Accessibility Analysis in Camera Placement Network Design for Vision Metrology

SUMMARY Automatic optimum network design for measuring complex industrial objects by vision metrology systems is a real challenge. In the absence of 3D simulated CAD models of the complex objects along with the absence of workspace information, several uncertain parameters are introduced into the camera placement decision-making process. These uncertain factors include the vision constraints such as visibility, accessibility and camera-object distance. The mutual effects of these uncertain factors make the decision making for the camera placement still more complicated. These parameters influence directly not only the mensuration quality but also visibility restrictions. If a priori 3D CAD model of the object and workspace information are available, the aforementioned ambiguities can be tackled. However, the 3D model and the workspace information may not be available in many circumstances. This makes the camera placement problem a nondeterministic process and hence demands a non-rigorous treatment. This paper is concentrated on the solution of the camera accessibility problem. For the complex objects, the accessibility is vastly influenced by the workspace vacancy and tools for putting camera into proper positions. In this paper we propose a novel method for computation of the so called accessibility uncertainty prediction (AUP) for the camera placement using a fuzzy logic approach. The test results indicate the high potential of the proposed method for automatic sensor placement in vision metrology without having access to any 3D CAD model and workspace information.