Object
Detection
Celestial bodies
TEAM members and their tasks
Sarthak
Kesarwani : Maintainance of
LogBook
Agrim Kumar : Video Editing
Om Tomar : Created the model
Introduction to Object
Detection
Definition: Object detection is the task of
identifying and locating objects within an
image or video.
Common Use Cases:
Surveillance systems
Self-driving cars
Medical imaging
In the Context of Celestial Bodies:
Detecting and classifying objects like stars,
planets, comets, asteroids, and galaxies in
space images.
Importance of Celestial Object
Detection
Title: Why Detect Celestial Objects?
Content:
Astronomy Research: Helps in discovering new
celestial bodies (e.g., exoplanets, asteroids).
Space Exploration: Detecting objects for space
missions (NASA’s exploration of Mars, detecting
satellites).
Astronomical Surveys: Identifying and
categorizing galaxies, stars, and nebulae for
cataloging.
Space Debris Monitoring: Detecting space
debris to ensure the safety of spacecraft.
Technologies Used in Object
Detection
Title: Technologies Used in Object Detection for Celestial
Bodies
Content:
Convolutional Neural Networks (CNNs):
Common for image classification and object detection tasks.
Used for analyzing images of celestial objects.
Deep Learning Algorithms:
YOLO (You Only Look Once)
SSD (Single Shot Multibox Detector)
Faster R-CNN (Region-based CNN)
Optical Telescopes & Space Telescopes:
Hubble Space Telescope
James Webb Space Telescope (JWST)
Challenges in Celestial Object
Detection
Noise in Data: Data from space telescopes can
contain noise, making object detection difficult.
Low Resolution: Images captured from distant
celestial bodies may have low resolution,
complicating detection.
Large Data Sets: The amount of data from
astronomical surveys is vast, making analysis
challenging.
Obscured Objects: Objects hidden by cosmic
dust or within dense galactic regions may be
difficult to detect.
Examples of Object Detection
in Space
Asteroid Detection:
AI used to analyze images from space
telescopes to detect asteroids.
Example: NEOWISE mission by NASA.
Galaxy Classification:
Deep learning models classify galaxies into
different types based on their features.
Example: Galaxy Zoo project.
Exoplanet Discovery:
Detecting exoplanets via the transit method
(e.g., Kepler Space Telescope data).
Case Study: Detecting
Asteroids Using AI
Problem: Asteroids pose a risk to Earth, and we
need to track them.
Solution: Machine learning models trained on
historical data of asteroid images.
Approach:
Use CNNs to analyze images from telescopes.
Training models on known asteroid positions and
features.
Results: AI algorithms can detect asteroids in
real-time, improving early warning systems.
Deep Learning Models in
Action
Model Comparison: YOLO vs Faster R-CNN
for real-time celestial object detection.
Example Application:
YOLO: Used for detecting various celestial
objects in real-time from telescope images.
Faster R-CNN: More accurate for detecting
galaxies and stars but slower.
Future of Object Detection in
Astronomy
Advancements in AI: More powerful models for
higher accuracy and faster processing.
Increased Space Missions: Growing data from
missions like James Webb Telescope, more data
for AI models.
Automated Space Exploration: AI-driven
spacecraft that can autonomously detect and
analyze celestial objects.
Global Collaboration: Sharing astronomical
data between research institutions to improve
models.
Conclusion
Object detection based on celestial bodies
is crucial for astronomy, space exploration,
and understanding the universe.
Deep learning models like CNNs and YOLO
are revolutionizing how we detect and
analyze celestial objects.
Despite challenges, continued
advancements in technology will make
object detection more accurate,
automated, and timely.
References
List of academic papers, websites, and
resources used in the presentation.
