Congratulations on 20 years of EMVA! Looking back: What was the most significant technological development or initiative of your association?

First of all, on behalf of the EMVA, I would also like to congratulate Computer&Automation on its 25th anniversary and on a quarter of a century of high-quality reporting. As far as the EMVA is concerned, in an industry as dynamic as image processing, it is the companies that drive innovation with technological developments. As an industry association, the EMVA supports these processes through networking events and conferences at which megatrends are discussed. One example was the Embedded Vision Europe Conference 2017 and its repetition two years later. The standardization efforts of the last 15 years or so have certainly been a decisive initiative for the industry. Standards have substantially advanced the vision tech industry, particularly in terms of customer acceptance of the technology. They simplify the introduction and integration of a technology, they reduce the development costs for suppliers and the learning costs for users. EMVA, as a founding association of the so-called G3 initiative, under whose umbrella existing standardization initiatives are promoted and new standardization fields are developed, has certainly played its part in this. The EMVA-hosted standards GenICam and EMVA 1288 are among the most important in the industry.

Is there an absolute highlight in the association's history? Which one and why?

The most important constant in the history of EMVA is the EMVA Business Conference. EMVA was founded at the very first conference, and the EMVA Business Conference has now been held in 19 major European cities, not to mention the two virtual conferences during the coronavirus pandemic. I would therefore not like to highlight a single event, but rather the success of this conference format, which has lasted for over two decades and brings together the top management of our industry once a year. However, one conference year does stand out: an unforgettable event for all participants was certainly the EMVA Business Conference 2010 in Istanbul, when the Eyjafjallajökull volcano in Iceland erupted as participants were arriving. Flights were canceled throughout Europe and no one was able to leave the city in the normal way. This led to adventurous but joint attempts to get home, from chartered Lear jets that were still allowed to fly to an 18-hour bus journey by around two dozen participants from Istanbul via the Balkans to Munich.

It's not all rosy in 20 years - what do you think was the most difficult time for EMVA?

There have certainly been difficult times in the past 20 years of EMVA. The global financial crisis from 2008 to 2009 led to considerable uncertainty in all markets, and the machine vision industry was no exception. However, the strong drivers of the industry and the strategic need for machine vision technologies have subsequently led to a rapid recovery. More recently, the global Covid-19 pandemic and the resulting supply chain restrictions have certainly impacted many companies and EMVA's activities, drastically limiting the ability to organize the in-person networking events that are so important to our members. Overall, EMVA has proven resilient to global events and we look forward to continuing to support our members in the future, even if we cannot predict future disruptive events.

What technological trends do you see that are or will be important for EMVA's work?

On the one hand, increasing digitalization is shaping the development of the vision tech market. The digitalization of industrial processes to create a uniform product life cycle, from design to production to application, offers many potential applications for image processing. These include the definition of inspection and measurement plans at the design stage or the use of image processing technology to analyze manual assembly processes. Furthermore, the increasing use of AI-based image processing systems is creating an entire technology ecosystem with many companies. These are developing low-code/no-code toolchains to facilitate the use of AI models. Linked to this, the use of fully synthetic data to train AI models will become more important, as will tools to quantify and validate the robustness and safety of models. Many companies are also recognizing the value of additional insights that can be gained from automated quality measurements, such as predicting the performance of a part and supporting predictive maintenance.

What do you think will be the biggest challenge in the coming years, business and/or technical?

Certainly, the machine vision industry will continue to face challenges in the future. However, as a sector that serves many different industries, machine vision has a natural resilience, and as a dynamic industry, unforeseen challenges often create new opportunities for innovative companies. Certainly, there are changes in the industry structure. These include increasing consolidation leading to larger machine vision companies, although this is partially offset by the continued entry of companies offering new approaches to solving machine vision applications. Pricing pressure, particularly from companies manufacturing hardware components, is likely to continue, so the industry as a whole will focus on offering a broader product portfolio or higher value solutions. On the technical side, the increasing use of AI in machine vision applications brings additional benefits but also leads to competitive challenges as barriers to entry for companies outside the traditional machine vision sector have decreased and AI performance needs to be made more transparent to end users, for example to meet stringent quality requirements.