Image Preprocessing Framework for Time-domain Astronomy in the Artificial Intelligence Era
The rapid advancement of image analysis methods in time-domain astronomy, particularly those leveraging artificial intelligence (AI) algorithms, has highlighted efficient image preprocessing as a critical bottleneck affecting algorithm performance. Image preprocessing, which involves standardizing i...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IOP Publishing
2025-01-01
|
| Series: | The Astronomical Journal |
| Subjects: | |
| Online Access: | https://doi.org/10.3847/1538-3881/adb842 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849395064862670848 |
|---|---|
| author | Liang Cao Peng Jia Jiaxin Li Yu Song Chengkun Hou Yushan Li |
| author_facet | Liang Cao Peng Jia Jiaxin Li Yu Song Chengkun Hou Yushan Li |
| author_sort | Liang Cao |
| collection | DOAJ |
| description | The rapid advancement of image analysis methods in time-domain astronomy, particularly those leveraging artificial intelligence (AI) algorithms, has highlighted efficient image preprocessing as a critical bottleneck affecting algorithm performance. Image preprocessing, which involves standardizing images for training or deployment of various AI algorithms, encompasses essential steps such as image quality evaluation, alignment, stacking, background extraction, gray-scale transformation, cropping, source detection, astrometry, and photometry. Historically, these algorithms were developed independently by different research groups, primarily based on central processing unit (CPU) architecture for small-scale data processing. This paper introduces a novel framework for image preprocessing that integrates key algorithms specifically modified for graphics processing unit architecture, enabling large-scale image preprocessing for different algorithms. To prepare for the new algorithm design paradigm in the AI era, we have implemented two operational modes in the framework for different application scenarios: eager mode and pipeline mode. The Eager mode facilitates real-time feedback and flexible adjustments, which could be used for parameter tuning and algorithm development. The pipeline mode is primarily designed for large-scale data processing, which could be used for training or deploying of AI models. We have tested the performance of our framework using simulated and real observation images. Results demonstrate that our framework significantly enhances image preprocessing speed while maintaining accuracy levels comparable to CPU-based algorithms. To promote accessibility and ease of use, a Docker version of our framework is available for download in the PaperData Repository powered by China-VO, compatible with various AI algorithms developed for time-domain astronomy research. |
| format | Article |
| id | doaj-art-e29a1a7c73d8451487291ac78eb7fea4 |
| institution | Kabale University |
| issn | 1538-3881 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IOP Publishing |
| record_format | Article |
| series | The Astronomical Journal |
| spelling | doaj-art-e29a1a7c73d8451487291ac78eb7fea42025-08-20T03:39:45ZengIOP PublishingThe Astronomical Journal1538-38812025-01-01169421510.3847/1538-3881/adb842Image Preprocessing Framework for Time-domain Astronomy in the Artificial Intelligence EraLiang Cao0Peng Jia1https://orcid.org/0000-0001-6623-0931Jiaxin Li2Yu Song3Chengkun Hou4Yushan Li5College of Physics and Optoelectronics, Taiyuan University of Technology , Taiyuan, 030024, People’s Republic of China ; robinmartin20@gmail.comCollege of Physics and Optoelectronics, Taiyuan University of Technology , Taiyuan, 030024, People’s Republic of China ; robinmartin20@gmail.comCollege of Physics and Optoelectronics, Taiyuan University of Technology , Taiyuan, 030024, People’s Republic of China ; robinmartin20@gmail.comCollege of Physics and Optoelectronics, Taiyuan University of Technology , Taiyuan, 030024, People’s Republic of China ; robinmartin20@gmail.comCollege of Physics and Optoelectronics, Taiyuan University of Technology , Taiyuan, 030024, People’s Republic of China ; robinmartin20@gmail.comCollege of Physics and Optoelectronics, Taiyuan University of Technology , Taiyuan, 030024, People’s Republic of China ; robinmartin20@gmail.comThe rapid advancement of image analysis methods in time-domain astronomy, particularly those leveraging artificial intelligence (AI) algorithms, has highlighted efficient image preprocessing as a critical bottleneck affecting algorithm performance. Image preprocessing, which involves standardizing images for training or deployment of various AI algorithms, encompasses essential steps such as image quality evaluation, alignment, stacking, background extraction, gray-scale transformation, cropping, source detection, astrometry, and photometry. Historically, these algorithms were developed independently by different research groups, primarily based on central processing unit (CPU) architecture for small-scale data processing. This paper introduces a novel framework for image preprocessing that integrates key algorithms specifically modified for graphics processing unit architecture, enabling large-scale image preprocessing for different algorithms. To prepare for the new algorithm design paradigm in the AI era, we have implemented two operational modes in the framework for different application scenarios: eager mode and pipeline mode. The Eager mode facilitates real-time feedback and flexible adjustments, which could be used for parameter tuning and algorithm development. The pipeline mode is primarily designed for large-scale data processing, which could be used for training or deploying of AI models. We have tested the performance of our framework using simulated and real observation images. Results demonstrate that our framework significantly enhances image preprocessing speed while maintaining accuracy levels comparable to CPU-based algorithms. To promote accessibility and ease of use, a Docker version of our framework is available for download in the PaperData Repository powered by China-VO, compatible with various AI algorithms developed for time-domain astronomy research.https://doi.org/10.3847/1538-3881/adb842GPU computingTime domain astronomyAstronomical techniques |
| spellingShingle | Liang Cao Peng Jia Jiaxin Li Yu Song Chengkun Hou Yushan Li Image Preprocessing Framework for Time-domain Astronomy in the Artificial Intelligence Era The Astronomical Journal GPU computing Time domain astronomy Astronomical techniques |
| title | Image Preprocessing Framework for Time-domain Astronomy in the Artificial Intelligence Era |
| title_full | Image Preprocessing Framework for Time-domain Astronomy in the Artificial Intelligence Era |
| title_fullStr | Image Preprocessing Framework for Time-domain Astronomy in the Artificial Intelligence Era |
| title_full_unstemmed | Image Preprocessing Framework for Time-domain Astronomy in the Artificial Intelligence Era |
| title_short | Image Preprocessing Framework for Time-domain Astronomy in the Artificial Intelligence Era |
| title_sort | image preprocessing framework for time domain astronomy in the artificial intelligence era |
| topic | GPU computing Time domain astronomy Astronomical techniques |
| url | https://doi.org/10.3847/1538-3881/adb842 |
| work_keys_str_mv | AT liangcao imagepreprocessingframeworkfortimedomainastronomyintheartificialintelligenceera AT pengjia imagepreprocessingframeworkfortimedomainastronomyintheartificialintelligenceera AT jiaxinli imagepreprocessingframeworkfortimedomainastronomyintheartificialintelligenceera AT yusong imagepreprocessingframeworkfortimedomainastronomyintheartificialintelligenceera AT chengkunhou imagepreprocessingframeworkfortimedomainastronomyintheartificialintelligenceera AT yushanli imagepreprocessingframeworkfortimedomainastronomyintheartificialintelligenceera |