Title :
A new algorithm of infrared gait detection based on immune ant colony
Author_Institution :
Fac. of Autom., Guangdong Univ. of Technol., Guangzhou, China
Abstract :
The accurate infrared gait detection is a research difficulty because of the unique thermal imaging characteristics. Therefore, the coarse result of infrared human detection is obtained by using the background subtraction method based on the single Gaussian background model, and is fine segmented by applying the improved pulse coupled neural network (PCNN). At the same time, the effect of the adaptive segmentation is reached by introducing the immune ant colony algorithm to fast determine the optimal segmentation parameters of PCNN. The simulation results show that this algorithm can achieve the ideal detection effect.
Keywords :
Gaussian processes; evolutionary computation; gait analysis; image segmentation; infrared detectors; infrared imaging; neural nets; optimisation; PCNN; background subtraction method; immune ant colony; infrared gait detection; optimal segmentation parameters; pulse coupled neural network; single Gaussian background model; thermal imaging characteristics; Adaptation models; Approximation algorithms; Biological neural networks; Image segmentation; Immune system; Mathematical model; Neurons; Gait detection; Immune ant colony; Infrared; Pulse coupled neural network (PCNN); single gauss model;
Conference_Titel :
Electronic and Mechanical Engineering and Information Technology (EMEIT), 2011 International Conference on
Conference_Location :
Harbin, Heilongjiang
Print_ISBN :
978-1-61284-087-1
DOI :
10.1109/EMEIT.2011.6024055
