Viewpoint: Industrial imaging
Martin Kersting, technical director at Stemmer Imaging, sees improvements in software functionality and improved usability of vision software packages going into 2015
M
achine vision soſtware has always played an important role for the efficiency of industrial imaging and automation
systems. With the smart factory developments we are facing, the importance of vision soſtware has increased – smart vision soſtware is a key element to reach higher levels of flexibility in production. Regarding the usability of machine vision
soſtware, we see two development trends: for the increasing number of end users, the soſtware packages need to get easier to handle so that this user group can develop their solutions quickly on a point-and- click basis. On the other hand, integrators and experienced users need the full flexibility of programmable soſtware suites to create specialised solutions. Machine vision soſtware suppliers need to introduce soſtware packages for both of these user groups in order to meet their requirements. Another aspect that supports the success of
machine vision soſtware is that machine vision is already part of the development process for machines that incorporate this technology – in former times, machines have oſten been built
Software is
changing from an inspection tool to a control system
without having the machine vision details and capabilities in mind from the very beginning of the machine design process, and the inspection system was only integrated aſter the basic production system had already been engineered. Nowadays, these machines are engineered with the vision elements included during the first stages of development. Today’s manufacturing machines use machine vision systems at the end of the production line to make a quality decision – and include machine vision at earlier stages of the process in order to prevent bad parts by controlling production parameters. Tis raises the cost effectiveness of manufacturing.
Finally, the functionality of machine vision
soſtware has changed from purely measuring to more of a controlling tool. It can help to recognise process trends at an early stage and, for example, make sure that tool wear is indicated in good time so the user can change tools or parameters of the process before producing rejects or damaging the machine. Soſtware is changing from an inspection tool to a control system. All these aspects will
stimulate the success of machine vision soſtware in the near future. Industrial camera suppliers continue to
develop cameras with new sensors that offer higher resolutions and higher frame rates. Despite interfaces with growing capabilities, the challenge still is how to process the growing amount of image data and how the machine vision soſtware can handle it. Technologies like shiſting the processing to the Graphical Processing Unit (GPU), working with FPGA-based frame grabbers in a user-friendly environment, or using distributed processing all offer interesting solutions to face that challenge. Te suppliers of these technologies need to develop solutions that can be used on the industrial factory floor which is a major technological challenge for the near future. With the Common Vision Blox soſtware development suite and other soſtware systems like Silicon Soſtware’s Visual Applets, a graphical tool that simplifies the programming of image pre-processing on FPGAs, as well as GPU and distributed processing support within Common Vision Blox, Stemmer Imaging can supply powerful solutions answering these challenges.
Peter Bhagat, CEO of Gardasoft Vision, argues that ‘better control and integration of lighting products is needed to improve overall vision system design’
I
t may seem surprising that I’d like to start with the trends in camera technology. Why? Well, lighting has always been a crucial ‘partner’
to cameras within any machine vision system, and as camera developments rapidly continue in areas such as processing speed, networking and standardisation of implementation, it places even greater emphasis on lighting technology. Lighting can even provide the lead function, whereby the processing overhead on the camera can be reduced – for example, high-speed, multi- imaging in the same application scene where different lighting levels are required. Tis view raises a paradox: machine vision
systems rely on highly effective integration between devices, with evolving developments of the various networking standards and enhancements from the leading image processing soſtware providers. Lighting is very obviously absent from this networked environment – yet,
as the industry agrees, lighting is a vital enabler within any machine vision system. I see the positive sign of manufacturers of LED lighting producing increasingly high-quality, increased output lighting devices – and so it’s the integration and control of these lights that I see as a major development requirement in resolving this paradox. Integration of lighting would
enable lighting parameters to be available right through to application level soſtware, bringing advantages in terms of increased lighting control functionality, and also speed of implementation and on-going maintenance and diagnostics. Another key area I recognise is the widening use of expert control functions for lighting, having spent many years addressing machine vision applications that have
14 Imaging and Machine Vision Europe • Yearbook 2014/2015 Techniques have
to be made approachable to a wider audience
been enhanced with the likes of finite pulse control, multi-pulsing, accurate overdriving and stabilised current supplies. However, the techniques are still performed nearly exclusively within an expert sector of the market, which I estimate to be about 10 per cent of all lights used within machine vision. Te techniques have to be made approachable to a wider audience. To achieve this, advanced light control techniques need to be made in much more of a plug-and-play
format. If this is achieved, I estimate that closer to 40 per cent of machine vision lighting will use advanced control functions, bringing extensive benefits to applications.
@imveurope
www.imveurope.com
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