A novel method for edge detection in vector images is proposed that do
es not require any prior knowledge of the imaged scenes. In the deriva
tion, it is assumed that the observed vector images are realizations o
f spatially quasistationary processes, and that the vector observation
s are generated by parametric probability distribution functions of kn
own form whose parameters are in general unknown. The method detects a
nd estimates the edge locations using a criterion derived by Bayesian
theory. It chooses the number of edges and their locations according t
o the maximum a posteriori probability (MAP) principle. We provide res
ults that demonstrate its performance on synthesized and real images.