How did you come to be part of the imaging/machine vision industry?
During my electronics studies at Heilbronn University, I first came into contact with image processing. The lecture called ‘process data processing’ imparted the basics of image processing and it was applied practically in the laboratory too. The subject of image processing was so inspiring to me that I wrote my diploma thesis on ‘pattern classification with neural networks’ and afterwards set up the Steinbeis Transfer Centre for Image Processing, together with Professor Dr Uwe Jäger, where we developed image processing applications for customers. As a user of image processing components, I was very dissatisfied with the programmability of the frame grabbers at that time. Register specifications were the only thing you had, and the initialisation extended to several code pages. Therefore, I was glad to join Matrix Vision. We constructed frame grabbers in a way I had missed as a user. As a German manufacturer, we quickly gained a foothold in the German market. Besides good products, the customers wanted efficient and direct support and a partner for image processing questions. We know what pressure customers face and understand that we are not only a supplier of off-the-shelf products, but also a partner who cares about customers. Just like then, this is still our recipe for success.
How do you convince customers that they need machine vision?
As a manufacturer of image processing components it is not our task to suggest image processing to customers, but rather to sell competitive products to system integrators who have practised in this market for years. Ten to 15 years ago, it was common to hear of antipathy against image processing from users. Since they had been deterred by negative experiences, a huge amount of energy was needed to satisfy them with solutions using image processing. Fortunately, this is not necessary any more. There are so many examples, like toll control, reverse vending machines, goods identification via barcode and mobile phones that use image processing every day. Other aspects, like 100 per cent checks as required in ISO standards, awaken interest in using industrial image processing solutions.
What role does Europe have in the development of machine vision?
Europe is a very innovative market. Especially in Germany, there are many system integrators with middle company sizes (20 to 50 employees). These companies can serve production companies very well and they can integrate image processing systems for production control. Naturally, a high demand stimulates development. This is the reason why many technical innovations come from European countries.
What do you see as the major growth sectors?
Besides traditional industrial image processing, other market areas like traffic engineering, surveillance and agricultural and medical engineering are developing well. More and more applications can be solved and realised with image processing. Many techniques used to solve an image processing task can be used comprehensively in other areas. For example, in surveillance there is an increased use of digital cameras resulting from smaller and cheaper cameras. In this area too, image interpretation has to be automated, because the number of cameras is growing and monitoring many cameras cannot be handled manually any more.
What do you see as the most important technological challenges facing the industry?
I see the change from analogue to digital camera technology as the greatest technological change in the last few years. I am thinking especially about the use of standard interfaces like USB and GigE. In comparison to the general information technology industry, image processing is very small and has to benefit from the technical developments of other areas. In addition, I believe that the attempt to define standards is a positive development, because it simplifies the ease of operation for customers and helps to expand the market.
