Edge-detection algorithm comparison for coherent imaging ladar

There has been much interest in the field of laser radar (ladar) owing to its high-resolution three-dimensional imagery. However, the coherent ladar images are affected badly by speckle, which is a multiplicative noise and has the statistical features of the negative exponential density. In the paper, the morphological filter based on the parametric edge detection is introduced in detail. Then, this detector and the ratio edge detector are compared with the conventional LOG (Laplacian of Gaussian) operator and Canny operator. At last, the edge detection results for coherent laser radar image are obtained. The experimental results show that the morphological filter based on parametric edge detection and the ratio edge detector are better than other algorithms for coherent ladar image, and the morphological filter based on the parametric edge detection is the best algorithm in this paper.