One of the challenging topics in image processing is extracting the shapes from noisy backgrounds. There are some methods for doing it from different kinds of noisy backgrounds. In this paper, we are going to introduce another method by using 4 steps to extract circular shapes from impulse noisy backgrounds. First step is applying median filter to disappear “salt and pepper” noise. This step causes edge smoothing. So, as the second step, a laplacian sharpening spatial filter should be applied. It highlights fine details and enhances the blurred edges. Using these two steps sequentially causes noise reduction in an impressive way. Third step is using Canny edge detection for segmenting the image. Its algorithm is talked during the paper. Finally, forth step is applying Circular Hough Transform (CHT) for detecting the circles in image. At the end of paper different use cases of this method is investigated.
One of the challenging topics in image processing is extracting the shapes from noisy backgrounds. There are some methods for doing it from different kinds of noisy backgrounds. In this paper, we are going to introduce another method by using 4 steps to extract circular shapes from impulse noisy backgrounds. First step is applying median filter to disappear "salt and pepper" noise. This step causes edge smoothing. So, as the second step, a laplacian sharpening spatial filter should be applied. It highlights fine details and enhances the blurred edges. Using these two steps sequentially causes noise reduction in an impressive way. Third step is using Canny edge detection for segmenting the image. Its algorithm is talked during the paper. Finally, forth step is applying Circular Hough Transform (CHT) for detecting the circles in image. At the end of paper different use cases of this method is investigated.