12 December 1997 Complex motion measurement using genetic algorithm
Author Affiliations +
Abstract
Genetic algorithm (GA) is an optimization technique that provides an untraditional approach to deal with many nonlinear, complicated problems. The notion of motion measurement using genetic algorithm arises from the fact that the motion measurement is virtually an optimization process based on some criterions. In the paper, we propose a complex motion measurement method using genetic algorithm based on block-matching criterion. The following three problems are mainly discussed and solved in the paper: (1) apply an adaptive method to modify the control parameters of GA that are critical to itself, and offer an elitism strategy at the same time (2) derive an evaluate function of motion measurement for GA based on block-matching technique (3) employ hill-climbing (HC) method hybridly to assist GA's search for the global optimal solution. Some other related problems are also discussed. At the end of paper, experiments result is listed. We employ six motion parameters for measurement in our experiments. Experiments result shows that the performance of our GA is good. The GA can find the object motion accurately and rapidly.
© (1997) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Jianjun Shen, Jianjun Shen, Dan Tu, Dan Tu, Zhenkang Shen, Zhenkang Shen, "Complex motion measurement using genetic algorithm", Proc. SPIE 3173, Ultrahigh- and High-Speed Photography and Image-based Motion Measurement, (12 December 1997); doi: 10.1117/12.294526; https://doi.org/10.1117/12.294526
PROCEEDINGS
8 PAGES


SHARE
Back to Top