Photo-mosaicing of images of pipe inner surface
This paper describes the algorithm for the construction of continuous visually consistent images of the inner surface of a pipe from a sequence of images acquired by a wide-angle camera that traveled inside the pipe. The algorithm is designed to be a proof of concept and performs well on simulated data (rendered images) even when camera poses (attitude and location) have errors as much as 5%. Photo-mosaics are suitable for traditional (visual) inspection or automatic processing for the detection of manufacturing faults, corroded areas, and cracks. It is demonstrated that the quality of the resulting mosaic depends how the camera is oriented with respect to the pipe axis and that the traditional orientation with an almost collinear camera optical axis and the pipe axis is not the optimal choice. The proposed system is useful for inspection of pipelines that cannot accommodate traditional devices (e.g., pipeline inspection gauges or crawlers), for example, small-scale boilers and gas systems.
KeywordsMosaicing Stereoscopic processing Pipe inspection Monocular stereoscopy
Unable to display preview. Download preview PDF.
- 3.Jokinen, O.: Reconstruction of quadric surfaces from disparity measurements. In: Tescher, A.G. (ed.) Applications of Digital Image Processing XVII, Proceedings of SPIE, vol. 2298, pp. 593–604. (1994)Google Scholar
- 4.Hartley, R., Zisserman, A.: Multiple view geometry. Second edn. Cambridge University Press, Cambridge. ISBN 0521 54051 8 (2003)Google Scholar
- 6.Harris, C., Stephens, M.: A combined corner and edge detector. In: Proceedings of the 4th Alvey Vision Conference, pp. 147–151. (1988)Google Scholar
- 7.Kanazawa, Y., Uemura, K.: Wide baseline matching using triplet vector descriptor. In: Proceedings of the 17th British Machine Vision Conference, pp. 267–276. (2006)Google Scholar
- 9.Agarwala, A., Dontcheva, M., Agrawala, M., Drucker, S., Colburn, A., Curless, B., Salesin, D., Cohen, M.: Interactive digital photomontage. In: Proceedings of the SIGGRAPH’04 (2004)Google Scholar
- 10.Chaperon, T., Goulette, F.: Extracting cylinders in full 3D data using a random sampling method and the Gaussian image. In: Proceedings of the Vision, Modelling and Visualisation, pp. 35–42 (2001)Google Scholar
- 11.Rabbani, T., van den Heuvel, F.: Efficient Hough transform for automatic detection of cylinders in point clouds. In: ISPRS, Workshop “Laser scanning 2005” (2005)Google Scholar