Product recalls are not only expensive; they can cause long-lasting reputational damage. To address this, one start-up has developed an AI-powered, photon-counting X-ray camera for production line inspection machines.
Deep Detection has developed PhotonAi - a photon-counting, multi-energy x-ray camera designed for food inspection. The innovation is designed to detect foreign bodies, defects, and flaws at a higher resolution and with greater energy efficiency than is currently possible.
“We wanted to improve industrial inspection to achieve safer and better-quality products,” explains Deep Detection co-founder and CEO David Ciudad.
“The food sector wants to produce safe food products, but one persistent problem has been the detection of foreign bodies such as plastics, bones, and glass. While manufacturers are willing to invest in solutions that can help ensure the quality reputation of their brand, systems today have been far from adequate.”
Ciudad and his colleagues thought they could help to address this challenge. They designed PhotonAi to deliver high-speed, high-volume throughput, and to deliver a very high Probability of Detection (POD) rate. The technology is also designed to be integrated into existing processing lines at a minimal cost.
“Our clients are manufacturers of X-ray inspection machines, who then sell these machines to food companies,” explains Ciudad. “They want to be able to design innovative solutions that ensure the quality and safety of the products. And in the end, food manufacturers want to eliminate recalls, which cost money and damage brand reputation.”
For this breakthrough, Deep Detection was awarded the Most Innovative Processing Technology or Technology Platform prize at Fi Europe’s Startup Innovation Challenge. "We are very focused on the food sector, and improving the inspection of products,” says Ciudad. “Regarding the award, we are pleased to have received this recognition, as it shows that our vision can add value to the industry.”
Fitting into automated production processes
Deep Detection was launched in July 2020, in the midst of the Covid-19 pandemic. “At that time everything was difficult,” Ciudad recalls. “We were finishing the first R&D prototype, and since we couldn't leave the house, development was stopped for a long time.”
Nonetheless, Ciudad and his colleagues persevered. The past two years have been spent refining the technology to develop quality cameras that fit seamlessly into automated production processes. “One of the things that sets us apart is the development of AI models to improve the probability of detection and identification of materials,” says Ciudad. “We think that we have a unique value proposition for the food sector.”
Adding value to the food sector
The start-up is currently in the pre-commercial phase, and a number of clients are carrying out pilot tests with industrial prototypes.
“The response has been excellent,” says Ciudad. “We have many clients now interested in trialling the technology, to see what it is capable of. Now the ball is in our court, and we are working on delivering more commercial cameras over the coming months. It will be a very entertaining 2024!”
The key aim now is to become established as a manufacturer of X-ray inspection cameras, with a rigorous and quality production process. The business intends to continue working closely with clients, to fully exploit the possibilities of the first camera model, PhontoAi DS. Feedback received will be taken into account in the development of the second model that is currently being developed.
“Our ambition is to lead industrial inspection with photon counting technologies powered by AI,” says Ciudad. “There are actually very few industrial hardware startups, and really very few focused on the food sector. We believe that it is a global sector with many opportunities, and we want to continue working to be able to add value to food companies.”
Main image: © Informa Markets
