This is the repo containing my second computer vision project.
In this project, different edge detection approaches are tested. The techniques used in this project are:
1-Sobels
2-Laplacian
3-Gradient Magnitude
4-Canny edge detection
After that, we improved the edge detection results with Gaussian Blurring. The best result achieved was canny edges on the blurred image.
Then different Hough transformations were tested. In the first test, we obtained railway tracks from an image using Hough Lines. After that, we found the circles in a couple of images using Hough Circles. The method did well in most cases, But it couldn't find the tiny circles.
In the last part, we found image contours in the images of the last part.