A New Approach for Object Boundary Extraction Based on Heuristic Search

摘要

This paper presents a boundary extraction algorithm for endoscopic images based on a d-ual step process. Firstly, a highly accurate and effective detection of the lumen region is performed by modified Iris filter which can enhance and detect the vague boundaries greatly. Then a new boundary extraction algorithm based on a heuristic search approach on the neighborhood pixels is employed to obtain a connected single pixel width outer boundary. It uses 2 preferential sequence windows. By using a set of heuristic rules, each window can change the priority of neighborhoods adaptively according to the irregular directions of the outer boundary, and thus to detect the boundary pixel correctly. The results of the experiments are presented to verify the effectiveness of the proposed technique.