Hough Transform for detecting Circles

The various results obtained by me as results of applying the circle detection routine is shown below.
The detected circles are marked in red.
 
 

First, I demonstrate the circle detecttion capabilities of the Hough transform
 

The Original Image Accumulator Maxima 114	The detected Circles Accumulator Threshold 100

 

Next, I demonstrate the Hough transform's capbility of detecting circles amongst other shapes.
 

The Original Image Accumulator Maxima 112	The detected Circles Accumulator Threshold 95

 
 

Next, I demonstrate the Hough transform's capability of detecting circles in noisy images.
 

The Original Image Accumulator Maxima 90	The detected Circles Accumulator Threshold 50

 
 

Now, I  demonstrate the Hough Transform's cpability on a real life image. The image is of the set of coins on a denim texture background.
The Hough transform is used to detect a the larger coins.
 
 

	The Set of Coins. Accumulator Maxima 42 (after gradient)
	The Gradient Image Gradient Threshold  0.8
	The detected coins marked in red

Other Experiments possible :

1.    The accumulator resolution and quantization can both be varied and the results studied.
2.    Each edge point can be made to vote for a neighbourhood of cells in the accumulator insted of just on cell and the effect studied.
 
 

Salient Features of this Implementation :

1. Highly efficient due to use of Mid - point circle drawing Algorithm.

2. Very Interactive - allows the user to vary a lot of parameters at runtime such as -
    a) The choice of calculating the Gradient Image
    b) The Gradient Threshold
    c) The Accumulator Threshold (The routine reports the accumulator Maxima for easy selection of this threshold).

Page last updated on 28 January, 2004.	AT cse.iitd.ac.in	© Parag Chaudhuri , 2009