SOFTWARE DEVELOPMENT AND SENSOR SELECTION FOR AUTOMATIC INSPECTION AND SORTING BASED ON VISION AND OTHER SENSOR TECHNOLOGY
Our competences in signal processing, algorithm development and machine learning are very valuable for solving difficult inspection and sorting problems. We specialize in inspection tasks that are impossible to automate with so-called "smart" camera's.
We are not only vision experts. We also have a broad array of expertise, tools and techniques at our disposal and can build decision logic that makes its assessments based on the fusion of different types of data: structured data like records in a database, as well as unstructured data like (3D) camera images and process data. We also have advanced process control expertise that enable us, when desired, to automatically adjust the machine settings to remedy the detected quality problems
We develop custom algorithms for complex machine vision integration projects, using for example 2D or 3D vision or thermal camera images for quality control.
Some Reference Projects
- Automatic measurement and inspection of pipelines, making use of 3D camera and regular RGB camera and accurate GPS with customer Voxdale.
- Detection of the orientation of onions on a conveyor using a high resolution camera for D-Grow. Read here more information.
- Detection of 12 different defects in valves used in shock absorbers with customer Tenneco.
- Inspection of the cross section of aluminium profiles by comparing the images with a CAD file of the desired cross section with customer Reynaers.
- Inspection of how brown the cookies leaving a tunnel oven are and automatically adjusting the oven setpoints to bake precisely with customer La Confiance.
- Inspection of the surface roughness of cardboard with customer ESKA.