The Visual Computer

, Volume 10, Issue 3, pp 160–172 | Cite as

SIMD algorithm for curved object recognition using Grimson and Lozano-Pérez matching

Original Articles
  • 30 Downloads

Abstract

The complexity of many computer-recognition problems is such that speed of processing is an important factor that cannot be overlooked, especially when real-time applications are sought. In order to reduce the processing time of matching, our work is implemented in parallel on SIMD architecture. We describe an extension of the Holder and Buxton algorithm (1989) using the surface normal and axis of rotation of objects as ‘edge’ features to increase the object-recognition scope to objects containing developable surfaces. In addition, we implement an improved data-sorting algorithm that gives impressive speed ups compared with the earlier sorting technique. We show the method to be highly effective in the fast determination of scene interpretations with tests using artificial scenes generated efficiently by a parallel ray caster incorporating constructive solid geometry (CSG). Accuracy and robustness are further tested by application to a real-world scene.

Key words

Computer vision Model matching SIMD parallelism Planar curve Ray tracing 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Appel A (1968) Some techniques for shading machine renderings of solids. AFIPS Spring Joint Computer Conference 32:37–45Google Scholar
  2. Cai LD, Porrill J, Pollard SB, Mayhew JEW, Frisby JP (1990) Segmentation of planar curves using local and global behaviour analysis. Proc BMVC 1:247–251Google Scholar
  3. Faugeras OD, Hébert M (1983) A 3D recognition and positioning algorithm using geometric matching between primitive surfaces. Proc Int Joint Conf AI 8:996–1002Google Scholar
  4. Flynn AM, Harris JG (1985) Recognition algorithms for the connection machine. Proc Int Joint Conf AI 9:57–60Google Scholar
  5. Grimson WEL, Lozano-Pérez T (1984) Model-based recognition and localization from sparse range or tactile data. Int J Robotics Research 3(3):382–414Google Scholar
  6. Holder D, Buxton H (1989) Polyhedral object recognition with sparse data in SIMD processing mode. Image and Vision Comput 7(1):71–78Google Scholar
  7. Kajiya IT (1984) Ray tracing volume densities. Comput Graph 18(3):165–174Google Scholar
  8. Murray DW, Buxton BF (1990) Experiments in the machine interpretation of visual motion. MIT Press, CambridgeGoogle Scholar
  9. Murray DW, Cook DB (1988) Using the orientation of fragmentary 3D edge segments for polyhedral object recognition. Int J Comput Vision 2:147–163Google Scholar
  10. Roth SD (1982) Ray casting for modeling solids. Comput Graph Image Proc 18:109–144Google Scholar
  11. Shankar RV, Ramamoorthy G, Suk M (1990) Three dimensional object recognition on the connection machine. Pattern Recog Lett 11(7):485–492Google Scholar
  12. Slater M (1986) Segments on bit mapped graphic displays. Software-Practice and Experience 16(11):965–980Google Scholar
  13. Usoh M (1992) Parallel process techniques for 3D model-based vision. Thesis, QMW CollegeGoogle Scholar
  14. Usoh M, Buxton H (1993) Using visualisation as a tool for model-based recognition. The Visual Computer (special supercomputing edition)Google Scholar
  15. Whitted T (1980) An improved illumination model for shaded display. Comm ACM 23(6):343–349Google Scholar
  16. Williams NS, Buxton BF, Buxton H (1987) Distributed ray tracing using an SIMD processor array. NATO Advanced Study Institute, Ciocco, ItalyGoogle Scholar

Copyright information

© Springer-Verlag 1993

Authors and Affiliations

  1. 1.Department of Computer ScienceQueen Mary and Westfield CollegeLondonEngland

Personalised recommendations