Probe: Noise-and-rotation Resistance of Hopfield Neural Network in Imaged Traffic Sign Recall

摘要：

This paper examines the noise and rotation resistance capacity of Hopfield neural network (HNN) based on the results from testing the four corrupted traffic sign images.In the study,signal-to-noise ratio,recall rate and pattern complexity are defined by the authors as the relevant specifications and employed to evaluate the recall performance of the network.The experimental results indicate that the HNN possesses significant recall capacity against the strong noise corruption,and certain restoring competence to the rotation.It is also found that combining noise with rotation or vice versa does not necessarily further compromise the corruption resistance capability of the HNN than the single type of corruption with noise or rotation.

关键词： Hopfield neural network traffic sign identification pattern complexity recall rate

作者: Ken Chen Shoujian Yang Jiaming He Celal Batur

作者单位: College of Information Science and Engineering Ningbo University Ningbo,Zhejiang 315211 China College of Engineering The University of Akron,Akron Akron,Ohio 44325,USA

会议类型: 国际会议

会议名称: 2011 3rd International Conference on Computer and Network Technology(ICCNT 2011)(2011第三届IEEE计算机与网络技术国际会议)

会议地点: 太原

会议语种:英文

页码: 24-28

在线出版日期: 2011-02-26（万方平台首次上网日期，不代表论文的发表时间）