Aydin Ayanzadeh1, Pourghaemi, H1, Seyfari, Y1
1Computer Science Department, University of Tabriz, Tabriz, Iran
In Knowledge-Based Engineering and Innovation (KBEI), 2015 2nd International Conference on (pp. 504-509). IEEE. DOi: 10.1109/KBEI.2015.7436096
inspired from the J-Tian Method.(Tian, J., Yu, W., Xie, S.: An ant colony optimization algorithm for image edge detection. In: Proc. IEEE Congress on Evolutionary Computation, Hongkong, China,pp. 751–756 (June 2008))
A modified ant colony based approach to digital image edge detection
The performance of proposed method is compared with traditional ant colony methods, also we have large number of experiments to find out the suitable threshold for proposed method. The experimental results clearly indicate how the ACO can extracts edges in efficient way, also we speed up the proposed method by modifying the effective parameters in speed of the problem and replacing them by optimized values. The results show that this method is faster and more efficient than other former Ant Colony-based edge detection methods.
Based on dimensions of the input image, there is a significant increase in the elapse time of the algorithm. The dimensional is directly proportional to the computation time. An increase in the computation time was witnessed when the dimensions of the images were 512 × 512 and higher. For images that has high dimensions, the algorithm took much Longer time and some extra load on the processor was noted. The computational time for the image Cameraman of 128 × 128 dimension was 32.5879s with our systems.
If you use the method in your project, please cite the following paper: