Detection of circular shapes from impulse noisy images using median and laplacian filter and Circular Hough Transform

Abstract

Shapes detection from noisy backgrounds is one of the hot topics in the field of image processing and pattern recognition. This paper investigates a new method which uses different filters like median and Laplacian filter for reducing the effect of noise in the image and applies the Canny edge detection algorithm. As the final step, a special form of Hough transform which is called Circular Hough Transform will be applied to detect the circular shapes in the image. To prove the efficiency of proposed method, results of implementation in real world use is presented. Results clearly show the good performance of it. This method is absolutely useful in the fields like: building extraction from aerial satellite images, digital filming, biometric security, astronomy, etc.