On Obtaining Reliable Spatial Information from Gas Structures with a Stereo Camera System

Spotting and localizing unwanted harmful gas releases is of great interest for the maintenance of facilities in the chemical, oil, gas and biogas industry due to their economic impact and negative effect on the environment. In this regard, stereo gas camera systems have been proposed in the literature for obtaining spatio-temporal information from gas releases. These permit gas localization and tracking. Nevertheless, these works rely on a photo-consistency assumption that holds for most textured opaque objects in the visual spectrum but not necessarily for semi-transparent continuous textures such as gases. In this work, first, a detailed measuring model for gaining gas structure’s spatial information with a stereo camera is proposed. A disparity calculation method and a quality measure is then implemented and tested for finding correspondences in stereo images of continuous textures. The proposed measuring model is tested and validated with the implemented disparity method and using synthetic and real stereo images sequences. The results indicate that the computed spatial information from gas structures approximates reliably the path-averaged and concentration-weighted gas position regarding the perspective of the stereo camera system.

[1]  M. Chamberland,et al.  Remote Detection and Flow rates Quantification of Methane Releases Using Infrared Camera Technology and 3D Reconstruction Algorithm , 2016 .

[2]  Gero Bonow,et al.  Gasleckortungsmethoden für autonome mobile Inspektionsroboter mit optischer Gasfernmesstechnik in industrieller Umgebung , 2015 .

[3]  Sayantan Bhattacharya,et al.  Stereo-particle image velocimetry uncertainty quantification , 2016 .

[4]  Tianshu Liu,et al.  Fluid flow and optical flow , 2008, Journal of Fluid Mechanics.

[5]  Daniel Cremers,et al.  Stereo Scene Flow for 3D Motion Analysis , 2011 .

[6]  Heiko Hirschmüller,et al.  Evaluation of Cost Functions for Stereo Matching , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[7]  Joachim Weickert,et al.  Optic Flow Goes Stereo: A Variational Method for Estimating Discontinuity-Preserving Dense Disparity Maps , 2005, DAGM-Symposium.

[8]  ByoungChul Ko,et al.  Fire detection and 3D surface reconstruction based on stereoscopic pictures and probabilistic fuzzy logic , 2014 .

[9]  Gary R. Bradski,et al.  Learning OpenCV 3: Computer Vision in C++ with the OpenCV Library , 2016 .

[10]  Andreas Kroll,et al.  Quantification of methane gas leakages using remote sensing and sensor data fusion , 2017, 2017 IEEE Sensors Applications Symposium (SAS).

[11]  Wolfgang Heidrich,et al.  Stochastic tomography and its applications in 3D imaging of mixing fluids , 2012, ACM Trans. Graph..

[12]  Andreas Kroll,et al.  Characterization and calibration of a stereo gas camera system for obtaining spatial information of gas structures , 2018, 2018 IEEE Sensors Applications Symposium (SAS).

[13]  P. Rusch,et al.  3-D Reconstruction of Gas Clouds by Scanning Imaging IR Spectroscopy and Tomography , 2010, IEEE Sensors Journal.

[14]  Jos Stam,et al.  Stable fluids , 1999, SIGGRAPH.