Get Smarter

Digitize one pair of eyes in a 4-Eyes-Check, where two people verify the successful outcome of a task independently from each other. Why? It combines the best of both worlds. Machines never get distracted, people do. People excel in recognizing new situations, machines not.
Use image recognition and artificial intelligence to automate one pair of eyes and build a prototype to perform a 4-eyes-check and can be used in many different scenarios.
Next steps
Next steps
Taking steps to bring the technology in house and scale up across Bayer production facilities

Grants4Tech is an open innovation competition by Bayer’s product supply.  One recent challenge was to build a digital4eyes device. Here is the story of the winning team.   

The Jiglabs team had participated in hackathons before, but Grants4Tech felt different. “It had a feeling of realism,” says JigLabs CEO Ryan Pulkrabek. The brief was ambitious: design and build a versatile and mobile digital4eyes prototype for three scenarios with a surprise 4th use challenge to solve during the final round of the competition at Cube Tech Fair in Berlin, in May 2018. 

They spent two days in Leverkusen to collect the data for three challenges: identify the on/off status of a machine, inspect a steel plate after cleaning, and check a mock assembly line after clearance. In Berlin, the fourth challenge was revealed: check if a polished steel bowl is clean. So the team collected thousands of images on the fly and then, back in the hotel, taught the device to distinguish between light reflections and a drop of liquid on the curvy bowl surface. 

Their hand-held device passed all the tests with flying colors, and the team jumped for joy every time the screen correctly said “Dirty”. They received the best results of all six finalist teams and gained additional points in the soft factor categories, such as easy-to-learn and easy-to-handle. The first prize is endowed with 40,000 Euros. 

3 questions to Ryan Pulkrabek 

JigLabs CEO talks about the impact of the competition
Ryan Pulkrabek
JigLabs CEO

1 Ryan, what kind of difference did this competition make to JigLabs?  

If you are a start-up, it is difficult to make potential customers believe you can solve things for them with a new technology. Now we can prove that we did it for Bayer. So winning the challenge has certainly added to our pedigree and continues to open doors for us.   

2 What do you offer your customers? 

Image recognition and artificial intelligence have progressed dramatically over the past few years and can do fantastic things. If a human can make a judgement by looking at something, our algorithm can do it better and faster. One of our customers is a chicken and egg processor who checks 5 million eggs a day. He could use traditional 100,000 € hardware sensors for that. Teaming up with us, he now uses a simple camera and an algorithm, the eggs roll by and our device tells him, if they are cracked or solid 

3 Was it an inspiration for you to meet the other finalists in Berlin?

Oh yes. The names of the others were familiar to us, but it was the first time we could benchmark our technology against some of these well-established participants. And it was a great chance to talk shop with them about industry trends and promising technologies.  Hyperspectral imaging, for example, can provide insights unknown to the human eye. In addition, we saw how the technology can be evolved from hand-held devices to something that you can wear on your head. I am a technical person and I want to create technologies that make a difference.  Being part of Grants4Tech was a great experience on this journey.