A NOVEL EDGE DETECTION ALGORITHM FOR IMAGE BASED ON NON-PARAMETRIC FISHER INFORMATION MEASURE

In this paper, we proposed a novel edge detection algorithm based on the nonparametric Fisher information (FI) measure. It does not depend on the gradient or Gaussian smoothing. It takes advantage of the local thresholding to find edges. The algorithm firstly created a binary image by choosing a local threshold value using the non-parametric FI measure. Secondly, the usual masks used to detect the edges. The efficiency of the proposed approach is proved by using examples from the real-world. The performance evaluation of the proposed technique based on peak signal to noise ratio (PSNR) is presented. Experimental results show that the effect of the proposed method is comparable to the classic methods, such as Canny, and it is better than Sobel, Prewitt, and Robert methods.