Improvements for volume self-calibration

Abstract

For multi-camera volumetric flow measurement techniques like tomographic PIV or shake-the-box, volume self-calibration (VSC) has become the standard procedure to correct remaining calibration errors using actual recorded images with particles. For high seeding densities and large allowed triangulation errors necessary to detect large camera shifts, the number of triangulated ghost particles can exceed the number of true particles by orders of magnitude, which makes the detection of the true disparity peaks more difficult. VSC has been improved here using the distinction between true particles with disparities always inside the true disparity peak in the disparity map for all cameras, while ghost particles are distributed over random position. This VSC with ghost particle suppression (VSC-GPS) makes VSC significantly more robust with many orders of magnitude fewer ghost particles and enables detection of disparities even larger than 10 pixels.In addition, an alternative volume self-calibration method is presented based on standard image correlation (VSC-IC) between dewarped images of two cameras similar to Stereo-PIV self-calibration without the need of particle detection and triangulation. For each camera combination, a correlation streak becomes visible in the correlation map, where the position of the streak is used for correcting the camera mapping functions. VSC-IC can easily detect very large camera shifts >10–30 pixels. For inline camera configurations, VSC-IC needs to be modified by some slower triple-image correlation technique. As a useful side-benefit, the intensity along the correlation streak provides directly the intensity profile across the measurement volume depth.Both methods, VSC-GPS and VSC-IC, have been tested on many experiments, and in particular for the most difficult case of vibrations where each camera image needs a modified calibration function. Guidelines are given to detect and correct vibrations.