Three-dimensional pattern recognition using joint transform correlation

Mohammad S. Alam; Sheue Feng Goh

doi:10.1117/12.558827

Translator Disclaimer

22 October 2004 Three-dimensional pattern recognition using joint transform correlation

Mohammad S. Alam; Sheue Feng Goh

Author Affiliations +

Proceedings Volume 5557, Optical Information Systems II; (2004) https://doi.org/10.1117/12.558827
Event: Optical Science and Technology, the SPIE 49th Annual Meeting, 2004, Denver, Colorado, United States

Abstract

A three-dimensional (3D) pattern recognition technique using classical joint transform correlation was proposed recently. However, it exhibits the same drawbacks of classical two-dimensional classical joint transform correlation by producing wide and broad correlation peak in the output plane. Thus, we propose a technique based on fringe-adjusted joint transform correlation to be used in 3D pattern recognition. The proposed technique yields better correlation discrimination ability compared to alternate 3D classical joint transform correlation by producing sharper and stronger correlation peak intensity. Simulation results verify the performance of the proposed technique.

Citation Download Citation

Mohammad S. Alam and Sheue Feng Goh "Three-dimensional pattern recognition using joint transform correlation", Proc. SPIE 5557, Optical Information Systems II, (22 October 2004); https://doi.org/10.1117/12.558827

ACCESS THE FULL ARTICLE