MANUFACTURES, RAPID prototyping, PRODUCTION engineering, MANUFACTURING processes, COMPUTER vision, and ELECTRONIC systems
The article presents information on some of the current applications of machine vision systems in manufacturing, their role in improving the manufacturing process, and the reasons they have not yet lived up to their promise. In recent years, increasing global competition has seriously challenged America's industrial and economic leadership. To remain competitive, organizations have found it necessary to explore new ways to improve their overall quality and manufacturing performance. The goal of machine vision is to electronically achieve visual perception. In order to do so, a vision system relies on electronic processors to acquire and analyze images seen through an imaging device such as a solid-state camera. Vision systems vary in complexity, depending on the number of possible digital values assigned to each pixel. Vision-guided robots have been developed in a variety of manufacturing environments. Even though the automotive and electronics industries continue to be the leading users of machine vision systems, applications can also be found in many other industries.
International Journal of Production Research. 9/10/2000, Vol. 38 Issue 13, p2891-2909. 19p. 23 Diagrams.
MANUFACTURING processes, ALGORITHMS, PRODUCTION engineering, INDUSTRIAL engineering, RAPID prototyping, MANUFACTURING execution systems, COMPUTER vision, and CONFIGURATION space
Visibility arising from computer vision, geometrical design and complexity analysis is widely used in manufacturing processes. According to the definition of the visibility cone of a point, two kinds of visibility of a feature, namely a complete visibility cone and a partial visibility cone, are defined, and the relation between visibility map and complete visibility cone is also discussed. To solve a kind of accessibility and setup problem in mold parting, NC-machining and CMMs inspection path planning, a procedure is proposed to perform visibility analysis with respect to the geometry of the part, the shape of the effector, and degrees of freedom between part/effector. A new method for computing visibility cones is formulated by identifying C-obstacles in Configuration Space (C-Space), in which a general and efficient algorithm is presented and implemented using visibility culling. Compared with previous methods, the proposed algorithm is efficient even in very complex scenes. Finally, the contributions and limits of our work are discussed. [ABSTRACT FROM AUTHOR]