Direkt zu




Baden-Württemberg Stiftung





twip optical solutions GmbH



Institut f. Technische Optik
Universität Stuttgart
Pfaffenwaldring 9
70569 Stuttgart

T: +49 (0)711/685-66074
F: +49 (0)711/685-66586



The processing of images is of fundamental importance for a lot of optical measurement systems. Therefore, image processing algorithms are constantly developed at ITO  for very different applications. Of course, a focus of our work is the design of the hardware of typical 2D/3D measurement systems and we are trying to implement the available expert knowledge concerning the design of such systems for technical surface inspection into assistance systems.

In addition, we work towards automatic optimization of image processing algorithms for the evaluation of surface inspection results.

The input to the optimization system is a library of real measurement data together with marked defects in each image. Using an optimization algorithm based on genetic programming, a chain of basic image processing steps together with the appropriate set of parameters is selected, such that the marked regions are automatically determined out of the training data set.



  1. Burla, A.; Haist, T.; Lyda, W.; Aissa, M. H.; Osten, W. „Assistant systems for efficient multiscale measurement and inspection“, Proc. SPIE 8082 (2011).
  2. Lyda, W.; Burla, A.; Haist, T.; Gronle, M.; Osten, W. „Implementation and Analysis of an Automated Multiscale Measurement Strategy for Wafer Scale Inspection of Micro Electromechanical Systems”, IJPEM 13 (2012).
  3. Gronle, M.; Lyda, W.; Burla, A.; Osten, W. “Extrinsic calibration of a fringe projection sensor based on a zoom stereo microscope in an automatic multiscale measurement system”, Proc. SPIE 8430 (2012).
  4. Burla, A.; Haist, T.; Lyda, W.; Osten, W. „Software Assistant System for Automatic Algorithm Design and Configuration in Multiscale Inspection based on Genetic Programming“, Opt. Eng. 51(6) (2012).
  5. Gronle, M.; Lyda, W.; Osten, W. “Concepts for an active inspection of three-dimensional objects using a multi-sensor measurement system”, Proc. Optics and Measurement, Liberec, CZ (2012).