Image processing tool and algorithm for star constellation detection from the image.
We had used list of constellations published by International Astronomical Union (IAU), to maintain a standard convention. We have used this th
A constellation is a recognized pattern of stars in the night sky. There are a total of 88 known constellations. During the ancient times, certain constellations had acquired special significance over time because of their appearance marking the start of a new season, guiding travelers and people, letting farmers know when to sow or reap a crop, and hunters to tell the best time to hunt. They also played a part in being the first GPS, and still play an important role in satellite placement and positioning. A list of constellations has now been published by International Astronomical Union (IAU). While identifying constellations is not very difficult for people who are experience in the activity of stargazing, it can be difficult for newcomers to explore the numerous patterns and the corresponding stars involved in the making of those patters. We try to apply our knowledge of various image processing techniques and algorithms to help in the detection of these constellations in the night sky. Our main focus is to build a detection tool for naked eye
constellations. The main detection tool are the templates we use, against which a test is compared and try to figure out the constellation in that test image. Along with processing the test images, the templates had to be processed as well to make them usable. There were three important steps in our constellation detection algorithm. The first was the creation of the template database. The original templates were obtained from an image gallery of constellations [3]. We select the 30 largest and most prominent constellations present in the night sky using a list available online [4]. Various image processing techniques are the applied to create the template database which is used in detection algorithm. The template creation is implemented based on an existing set of modified constellation charts [5]. The second was the processing of test images. The test images were obtained from a night sky observation application [6] and then processed to make detection of constellations feasible. The third was creation of a detection algorithm, that would let us detect a constellation irrespective of the way it was present in the image. The performed procedures and techniques are discussed in detail in the following sections.
To use this project, first clone this git repository using following commands
git init
git clone https://github.com/Kartikeya18153/Constellation-detection.git
Then comes the, installation of dependences
I have use following dependences
- numpy
- matplotlib
- cv2
- math
- os
- pickle
- copy
for installing all the dependencies, use requirement file with following command
pip3 install -r requirment.txt
You need to run the DIP_Project.py. It will crunch all the templates and test data to form Normalised immages for genrating results.
See also the list of contributors who participated in this project.
This project is licensed under the MIT License - see the LICENSE.md file for details
- If you want to contribute, fork the repository for yourself. Check out Fork list
- If you liked the project, leave a star. Check out stargazers
TODO :
- delete the unrequired file
- All images in main directory (like final.png, test.png)
- Report doc file