5 May 2014 Target tracking using log-polar transform-based shifted phase-encoded joint transform correlation
Author Affiliations +
Abstract
Automatic target detection and tracking requires efficient recognition of the target pattern in variable environmental conditions. Optical joint transform correlation (JTC) method has been proven to be efficient in recognizing a target without requiring complex optical set up. However, the classical JTC suffers from poor correlation performance, which can be improved through the use of different and modified designs. A very successful scheme is developed by employing phase-shifted and phase-encoded fringe-adjusted JTC (SPFJTC), which provides with a high discrimination between a target and non-target objects in a given scene and better utilization of the space-bandwidth resource. Further enhancement of the target detection performance can be achieved by incorporating log-polar transform in the SPFJTC technique. We applied the SPFJTC technique to the log-polar transformation of both the reference image and the input scene that makes the pattern recognition invariant to rotation and scale variations. Peak-to-side lobe ratio is measured and a threshold operation is employed to detect and track a target in an unknown input scene.
© (2014) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Mohammed Nazrul Islam, Worku T. Bitew, "Target tracking using log-polar transform-based shifted phase-encoded joint transform correlation", Proc. SPIE 9094, Optical Pattern Recognition XXV, 90940B (5 May 2014); doi: 10.1117/12.2051575; https://doi.org/10.1117/12.2051575
PROCEEDINGS
8 PAGES


SHARE
Back to Top