Industrial Vision Systems Research Articles

Efficient 3-D Object Representations for Industrial Vision Systems

The representation of 3-D objects is an important step in solving many problems in scene analysis. One of the most successful techniques is that based on the surfaces of objects. We describe several methods for obtaining such surface representations from various types of intrinsic images. In particular, previous work is reviewed and an algorithm based on region growing is investigated in terms of its efficiency in segmenting a set of points in 3-D space into planar faces. Information on the neighborhood structure of the points in the form of a spatial proximity graph is used to direct the segmentation. Applications to industrial objects are demonstrated.

Computer vision for industrial applications

In recent years, great interest has developed in the possibilities for using computer vision systems in a variety of industrial applications. These range from complex inspection or monitoring tasks to supplying precise position and orientation information for part feeders or robotic manipulators. The paper describes the work being done at the GEC Hirst Research Centre to develop a modular vision system with its own structured interpretive language. Such a system will provide an excellent tool for building and programming industrial vision systems. The vision system has been incorporated into a robot workcell, and an experiment is described which demonstrates the use of several co-operating processors to perform a simple assembly task

Critical points on a perfectly 8- or 6-connected thin binary line

A set of “logical pixel logic” based procedures is described and illustrated for detecting the near-minimal number of pixels on a perfectly 8- or 6-connected binary thin line which have to be retained to adequately approximate the original pixel string with a connected sequence of straight-line segments. The found critical points or pixels on the line are for practical purposes the same as those chosen by a person, were he asked to perform the same task. The method is fast and in practically all cases the approximating connected sequence of straight-line segments does not deviate by more than one pixel distance from the original pixel string. Experiments on Chinese characters have given very good results. The method could also be used as a preprocessing step to conventional iterative methods based on minimizing a distance criterion between the approximating straight-line segments and the pixels on the original binary line. It will be very useful in any field in which computational speed is one of the major concerns, such as industrial vision systems.

Computer vision for industrial applications

In recent years, great interest has developed in the possibilities for using computer vision systems in a variety of industrial applications. These range from complex inspection or monitoring tasks to supplying precise position and orientation information for part feeders or robotic manipulators. The paper describes the work being done at the GEC Hirst Research Centre to develop a modular vision system with its own structured interpretive language. Such a system will provide an excellent tool for building and programming industrial vision systems. The vision system has been incorporated into a robot workcell, and an experiment is described which demonstrates the use of several co-operating processors to perform a simple assembly task

A new general triangulation method for planar contours

The problem of approximating the surface spanning a given set of 3D points as a polyhedron of triangular faces (“triangulation”) is a significant one, and has many applications in the fields of computer graphics and computer vision. In this paper, several solutions to this problem are reviewed. These solutions can be grouped into two classes, and particular emphasis is given to the class of surfaces spanned by parallel planar contours. For a contour pair P 0 ,P 1 ,...P m−1 and Q 0 ,Q 1 ,...Q n−1 , a graph theoretic approach can be used to arrive at a class of solutions, each requiring exactly m+n steps to triangulate the pair. Existing methods (both rigorous and heuristic) for extracting a particular solution from this group are reviewed, and a new heuristic based on inter-contour coherence is proposed. This heuristic is being used in the field of Ultrasonic Non-destructive Evaluation to produce images of flaws in pressure vessels, and its performance is shown to compare favorably with methods of greater computational complexity. It is believed that this heuristic can also be used with success in industrial vision systems where similar contours are obtained using a laser range finder.

Neural Network Based Object Recognition System ForManufacturing Applications

The manufacturing environment these days is characterised by rapid change, posing new challenges and problems to the production and operations manager in the industry. Under this circumstances, process flexibility is fast becoming a major priority for many organisations as they attempt to deal with these changes. In response to the need for process flexibility, increasing attention is being given to integration of computing technologies with the manufacturing systems leading to the development of flexible manufacturing systems aided with high performance vision capabilities. The flexible manufacturing between various batches. Ideally, FMS should be able to accommodate the batch size of one component. This is possible if the programs for machining of different batches can be loaded instantly without operator intervention. An automatic component (object) recognition system, capable of identifying the component loaded on any given machine is required for meeting this objective. Existence of such a system provides an additional latitude in the scheduling of FMS to reduce the down time and improve the utilisation. This requires development of high performance industrial vision systems, which is quite a challenging task, as vision is a complex and computational intensive task, and the typical manufacturing environment is characterised by several different parts with simple or complex geometric descriptions, partially occluded parts, or parts with surface anomalies, blemishes and varying shapes. In this paper the recognition performance of the neural network based classifiers is reported for an industrial object recognition system being developed at the author’s laboratory. The intelligent object recognition system being developed has a modular structure with each module dedicated to different levels of vision in the vision hierarchy. A comparison of two different cases where the proposed neural network based classifiers perform much better than the general statistical classifiers is reported. The first case considered is that of the one degree of freedom tools where the backpropagation neural net has the better recognition performance and the second case considered is that of partially occluded objects where the Hopfield neural net yields better performance.