Class-associative pattern recognition using joint transform correlation

Mohammad S. Alam

doi:10.1117/12.456550

Translator Disclaimer

20 November 2002 Class-associative pattern recognition using joint transform correlation

Mohammad S. Alam

Author Affiliations +

Proceedings Volume 4803, Photorefractive Fiber and Crystal Devices: Materials, Optical Properties, and Applications VIII; (2002) https://doi.org/10.1117/12.456550
Event: International Symposium on Optical Science and Technology, 2002, Seattle, WA, United States

Abstract

In this paper, we investigate the latest advancements in real time pattern recognition using the joint transform correlator (JTC) architectures and algorithms. We propose two class associative correlation filters to detect a class of objects consisting of dissimilar patterns. For enhanced performance, both phase and amplitude information is incorporated in the class detection filters. To suppress undesired crosscorrelation between selected objects a new algorithm is introduced. In addition fringe-adjusted joint transform correlation is utilized to enhance the correlation performance, thus ensuring strong and equal correlation peak for each element of the selected class. The feasibility of the proposed technique has been tested by computer simulation.

Citation Download Citation

Mohammad S. Alam "Class-associative pattern recognition using joint transform correlation", Proc. SPIE 4803, Photorefractive Fiber and Crystal Devices: Materials, Optical Properties, and Applications VIII, (20 November 2002); https://doi.org/10.1117/12.456550

ACCESS THE FULL ARTICLE