Star Matching

A python implementation of a star matching algorithm for finding corresponding stars in two or more images.

Features

• Uses the Hough Circle Transform to detect stars in an image
• Implements RANSAC to match corresponding stars between two images
• Draws lines connecting the matched stars in the two images

Usage

To use the code, you need to provide the path to two images as input to the StarMatcher class. The code will then return a list of matched stars and also create an image with lines connecting the matched stars.

Requirements

The code requires the following libraries:

• OpenCV
• Numpy
• scikit-image

To install the dependencies, run the following commands in your terminal: pip install opencv-python pip install numpy pip install scikit-image

image_path1 = "/path/to/image1.png"
image_path2 = "/path/to/image2.png"
output_matched_lines = "/path/to/output_matched_lines.jpg"

star_matcher = StarMatcher(image_path1, image_path2)
matched_stars = star_matcher.match_stars()
star_matcher.draw_matched_lines(output_matched_lines, matched_stars)

Summarize

in this project we have to code an algorithm to match between stars in two images or more , trying to detect stars in the first image using CHT - use the OpenCv function HoughCircles(), it works in roughly analogous way, that loads an image(reduce noise by blur it)->applies the houghCircleTransform ->display the detecedd circle in the window. Then we impelement the ransac - a statistical approach for curbing outliers , to match between the corresponding stars in the images and to connect between them by drawing lines.
More information about the code: in the code we use an externalLibrary(Numpy), and OpenCv, and OpenSource(scikit-image), they all must be installed before running the code. in the star_coordinates library we can get the star's coordinates and save them, also we draw circles round matched stars , in that we save coordinates of image1 and match stars of image2