Fast restoration of geometric details of automobile castings scanned by RGB-D sensor

The depth data of automobile castings obtained by RGB-D sensor are usually combined with noise, the classical regularization method can eliminate the noise efficiently. Yet the regularization step is too time-consuming to reconstruct the geometric details of automobile castings efficiently. Given this, we present a fast method called fast restoration of automobile castings (FRAC) to restore the geometric details of automobile castings in fast manner. First, the implicit surface data is extracted from globally aligned RGB-D images, the voxel data structure is extended to index and process the implicit surface in real time. Then, an inverse shading formula is constructed to compute TSDF (truncated signed distance field) values of casting surfaces quickly, and an objective function is designed to optimize the geometric details of casting surfaces in real time. Finally, a GPU-based Gauss–Newton solver is used to accelerate restoration of castings further. The defective casting models scanned by RGB-D sensor are quickly refined to a complete model with better accuracy. Experimental results show that with respect to the sampled automobile castings which include 359,470 points in average, the average optimization time reaches 0.66 s per frame, the average restoration time is about 6.48 s. Computing TSDF requires only about 34.8 MB GPU caches in average. FRAC is able to restore the geometric details of automobile castings in real time.

[1]  John Stavridis,et al.  Quality assessment in laser welding: a critical review , 2017, The International Journal of Advanced Manufacturing Technology.

[2]  Matthias Nießner,et al.  Shading-based refinement on volumetric signed distance functions , 2015, ACM Trans. Graph..

[3]  Vo Hoai Viet,et al.  WSDF: Weighting of Signed Distance Function for Camera Motion Estimation in RGB-D Data , 2016 .

[4]  Murilo G. Coutinho,et al.  Guide to Dynamic Simulations of Rigid Bodies and Particle Systems , 2012, Simulation Foundations, Methods and Applications.

[5]  C. C. Chang,et al.  Three-dimensional image reconstructions of complex objects by an abrasive computed tomography apparatus , 2003 .

[6]  José García Rodríguez,et al.  Interactive 3D object recognition pipeline on mobile GPGPU computing platforms using low-cost RGB-D sensors , 2016, Journal of Real-Time Image Processing.

[7]  A. Parvizi,et al.  Optimum curved die profile for tube drawing process with fixed conical plug , 2018 .

[8]  Seiichiro Kagei,et al.  3D reconstruction and multiple point cloud registration using a low precision RGB-D sensor , 2016 .

[9]  Murilo G. Coutinho.  Appendix G: Constructing Signed Distance Fields for 3D Polyhedra , 2013 .

[10]  M. Goesele,et al.  Floating scale surface reconstruction , 2014, ACM Trans. Graph..

[11]  Thomas Sangild Sørensen,et al.  Connected Components Labeling on the GPU with Generalization to Voronoi Diagrams and Signed Distance Fields , 2013, ISVC.

[12]  Ming C. Lin,et al.  Example-guided physically based modal sound synthesis , 2013, ACM Trans. Graph..

[13]  Matthias Nießner,et al.  Real-time 3D reconstruction at scale using voxel hashing , 2013, ACM Trans. Graph..

[14]  Hans-Peter Seidel,et al.  Coherent Spatiotemporal Filtering, Upsampling and Rendering of RGBZ Videos , 2012, Comput. Graph. Forum.

[15]  Xiaoyang Liu,et al.  Real-Time Geometry, Albedo, and Motion Reconstruction Using a Single RGB-D Camera , 2017, ACM Trans. Graph..

[16]  Hujun Bao,et al.  Robust 3D Reconstruction With an RGB-D Camera , 2014, IEEE Transactions on Image Processing.

[17]  Sehoon Ha,et al.  Iterative Training of Dynamic Skills Inspired by Human Coaching Techniques , 2014, ACM Trans. Graph..

[18]  Yu Yao,et al.  Detection of a casting defect tracked by deep convolution neural network , 2018, The International Journal of Advanced Manufacturing Technology.

[19]  Radovan Kovacevic,et al.  Detection of defects in laser welding of AZ31B magnesium alloy in zero-gap lap joint configuration by a real-time spectroscopic analysis , 2014 .

[20]  Weiwei Zhang,et al.  Real-time vehicle type classification with deep convolutional neural networks , 2017, Journal of Real-Time Image Processing.

[21]  Xiaotong Jiang,et al.  Interior structural optimization based on the density-variable shape modeling of 3D printed objects , 2016 .

[22]  Luigi di Stefano,et al.  On-Line Large Scale Semantic Fusion , 2016, ECCV Workshops.

[23]  Ayoub Al-Hamadi,et al.  Truncated Signed Distance Function: Experiments on Voxel Size , 2014, ICIAR.

[24]  Ankit Chaudhary,et al.  Both Hands' Fingers' Angle Calculation from Live Video , 2012, Int. J. Comput. Vis. Image Process..

[25]  Nassir Navab,et al.  SDF-2-SDF Registration for Real-Time 3D Reconstruction from RGB-D Data , 2017, International Journal of Computer Vision.

[26]  Vimal Dhokia,et al.  Additive manufacturing simulation using signed distance fields , 2016 .

[27]  John J. Leonard,et al.  Robust real-time visual odometry for dense RGB-D mapping , 2013, 2013 IEEE International Conference on Robotics and Automation.

[28]  Bernd Hamann,et al.  On-line CAD Reconstruction with Accumulated Means of Local Geometric Properties , 2014, Curves and Surfaces.

[29]  Brian Wyvill,et al.  Robust iso-surface tracking for interactive character skinning , 2014, ACM Trans. Graph..

[30]  Lize Gu,et al.  Real-time image recognition using weighted spatial pyramid networks , 2017, Journal of Real-Time Image Processing.

[31]  Dario J. Toncich,et al.  Non-contact inspection for the detection of internal surface defects in hollow cylindrical work-pieces , 1996 .

[32]  Y. Qin,et al.  Micro-manufacturing: research, technology outcomes and development issues , 2010 .

[33]  TheobaltChristian,et al.  Coherent Spatiotemporal Filtering, Upsampling and Rendering of RGBZ Videos , 2012 .