Transient Object Detection in the Night Sky Image Using Computer Vision
A constellation is an area on the celestial sphere in which a group of visible stars forms a perceived outline or pattern, typically representing an animal, mythological person or creature, or an inanimate object. We propose an algorithm that detects constellation directly from a wide-field i
2025-06-28 16:36:27 - Adil Khan
Transient Object Detection in the Night Sky Image Using Computer Vision
Project Area of Specialization Artificial IntelligenceProject SummaryA constellation is an area on the celestial sphere in which a group of visible stars forms a perceived outline or pattern, typically representing an animal, mythological person or creature, or an inanimate object.
We propose an algorithm that detects constellation directly from a wide-field image of the night sky. Since, the constellations in the sky have fixed patterns and neighborhood relationships, it is probable to apply a pattern recognition algorithm to detect the real star patterns on the image. The algorithm will be further used to compare source image of the constellation with the new input image to find any celestial object that have passed through it.
This would help us in detecting objects that pass through the constellation and may also lead to discovering objects in space that have yet not been seen. For the project working, we will take a certain constellation in consideration.
Project Objectives- Development of an algorithm that maps stars into a constellation by understanding its dynamics.
- To detect objects that pass a certain constellation using Computer Vision. This object could be an already discovered object or it could lead to a new discovery.
- Implementation of this Transient Object Detection Algorithm to contribute to the Astronomical Domain.
- To eliminate the human inefficiency in comparison of the astronomical data to detect objects in night sky images and to increase throughput and the space of discoveries by the introduction of Artificial Intelligence.
Creation of an algorithm that performs image processing and filtering, such as noise reduction and resolution standardization, on the night sky image so that the accuracy in detecting the constellation and transient object in the night sky image increases.
Applying pattern recognition techniques, to the input night sky image, to detect the constellation pattern and adjusting the input image with respect to the source night sky image present in the database. Furthermore, applying comparison techniques to both the source and input images to find the dissimilarity (Transient Object Detection).
Benefits of the ProjectArtificial intelligence could be used to create information, filling in blind spots in our observations of the Universe and performing a sort of scientific alchemy, helping us turn old knowledge into new. And we’d be able to explore space like never, without even leaving Earth.
The project might become a contribution to the Astronomical Domain as the proposed algorithm might help in the discovery of a new object in the night sky image.
Technical Details of Final DeliverableThe tools that are to be used are:
- Python Language: One of the most frequently used language for data analysis and image processing.
- Computer Vision Libraries (OpenCV, scikit-image etc)
The deliverables,
- Astronomical, Computational knowledge alongside filtered Data.
- Pattern Matching Algorithm
- Comparison Algorithm to detect transient object
- Accuracy of the model
- Algorithm to detect objects in the vicinity of our specific constellation
| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 71000 | |||
| GPU | Equipment | 1 | 70000 | 70000 |
| Printing | Miscellaneous | 1 | 1000 | 1000 |