COSMICA: A Novel Dataset for Astronomical Object Detection with Evaluation Across Diverse Detection Architectures
Accurate and efficient detection of celestial objects in telescope imagery is a fundamental challenge in both professional and amateur astronomy. Traditional methods often struggle with noise, varying brightness, and object morphology. This paper introduces COSMICA, a novel, curated dataset of manua...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-06-01
|
| Series: | Journal of Imaging |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2313-433X/11/6/184 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839653666953363456 |
|---|---|
| author | Evgenii Piratinskii Irina Rabaev |
| author_facet | Evgenii Piratinskii Irina Rabaev |
| author_sort | Evgenii Piratinskii |
| collection | DOAJ |
| description | Accurate and efficient detection of celestial objects in telescope imagery is a fundamental challenge in both professional and amateur astronomy. Traditional methods often struggle with noise, varying brightness, and object morphology. This paper introduces COSMICA, a novel, curated dataset of manually annotated astronomical images collected from amateur observations. COSMICA enables the development and evaluation of real-time object detection systems intended for practical deployment in observational pipelines. We investigate three modern YOLO architectures, YOLOv8, YOLOv9, and YOLOv11, and two additional object detection models, EfficientDet-Lite0 and MobileNetV3-FasterRCNN-FPN, to assess their performance in detecting comets, galaxies, nebulae, and globular clusters. All models are evaluated using consistent experimental conditions across multiple metrics, including mAP, precision, recall, and inference speed. YOLOv11 demonstrated the highest overall accuracy and computational efficiency, making it a promising candidate for real-world astronomical applications. These results support the feasibility of integrating deep learning-based detection systems into observational astronomy workflows and highlight the importance of domain-specific datasets for training robust AI models. |
| format | Article |
| id | doaj-art-6ceb71aab1ea4a2892cc4e1a33e8f256 |
| institution | Matheson Library |
| issn | 2313-433X |
| language | English |
| publishDate | 2025-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Imaging |
| spelling | doaj-art-6ceb71aab1ea4a2892cc4e1a33e8f2562025-06-25T14:00:25ZengMDPI AGJournal of Imaging2313-433X2025-06-0111618410.3390/jimaging11060184COSMICA: A Novel Dataset for Astronomical Object Detection with Evaluation Across Diverse Detection ArchitecturesEvgenii Piratinskii0Irina Rabaev1Software Engineering Department, Shamoon College of Engineering, 56 Bialik St., Be’er Sheva 8410802, IsraelSoftware Engineering Department, Shamoon College of Engineering, 56 Bialik St., Be’er Sheva 8410802, IsraelAccurate and efficient detection of celestial objects in telescope imagery is a fundamental challenge in both professional and amateur astronomy. Traditional methods often struggle with noise, varying brightness, and object morphology. This paper introduces COSMICA, a novel, curated dataset of manually annotated astronomical images collected from amateur observations. COSMICA enables the development and evaluation of real-time object detection systems intended for practical deployment in observational pipelines. We investigate three modern YOLO architectures, YOLOv8, YOLOv9, and YOLOv11, and two additional object detection models, EfficientDet-Lite0 and MobileNetV3-FasterRCNN-FPN, to assess their performance in detecting comets, galaxies, nebulae, and globular clusters. All models are evaluated using consistent experimental conditions across multiple metrics, including mAP, precision, recall, and inference speed. YOLOv11 demonstrated the highest overall accuracy and computational efficiency, making it a promising candidate for real-world astronomical applications. These results support the feasibility of integrating deep learning-based detection systems into observational astronomy workflows and highlight the importance of domain-specific datasets for training robust AI models.https://www.mdpi.com/2313-433X/11/6/184astronomical object detectiondeep learningYOLO modelstelescope image analysisastronomy dataset |
| spellingShingle | Evgenii Piratinskii Irina Rabaev COSMICA: A Novel Dataset for Astronomical Object Detection with Evaluation Across Diverse Detection Architectures Journal of Imaging astronomical object detection deep learning YOLO models telescope image analysis astronomy dataset |
| title | COSMICA: A Novel Dataset for Astronomical Object Detection with Evaluation Across Diverse Detection Architectures |
| title_full | COSMICA: A Novel Dataset for Astronomical Object Detection with Evaluation Across Diverse Detection Architectures |
| title_fullStr | COSMICA: A Novel Dataset for Astronomical Object Detection with Evaluation Across Diverse Detection Architectures |
| title_full_unstemmed | COSMICA: A Novel Dataset for Astronomical Object Detection with Evaluation Across Diverse Detection Architectures |
| title_short | COSMICA: A Novel Dataset for Astronomical Object Detection with Evaluation Across Diverse Detection Architectures |
| title_sort | cosmica a novel dataset for astronomical object detection with evaluation across diverse detection architectures |
| topic | astronomical object detection deep learning YOLO models telescope image analysis astronomy dataset |
| url | https://www.mdpi.com/2313-433X/11/6/184 |
| work_keys_str_mv | AT evgeniipiratinskii cosmicaanoveldatasetforastronomicalobjectdetectionwithevaluationacrossdiversedetectionarchitectures AT irinarabaev cosmicaanoveldatasetforastronomicalobjectdetectionwithevaluationacrossdiversedetectionarchitectures |