3D measurement and recognition

In recent years, the development of 3D sensors for digitizing single objects or complete scenes has progressed rapidly. Meanwhile, optical systems are capable of acquiring a high density of measuring points in less than a second. Consequently, 3D image-processing systems can now be used in manufacturing applications without any problem. However, in order to be able to implement applications in measuring and testing technology or automation, faster and more intelligent 3D data analysis methods are needed. The department “Machine Vision and Signal Processing” has developed various algorithms for analyzing 3D image data and point clouds for use in measuring technology as well as in 3D object recognition and scene analysis. These include diverse fit algorithms and automated segmentation methods, which are particularly suitable for processing point cloud data. Calibration techniques can also be implemented, which enable the versatile application of new sensor technologies, e.g. time-of-flight or 3D smart cameras

 

Fit method

Fitting curves and surfaces into measuring points is a key aspect of many tasks in industrial image processing and coordinate measuring technology. The fitting of geometric elements is also important in object recognition and scene analysis.

 

Segmentation method

The surface of test objects can be recorded using a high number of measuring points with optical sensors, such as laser line scanners or fringe projection systems. The object’s areas of interest then have to be extracted from the resulting measurement data.

 

3D tolerance fit

In coordinate measuring technology, it is important to fit shaped elements according to their tolerances. With the aid of the 3D tolerance best-fit method, mechanical gauging can be carried out virtually by computer.