Edge Detection and Segmentation of Multiple Contours from CT Scan Images

Abstract

In conceptual design, behavior information often plays a pivotal role in mapping from function to structure. Behavioral process is a detailed description of a number of linked sub-level behaviors, which characterize the process of an overall behavior. To understand this process, as well as to provide assistance to designers for design synthesis, it is necessary to construct an appropriate behavioral process model, for which two aspects of design information need to be identified first. One is the characteristics of behavior itself, and the other is the relations between individual behaviors. In this paper, the characteristics of behavior are clarified and the relations of behaviors such as loop relation, temporal relation and state relation together with causal relation are identified. Based on the existing Causal Behavioral Process (CBP) model, an extended CBP (E-CBP) model is proposed by adding these new behavioral characteristics and relations. An Extended Backus-Naur Form (EBNF) representation method is also proposed to represent the E-CBP model. An electric nailing device design case and a lever-clamp assembly system design cases are illustrated to demonstrate the usefulness and applicability of the E-CBP model as well as its representation method.

Diagnostic accuracy of ovarian cyst segmentation in B-mode ultrasound images

“…The algorithm restarts each time a new black pixel is found and the scanning process stops when the starting pixel is touched. The black pixels touched in this process represent the contour of the object [4], [5].…”