SIMULATION
Virtualisation core Te key to virtual image processing lies in the virtualisation core, consisting of two interconnected components: planning and simulation. Simulating what the camera sees can be used to evaluate the design plan of an inspection system. Te core is fed by a CAD model – the geometry – of a product, along with different inspection parameters, for example the types of defects, product material, and inspection speed. Based on these parameters, the core will output a set of possible solutions and parameters, which an engineer can then
Current inspection
systems offer no, or very little, flexibility when it comes to production lines
use to build an inspection system, as well as the expected results, for example sensing viewpoints, light positions, and simulated inspection images. Te framework is currently being researched
and developed on several fronts in parallel: parametric surface estimation; active model- based position planning; camera lens modelling; position-based defect augmentation; and surface light response modelling. Te emphasis is, firstly, on making the position planning accessible to a broader audience, since it is considered to be the backbone of the overall framework. Tis can then be built on and features added. Te planning backbone will solve the
fundamental inspection problem: maximising object coverage regardless of the surface’s
Even slight variations between the angle of the camera and the surface can make a defect completely invisible
geometrical complexity, while producing a quantifiable coverage measurement. Te requirement is a CAD model of the product, also known as a digital twin. Te model is used as an active component, meaning that information about surface complexity is directly drawn from it and used to generate a list of camera viewpoint candidates – i.e. a list of points in space that might be required to cover all interesting parts of the product. Te viewpoint candidate list is then optimised by modelling the complete inspection environment, using physical-based rendering to simulate sensor response and taking inspection parameters,
such as the number of viewpoints or overall inspection time into account. Te final output is a list of viewpoints for both illumination and camera necessary to carry out the inspection. In the current development phase, the
pipeline produces a list of camera viewpoints in order to inspect the entire geometry of the object. Te viewpoints are also used for manipulator trajectory optimisation, with a key point being the fact that the choice of manipulator rests solely on the system designer. As mentioned earlier, current inspection
systems offer no, or very little, flexibility when it comes to production lines. Terefore, the idea of a flexible inspection system, capable of adapting to small-series production lines, is currently just a dream. By developing virtual image processing and implementing it into the inspection system development process, automated inspection will mature. Surface complexity of the product or its actual size will no longer pose a problem when it comes to system design. Te inspection system development phase will
be shortened thanks to environment modelling capabilities, reducing the amount of physical testing, which will also reflect on the overall cost – not only will it be reduced, it will also be possible to give a more accurate prediction. O
Interface of an inspection system developed for adaptive product inspection. Camera and illumination are mounted on a robot arm
62 Imaging and Machine Vision Europe • October/November 2018
Petra Gospodnetic is completing her PhD at Fraunhofer ITWM. She presented her work at the European Machine Vision Association’s business conference in June in Dubrovnik, Croatia
@imveurope
www.imveurope.com
Slight variations in exposure time (from left; 137ms, 700ms, 2,309ms) might reveal defects in some parts of the object, while, at the same time, making other parts unusable because of over or under saturation
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72
