The quality of a product or service is important to companies. Using Computer Vision you can carry out automated visual assessments of the quality. This contributes to a smart industry: a higher quality, and cheaper, more flexible and/or labour-friendly (production) processes.
Computer Vision - the automation of visual inspections - involves making interpretations with the help of computer camera images, after which this information is used to control other processes. Take, for example, quality control, automatic position and orientation determination, disease detection, defect measurements and sorting products.
The professorship of applied sciences works closely with the professorship of applied sciences Data Science in a single research laboratory. Together they form the Centre of Expertise of Computer Vision & Data Science within higher vocational education in the Netherlands. The strength of professorship is that it has both the know-how and equipment for the entire chain of lighting, cameras, optics, configuration, vision algorithms, deep learning algorithms and the embedding of computer vision solutions in existing systems.
Download Computer Vision & Data Science Horizon magazine eng (pdf)
A recent development within Computer Vision & Data Science is the use of Unmanned Aerial Vehicles (UAVs) or 'drones' for carrying out the visual inspections. For example when inspecting wind turbines or surveying ditches.
The professorship carries out research into the automated (real-time) processing of the recorded images into information that supports processes and operations. Research is also carried out into realising the fully automated flying of UAVs. NHL Stenden Hogeschool, Computer Vision & Data Science is a certified aviation company and has its own UAVs.
Smart Vision for UAVs Smart Industries
Identifying moulds in potato fields, detecting the source of a fire or inspecting the vanes of a windmill. With an Unmanned Aerial Vehicle (UAV) - or drone - this is all possible these days. The Smart Vision project for UAVs of the Computer Vision professorship provides innovative solutions.
The use of unmanned aerial vehicles has experienced unprecedented growth over the last few years. Drones are being used for practical purposes ever more frequently, whereby high-tech cameras act as the eyes of the client on the ground. At NHL Stenden, the Smart Vision for UAVs project has been focusing its full attention over the last few years on the development of small civil drones, that can fly automatedly as well process the recorded images into concrete information that can support the processes or operations of the customer. And with success, because the project has already delivered innovative solutions for complex issues from the field.
Intensive contact with the field
"We work together with a large consortium of businesses that wish to deploy UAVs for practical purposes," says researcher Klaas Dijkstra. "There is Wetterskip Fryslân for surveying ditches, research and consultancy agency HLB for whom we carry out research into diseases in arable farming, Firefighting training centre Noord which we support by identifying the sources of fires and the Leeuwarden-based company Epomat which is specialized in the inspection and maintenance of turbine blades. These are the main customers of innovative solutions with drones, although there are many more technological suppliers in our consortium.”
LapVas Smart Industries
'Suture leakage' after an operation is cause for concern. The Computer Vision professorship searched in the LapVas project for a measurement instrument that could identify the quality of the tissue of the intestine wall.
With the measurement instrument, it is possible to make the micro (blood) circulation of the tissue visible. This helps the surgeon operating to make better decisions. The aim is to reduce the number of intensive care admissions, to have patients recover more quickly and to reduce the number of necessary reversal operations. The measurement instrument is used to carry out various tests.
In addition to the MCL in Leeuwarden, the Biomedical Photonic Imaging department of Twente University is also actively involved in the LapVas research. The project is part of the broader research project OR of the future of innovation agency LIMIS, which carries out research into healthcare innovations.
Circular Plastics Recycling plastic is a well-known social challenge. The Computer Vision & Data Science professorships are working on this issue in collaboration with the Circular Plastics research group. If plastics can be sorted with sufficient accuracy, the recycling chain will work more effectively and it will no longer be necessary to burn plastic. The central question is: How can this sorting be improved with the help of techniques from the Computer Vision and Data Science?
This project uses a hyperspectral camera that is sensitive to Short Wave Infrared (SWIR). In this part of the electromagnetic spectrum, it is possible to distinguish different polymers based on chain length (and therefore type, PE, PP, PVC, etc.).
By using the latest techniques from artificial intelligence (including deep learning), recycling systems can be trained data-driven. This approach not only takes into account the spectral properties of polymers in the sorting, but also the morphological properties. In the first phase of the research, promising results were achieved that demonstrate the added value of using deep learning in the context of polymer recycling.
With whom do we work together?
The Computer Vision professorship is the connecting link between higher vocational education and the professional field. We help businesses to innovate by making academic knowledge applicable. We share this knowledge with students, so that in the future they, as innovative professionals, will be able to strengthen businesses.
In order to realise this, the professorship is active as a researcher and educator as well as an entrepreneur. The Computer Vision professorship has for years now been working intensively with other parties. As a result, we have built up an extensive network of businesses, suppliers, universities and knowledge institutes with whom we share knowledge and experiences.
The Computer Vision Northern Netherlands Cluster is a platform set up by the professorship and consists of more than thirty businesses in the North of the Netherlands who are actively involved with computer vision (www.ccvnn.nl). The Image Processing Platform has also been set up for higher vocational education, in which NHL Stenden plays a leading role.
The Computer Vision professorship falls under the Research Unit of the Technology & Innovation Academy, together with the Data Science, Sustainable Plastics, Circular Plastics and Water Technology professorships of applied sciences. The professorship is linked to the Smart Sustainable Industries spearhead of NHL Stenden University of Applied Sciences.
The Computer Vision & Data Science professorships offer every semester the opportunity to work within the research laboratory on a research assignment in the field of Computer Vision & Data Science in the form of a minor or (graduation) internship. The semester starts with a two-week Kick-Off course in which the basic principles of Computer Vision & Data Science are taught. Then the students are linked to a project and intensively supervised by the team. More information and the registration option can be requested from our secretariat.
Within the Computer Vision professorship, various disciplines work together on the development of technological innovations. Professor of applied sciences Jaap van Loosdrecht plays a pivotal role. Along with the researcher's students also have an active role in the team.
Over the years the Computer Vision professorship has built a large international network. In addition to students from NHL Stenden's own programmes such as Information Technology, Mechanical Engineering, Chemical Technology, Technical Information Science, Applied Mathematics and Electrical Engineering, there are also contacts with universities of applied sciences abroad. Students from higher vocational educational institutes from Spain, China and Mexico regularly come to Leeuwarden in order to take part here in research carried out at Computer Vision.
NHL Stenden University of Applied Sciences
Academy Technology & Innovation Professorship Computer Vision