Title :
Research of pattern recognition based on the local structure topological relationship modeling
Author :
Qin, Yinxue ; Ma, Lin ; Li, Haifeng ; Wei, Qiuhua
Author_Institution :
Sch. of Comput. Sci. & Technol., Harbin Inst. of Technol., Harbin, China
Abstract :
The article is mainly to study a pattern recognition method which is based on the local structure feature and spatial topological relationship modeling. From the perspective of human cognition and bio-mimetic pattern recognition principle, analyzing things\´ local structural feature and the overall topological relationship we can achieve things "cognitive" by using ANN(Artificial Neural Network) and HMM(Hidden Markov Model), so in this paper we propose a common pattern recognition method. This method with a strong anti-interference and resistance to deformation reflects the characteristics of things which is similar to human cognition. The experimental results show that this method can apply to the general pattern recognition.
Keywords :
biomimetics; cognition; hidden Markov models; neural nets; pattern recognition; ANN; HMM; artificial neural network; biomimetic pattern recognition principle; deformation resistance; hidden Markov model; human cognition; local structure topological relationship modeling; spatial topological relationship modeling; Artificial neural networks; Feature extraction; Hidden Markov models; Pattern recognition; Predictive models; Training; Vectors;
Conference_Titel :
Audio, Language and Image Processing (ICALIP), 2012 International Conference on
Conference_Location :
Shanghai
Print_ISBN :
978-1-4673-0173-2
DOI :
10.1109/ICALIP.2012.6376704