Edge linking by sequential search

摘要

Edge detection is mainly a two-stage process: edge enhancement followed by edge linking. In this paper we develop a linking algorithm for the combination of edge elements enhanced by an optimal filter. The linking algorithm (LINK) is based on sequential search. From a starting node, transitions are made to the goal nodes based on a maximum likelihood metric. Results of our search algorithm are compared to the sequential edge linking algorithm (SEL) as well as to two common nonsequential algorithms based on tracing the zero-crossings loci in the convolution output of the ▿2G operators, and based on nonmaximal suppression and thresholding of the convolution output of the ▿G operator. It is shown that the LINK algorithm provides comparable results, require only local calculations, and can accommodate any a priori information in the path metric used to guide the search.