Edge detection is one of the most important and commonly used operations performed in image processing. Edges characterize object boundaries and are therefore useful for segmentation, registration, and identification of objects. In the thresholding method of edge detection, the gradient values of pixels exceeding a certain threshold are considered as edge pixels. In this paper different thresholding methods are used for determining a reasonable threshold value for canny edge detection algorithm and then results obtained are compared on the basis of visual inspection and quantitative parameters i. e. Mean Square Error (MSE) and peak Signal-to-Noise Ratio (PSNR). From experimental results it has been concluded that entropy based method achieves better performance than Otsu and Iterative threshold selection methods. So the results obtained by entropy-based thresholding approach are quite promising.

Thresholding MSE PSNR Entropy