-06fd8f6a8f6746ae8a8516467094e9a7 |
| institution | Kabale University |
| issn | 0067-0049 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IOP Publishing |
| record_format | Article |
| series | The Astrophysical Journal Supplement Series |
| spelling | doaj-art-06fd8f6a8f6746ae8a8516467094e9a72025-08-20T03:30:55ZengIOP PublishingThe Astrophysical Journal Supplement Series0067-00492025-01-0127911010.3847/1538-4365/add896BCDDM: Branch Correction Denoising Diffusion Model for Black Hole Image GenerationAo Liu0Zelin Zhang1Songbai Chen2https://orcid.org/0000-0001-5531-9113Cuihong Wen3https://orcid.org/0000-0003-2668-4503Jieci Wang4https://orcid.org/0000-0001-5072-3096College of Information Science and Engineering, Hunan Normal University , Changsha, 410081, People’s Republic of China ; cuihongwen@hunnu.edu.cnDepartment of Physics , Institute of Interdisciplinary Studies, Key Laboratory of Low Dimensional Quantum Structures and Quantum Control of Ministry of Education, Synergetic Innovation Center for Quantum Effects and Applications, Hunan Normal University, Changsha, Hunan 410081, People’s Republic of China ; csb3752@hunnu.edu.cn, jieciwang@hunnu.edu.cnDepartment of Physics , Institute of Interdisciplinary Studies, Key Laboratory of Low Dimensional Quantum Structures and Quantum Control of Ministry of Education, Synergetic Innovation Center for Quantum Effects and Applications, Hunan Normal University, Changsha, Hunan 410081, People’s Republic of China ; csb3752@hunnu.edu.cn, jieciwang@hunnu.edu.cn; Center for Gravitation and Cosmology , College of Physical Science and Technology, Yangzhou University, Yangzhou 225009, People’s Republic of ChinaCollege of Information Science and Engineering, Hunan Normal University , Changsha, 410081, People’s Republic of China ; cuihongwen@hunnu.edu.cnDepartment of Physics , Institute of Interdisciplinary Studies, Key Laboratory of Low Dimensional Quantum Structures and Quantum Control of Ministry of Education, Synergetic Innovation Center for Quantum Effects and Applications, Hunan Normal University, Changsha, Hunan 410081, People’s Republic of China ; csb3752@hunnu.edu.cn, jieciwang@hunnu.edu.cnThe properties of black holes and accretion flows can be inferred by fitting Event Horizon Telescope data to simulated images generated through general relativistic ray tracing (GRRT). However, due to the computationally intensive nature of GRRT, the efficiency of generating specific radiation flux images needs to be improved. This paper introduces the Branch Correction Denoising Diffusion Model (BCDDM), a deep learning framework that synthesizes black hole images directly from physical parameters. The model incorporates a branch correction mechanism and a weighted mixed-loss function to enhance accuracy and stability. We have constructed a data set of 2157 GRRT-simulated images for training the BCDDM, which spans seven key physical parameters of the radiatively inefficient accretion flow model. Our experiments show a strong correlation between the generated images and their physical parameters. By enhancing the GRRT data set with BCDDM-generated images and using ResNet50 for parameter regression, we achieve significant improvements in parameter prediction performance. BCDDM offers a novel approach to reducing the computational costs of black hole image generation, providing a faster and more efficient pathway for data set augmentation, parameter estimation, and model fitting.https://doi.org/10.3847/1538-4365/add896Astrophysical black holesNeural networksAstronomy image processingGeneral relativity |
| spellingShingle | Ao Liu Zelin Zhang Songbai Chen Cuihong Wen Jieci Wang BCDDM: Branch Correction Denoising Diffusion Model for Black Hole Image Generation The Astrophysical Journal Supplement Series Astrophysical black holes Neural networks Astronomy image processing General relativity |
| title | BCDDM: Branch Correction Denoising Diffusion Model for Black Hole Image Generation |
| title_full | BCDDM: Branch Correction Denoising Diffusion Model for Black Hole Image Generation |
| title_fullStr | BCDDM: Branch Correction Denoising Diffusion Model for Black Hole Image Generation |
| title_full_unstemmed | BCDDM: Branch Correction Denoising Diffusion Model for Black Hole Image Generation |
| title_short | BCDDM: Branch Correction Denoising Diffusion Model for Black Hole Image Generation |
| title_sort | bcddm branch correction denoising diffusion model for black hole image generation |
| topic | Astrophysical black holes Neural networks Astronomy image processing General relativity |
| url | https://doi.org/10.3847/1538-4365/add896 |
| work_keys_str_mv | AT aoliu bcddmbranchcorrectiondenoisingdiffusionmodelforblackholeimagegeneration AT zelinzhang bcddmbranchcorrectiondenoisingdiffusionmodelforblackholeimagegeneration AT songbaichen bcddmbranchcorrectiondenoisingdiffusionmodelforblackholeimagegeneration AT cuihongwen bcddmbranchcorrectiondenoisingdiffusionmodelforblackholeimagegeneration AT jieciwang bcddmbranchcorrectiondenoisingdiffusionmodelforblackholeimagegeneration |