Hybrid human-machine binary morphological operator design. An independent constraint approach

A basic paradigm in mathematical morphology is the construction of set oper ators by concatenations of dilations and erosions via the operations of com position, union, intersection and complementation. Since its introduction, in the 1960s by Matheron and Serra, this paradigm has been applied on image analysis for designing set operators, that were called morphological opera tors. Classically, morphological operators are constructed based on the exp erience and intuition of human beings. Recently, an approach for the automa tic design of morphological operators, based on statistical optimization fr om the observation of collections of image pairs, was proposed. The two app roaches have drawbacks: usually, the first approach is slow and depends on an expert in mathematical morphology, while the second requires large amoun ts of observed data. This paper proposes a symbiosis between the human and the statistical design approaches. The idea is that the design procedure be composed of simplified forms of both. Thus, avoiding difficulties that ari se when applying each one independently. The strategy adopted to achieve th is goal is restricting the space of operators considered with a human-desig ned envelope. (C) 2000 Elsevier Science B.V. All rights reserved.