With the growing complexity and optimization of industrial production, quality control is increasingly important. The inability to detect important product defects can cause serious problems or even dangers in their use, resulting in irreparable damage to the manufacturer’s reputation. The key challenge in this domain is to detect defects as early as possible, and make adjustments to the production process to improve production yields.
The Computational Imaging group develops advanced image processing and image reconstruction methods that are capable of detecting defects from a limited number of sensor measurements, and are fast enough to be used in an online factory environment. Moreover, these imaging techniques also yield valuable quantitative information about each individual product. In low-cost, high-volume sectors such as the food industry, our techniques can be used to effectively detect product defects, while in high-cost, low-volume sectors our methods will lead to adaptive industry processes, allowing the repair of defects as early as possible in the production process.
This research provides insights into the use of computational imaging tools and algorithms for optimizing the industrial quality-inspection process, as well as their implementation on high-performance computing hardware.
Contact person: Joost Batenburg
Research group: Computational Imaging (CI)
Research partner: FEI Company