Generation of differential topographic images for surface inspection of long products
- 11 Downloads
The current manufacturing industries need efficient quality control systems to ensure their products are free of defects. In most cases, surface inspection is carried out by automatic systems that process 2D images which lack measurable information such as the height or depth of the surface defects. An alternative technology for surface inspection is laser scanning. Using this technique, a 3D representation of a product can be generated and therefore, defects can be easily measured. This paper proposes a real-time algorithm to generate differential topographic images of the surface of a product using laser scanning. The images generated by the proposed method are a flattened representation of the surface of the product which compare it to a perfect-shaped product. In these images, the volumetric defects can be easily segmented and measured using computer vision techniques to fulfill the requirements of the international standards of quality. The proposed algorithm is tested on 500,000 profiles meeting the constraints of real time.
KeywordsReal-time imaging 3D reconstruction Differential image Laser scanning
This work has been partially funded by the project TIN2014-56047-P of the Spanish National Plan for Research, Development and Innovation.
- 2.Deutschl, E., Gasser, C., Niel, A., Werschonig, J.: Defect detection on rail surfaces by a vision based system. In: Intelligent Vehicles Symposium, 2004 IEEE, pp. 507–511. IEEE, (2004)Google Scholar
- 4.Garrett, T., Radkowski, R., Sheaffer, J.: Gpu-accelerated descriptor extraction process for 3d registration in augmented reality. In: 2016 23rd International Conference on Pattern Recognition (ICPR), pp. 3085–3090. (2016)Google Scholar
- 5.Gautier, Q., Shearer, A., Matai, J., Richmond, D., Meng, P., Kastner, R.: Real-time 3d reconstruction for fpgas: a case study for evaluating the performance, area, and programmability trade-offs of the altera opencl sdk. In: 2014 International Conference on Field-Programmable Technology (FPT), pp. 326–329. (Dec 2014)Google Scholar
- 6.Grant, W.S., Voorhies, R.C., Itti, L.: Finding planes in lidar point clouds for real-time registration. In: 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems, pp. 4347–4354. (2013)Google Scholar
- 8.Ikeda, O., Duan, Y.: Color photometric stereo for albedo and shape reconstruction. In: Applications of Computer Vision, 2008. WACV 2008. IEEE Workshop on, pp. 1–6. IEEE, (2008)Google Scholar
- 11.Molleda, J., Usamentiaga, R., Millara, Á. F., García, D.F., Manso, P., Suárez, C.M., García, I.: A profile measurement system for rail manufacturing using multiple laser range finders. In: Industry Applications Society Annual Meeting, 2015 IEEE, pp. 1–8. IEEE, (2015)Google Scholar
- 13.Santur, Y., Karaköse, M., Akin, E.: A new rail inspection method based on deep learning using laser cameras. In: Artificial Intelligence and Data Processing Symposium (IDAP), 2017 International, pp. 1–6. IEEE, (2017)Google Scholar
- 14.Seçil, S., Turgut, K., Parlaktuna, O., Özkan, M.: 3-d visualization system for geometric parts using a laser profile sensor and an industrial robot. In: Robotics and Manufacturing Automation (ROMA), 2014 IEEE International Symposium on, pp. 160–165. IEEE, (2014)Google Scholar
- 16.Skinner, K.A., Johnson-Roberson, M.: Towards real-time underwater 3d reconstruction with plenoptic cameras. In: 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 2014–2021. (2016)Google Scholar