Preprocessing and algorithms selection key points in the problem of simultaneous 3d reconstruction of underwater objects and constructing the camera movement

by

B.A. Skorohod, A.V. Statsenko, S.I. Fateev

 Sevastopol State University, RF, Sevastopol, Universitetskaya St., 33

E-mail: boris.skorohod@mail.ru, lex00x1@mail.ru, fateev-si@ya.ru

DOI: 10.33075/2220-5861-2019-2-30-36

UDC 004.9:004.41

Abstract:

     Distortion of underwater images can reduce both the accuracy and robustness of 3D scene reconstruction and visual odometry algorithms, resulting in a reduction in the number of detected conjugate points on pairs of consecutive images.  In this regard, preprocessing of images and procedures for selecting key points on them are important factors in this task. The article investigates the influence of various algorithms of preprocessing and construction of key points on the properties of 3D scene reconstruction and visual odometry algorithms in the conditions of underwater shooting and uncontrolled camera movement.

     The obtained results showed that the number of conjugate key points to be determined on the image sequence, and consequently, the accuracy of the 3D model of the underwater scene and the trajectory of the camera can depend fundamentally on both the preprocessing and the detector used. As recommendations for practical application, the authors propose joint detection of key points by different detectors, correction of the trajectory of the camera 3D model of the scene, based on several algorithms of preprocessing, the use for visualization the algorithm of preprocessing homomorphic filter with a map of object distances  to the camera.

Keywords: underwater video images, 3D reconstruction of the scene, conjugate  and key points.

Full text in PDF(RUS)

LIST OF REFERENCES:

  1. Casalino G., Caccia M., Caiti A., Antonelli G., Indiveri G., Melchiorri C. and MARIS A.: a National Project on Marine Robotics for Interventions. 2014 22nd Mediterranean Conference on Control and Automation (MED) University of Palermo. June 16–19, 2014. Palermo, Italy.
  2. Lan D., Maurelli F., Larkworthy T., Caldwell D., Salv J.i, Fox M., Kyriakopoulosy K. PANDORA: Persistent Autonomy through Learning, Adaptation, Observation and Re-planning. IFAC-PapersOnLine Vol. 48, Issue 2, 2015. P. 238–243.
  3. San S.,, Ridao P., Olive G., Casalino G., Petillot I., Silvestre C., Melchiorri C., Turetta A. TRIDENT An European Project Targeted to Increase the Autonomy Levels for Underwater Intervention Missions. 2013 OCEANS – San Diego, 23-27 Sept. 2013.
  4. Bonin, F., Burguera, A. & Oliver, G., 2011. Imaging Systems for Advanced Underwater Vehicles. Journal od Maritime Research, VIII(1), P. 65-86.
  5. Hartley R, Zisserman A..Multiple View Geometry in Computer Vision. Cambridge University Press, 2003– Computers – 655 p.
  6. Zuiderveld K. Contrast Limited Adaptive Histogram Equalization // Graphic Gems IV, P. 474–485. 1994.
  7. Arroy A., Sanch G. Analyzing pre-processing filters sequences for underwater-image Enhancement // Contemporary Engineering Sciences, October 2017.
  8. He, K., Sun, J. and Tang, X.: ‘Single image haze removal using dark channelprior’, IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33, P. 2341-2353.
  9. Lindbloom B. Chromatic Adaptation. http://www.brucelindbloom. com /index.html.
  10. Harris C., Stephens M. “A Combined Corner and Edge Detector,” Proceedings of the 4th Alvey Vision Conference, August 1988, P. 147-151.
  11. Rosten, E., and Drummond T. “Fusing Points and Lines for High Performance Tracking,” Proceedings of the IEEE International Conference on Computer Vision, Vol. 2 (October 2005): P. 1508–1511.
  12. Leutenegger, S., Chli M. and Siegwart.R. “BRISK: Binary Robust Invariant Scalable Keypoints”, Proceedings of the IEEE International Conference, ICCV, 2011.
  13. Bay, H., Ess T., Tuytelaars T.,, and Gool.I. “SURF:Speeded Up Robust Features.” Computer Vision and Image Understanding (CVIU). Vol. 110, No. 3, P. 346–359, 2008.
  14. Matas J., Chum O., Urba M. and T. Pajdla. “Robust wide baseline stereo from maximally stable extremal regions.” Proc. of British Machine Vision Conference, pages 384-396, 2002.
  15. Alcantarilla, P., Bartoli A. and Davison A.. “KAZE Features.” ECCV 2012, Part VI, LNCS 7577. 2012, p. 214.
  16. Meline,A., Triboulet J., Jouvence Bl. Comparative Study of Two 3D Reconstruction Methods for Underwater Archaeology. IROS: Intelligent Robots and Systems, Oct 2012, Vilamoura, Algarve, Portugal. 2012.

Loading