A neural network model for visual selection and shifting

doi:10.1142/S0219635216500205

CORC > 计算技术研究所 > 中国科学院计算技术研究所 > 中国科学院计算技术研究所期刊论文 > 英文

	A neural network model for visual selection and shifting
	Qiao, Yuanhua 1; Liu, Xiaojie 1; Miao, Jun 2; Duan, Lijuan 3
刊名	JOURNAL OF INTEGRATIVE NEUROSCIENCE
	2016-09-01
卷号	15 期号:3 页码:321-335
关键词	Visual selection and shifting neural network periodic activity synchronization mapping dynamic system
ISSN号	0219-6352
DOI	10.1142/S0219635216500205
英文摘要	In this paper, a two-layer network is built to simulate the mechanism of visual selection and shifting based on the mapping dynamic model for instantaneous frequency. Unlike the differential equation model using limit cycle to simulate neuron oscillation, we build an instantaneous frequency mapping dynamic model to describe the change of the neuron frequency to avoid the difficulty of generating limit cycle. The activity of the neuron is rebuilt based on the instantaneous frequency and in this work, we use the first layer of neurons to implement image segmentation and the second layer of neurons to act as visual selector. The frequency of the second neuron (central neuron) is always changing, while central neuron resonates with the neurons corresponding to an object, the object is selected, then with the central neuron frequency changing, the selected object loses attention, the process goes on.
资助项目	Natural Science Foundation of China[61572004] ; Natural Science Foundation of China[61175115] ; Natural Science Foundation of China[60702030] ; Natural Science Foundation of China[61370113] ; Natural Science Foundation of China[60970087] ; Natural Science Foundation of China[61070149] ; Natural Science Foundation of China[61272320] ; Beijing Natural Science Foundation[4152005] ; Jing-Hua Talents Project of Beijing University of Technology[2014-JH-L06] ; International Communication Ability Development Plan for Young Teachers of Beijing University of Technology[2014-16]
WOS研究方向	Neurosciences & Neurology
语种	英语
出版者	IMPERIAL COLLEGE PRESS
WOS记录号	WOS:000388756700004
内容类型	期刊论文
源URL	[http://119.78.100.204/handle/2XEOYT63/7882]
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Qiao, Yuanhua
作者单位	1.Beijing Univ Technol, Coll Appl Sci, Beijing 100124, Peoples R China 2.Chinese Acad Sci, Inst Comp Technol, Key Lab Intelligent Informat Proc, Beijing 100190, Peoples R China 3.Beijing Univ Technol, Coll Comp Sci, Beijing 100124, Peoples R China
推荐引用方式 GB/T 7714	Qiao, Yuanhua,Liu, Xiaojie,Miao, Jun,et al. A neural network model for visual selection and shifting[J]. JOURNAL OF INTEGRATIVE NEUROSCIENCE,2016,15(3):321-335.
APA	Qiao, Yuanhua,Liu, Xiaojie,Miao, Jun,&Duan, Lijuan.(2016).A neural network model for visual selection and shifting.JOURNAL OF INTEGRATIVE NEUROSCIENCE,15(3),321-335.
MLA	Qiao, Yuanhua,et al."A neural network model for visual selection and shifting".JOURNAL OF INTEGRATIVE NEUROSCIENCE 15.3(2016):321-335.