Advertisement

Feature Extraction and Correlation for Time-to-Impact Segmentation Using Log-Polar Images

  • Fernando Pardo
  • Jose A. Boluda
  • Esther De Ves
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3046)

Abstract

In this article we present a technique that allows high-speed movement analysis using the accurate displacement measurement given by the feature extraction and correlation method. Specially, we demonstrate that it is possible to use the time to impact computation for object segmentation. This segmentation allows the detection of objects at different distances.

There are several methods to measure movement in front of a mobile vehicle (robot) equipped with a camera. Some methods detect movement from the analysis of the optical flow, while other methods detect movement from the displacement of objects or part of the objects (corners, edges, etc). Those methods based on the optical flow are suitable for high speed analysis (say 25 images per second) but they are not very accurate and treat the image as a whole, being it difficult to separate different objects in the scene. Those methods based on image feature extraction are good for object recognition and clustering, that can be more precise than other methods, but they usually require many calculations to yield a result, making it difficult to implement these methods in a navigation system of a robot or mobile vehicle.

Keywords

Feature Extraction Interest Point Black Ring Object Segmentation Mobile Vehicle 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Daniilidis, K.: Optical flow computation in the log-polar plane. In: Hlaváč, V., Šára, R. (eds.) CAIP 1995. LNCS, vol. 970, pp. 65–72. Springer, Heidelberg (1995)Google Scholar
  2. 2.
    Rojer, A., Schwartz, E.: Design considerations for a space variant visual sensor with complex logarithmic geometry. In: Proc. Int. Conf. on Pattern Recognition, Philadelphia, PA (1990)Google Scholar
  3. 3.
    Questa, P., Sandini, G.: Time to contact computation with a space-variant retinalike CMOS sensor. In: IEEE/RSJ Int. Conf. on Intelligent Robots and Systems, IROS 1996, Osaka, Japan (1996)Google Scholar
  4. 4.
    Ancona, N., Poggio, T.: Optical flow from 1d correlation: Aplication to a simple time-to-crash detector. International Journal of Computer Vision 14, 131–146 (1995)CrossRefGoogle Scholar
  5. 5.
    Pardo, F., Dierickx, B., Scheffer, D.: Space-variant non-orthogonal structure CMOS image sensor design. IEEE Journal of Solid State Circuits 33-6, 842–849 (1998)CrossRefGoogle Scholar
  6. 6.
    Tistarelli, M., Sandini, G.: On the advantages of polar and log-polar mapping for direct estimation of time-to-impact from optical flow. IEEE Trans. on PAMI 15, 401–410 (1993)Google Scholar
  7. 7.
    Díaz, M., Domingo, J., Ayala, G.: A grey-level 2d feature detector using circular statistics. Pattern Recognition Letters 18, 1083–1087 (1997)CrossRefGoogle Scholar
  8. 8.
    Pardo, F., Boluda, J.A., Coma, I., Mico, F.: High-speed log-polar time to crash calculation for mobile vehicles. Image Processing & Communications 8-2, 23–32 (2002)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Fernando Pardo
    • 1
  • Jose A. Boluda
    • 1
  • Esther De Ves
    • 1
  1. 1.Dpt. InformáticaUniversidad de ValenciaBurjassotSpain

Personalised recommendations