Model-based automatic calculation and evaluation of camera positions for industrial machine vision

Abstract

One of the key issues for a successful inspection process is the determination of the necessary number of cameras and their respective positions given a specific inspection task and a geometric model of the inspected work-piece and its surroundings. In the last decades, a number of approaches concerning camera positioning strategies have been proposed. Generally, these approaches define an inspection task in terms of good visibility of certain features on the surface of the inspected objects. However, these approaches neither provide general means to include arbitrary inspection requirements, nor do they minimize the number of required cameras. Others use only hard constraints to determine the area of feasibility for certain task requirements. To overcome these shortcomings, we propose a model-based approach to optimize one or more camera positions by optimizing cost-functions derived from the inspection task. The goal is to use a minimum number of cameras / camera positions to fulfill the inspection task. Feature-visibility is represented using a novel concept: the visibility map. It can be calculated quickly by using a projective approach, consumes little storage memory and allows for quick feature-visibility checks. The system is evaluated on several examples using real inspection tasks from current production processes.

Reflectivity function based illumination and sensor planning for industrial inspection

“…In a previous publication, 3 we have shown how to calculate one or more optimal viewpoints given a geometric description of the inspected object, the location of two-dimensional feature areas that are to be inspected and a description of the sensor. However, illumination, its interaction with the inspected object and its influence on the internal sensor settings was not considered.…”

“…The optimum sensor position for each inspection task was calculated using the existing viewpoint planning framework described by Ellenrieder and Komoto. 3 The employed 8-bit camera had a resolution of 1032x776 pixel at a focal length of 25 mm. For all experiments, we used LED illumination with an effective ligh power of 230.43 mW and reflector opening angle of γ = 12 • .…”

“…Some of the key issues for a successful inspection process are suitable positions for the cameras and illumination devices. While several approaches for automatic sensor planning are available, [1][2][3] interaction between illumination, inspected object and sensor settings is considered only by very few publications. Some approaches, e.g.…”

A Versatile Model-Based Visibility Measure for Geometric Primitives

“…to compare all possible assignments is introduced in [4]. Based on this equation, an algorithm to find the optimal sensor-feature association matrix C opt with a minimum number of sensors is also presented.…”

“…Another application using visibility maps is presented in [4]. There, the optimal viewpoints for industrial inspection tasks are calculated from a geometric model of the inspected objects (Fig.…”