Is edge detection better in resized image (increased image size)?

I am detecting edges in images for quality estimation and I found that after resizing (double the original image), we get more efficient result than original size. Please help to find proper reason.

26 April 2016 1,805 10 View