Title :
Decomposition of morphological structuring elements with integer linear programming
Author :
Yang, H.-T. ; Lee, S.-J.
Author_Institution :
Dept. of Electr. Eng., Nat. Sun Yat-Sen Univ., Kaohsiung, Taiwan
fDate :
4/8/2005 12:00:00 AM
Abstract :
A method is proposed for decomposing morphological structuring elements based on integer linear programming. A decomposition problem is formulated into a set of linear constraints, and an optimal decomposition is a solution to the constraints, obtained by the cutting-plane simplex algorithm. The method has several advantages. It provides a systematic way of decomposing arbitrarily shaped structuring elements. For convex images, factors can be of any size, not restricted to 3 × 3; the candidate set can be freely assigned by the user; and the optimality criteria can be flexible.
Keywords :
image processing; integer programming; linear programming; mathematical morphology; convex images; cutting-plane simplex algorithm; integer linear programming; linear constraints; morphological structuring elements; optimal decomposition;
Journal_Title :
Vision, Image and Signal Processing, IEE Proceedings -
DOI :
10.1049/ip-vis:20041072