您的位置:首页  > 论文页面

脉冲发放皮层模型实现特征捆绑

发表时间:2015-09-30  浏览量:1837  下载量:490
全部作者: 绽琨,滕骥才,王明英,周昊乐
作者单位: 兰州大学信息科学与工程学院
摘 要: 为实现图像基本特征捆绑,提出一种以脉冲发放皮层为模型的赋时矩阵实现特征捆绑方法。发现赋时矩阵可实现对原始图像信息较完整的表示,通过对脉冲发放皮层模型输出的一系列脉冲图像与对应的迭代时间相乘求和得到赋时矩阵。脉冲图像表达了原始输入图像的形状等空间几何信息,而迭代时间表达了原始输入图像的灰度信息,捆绑空间几何特征与灰度特征是对原始输入图像分离信息的整体表述。实验结果证明,脉冲发放皮层模型能较好地实现对形状等几何信息与灰度信息的捆绑。
关 键 词: 图像处理;脉冲发放皮层模型;特征捆绑;Gamma带神经振荡
Title: Spiking cortical model for feature binding
Author: ZHAN Kun, TENG Jicai, WANG Mingying, ZHOU Haole
Organization: School of Information Science & Engineering, Lanzhou University
Abstract: In order to achieve the basic features images, a feature binding method is proposed based on the time matrix of the spiking cortical model. It is found that the matrix can be expressed as a complete representation of the original image information. A natural image can be represented by the time matrix, which is obtained from the series spiking images and their iterative times. The series spiking images represent the edges and segments of the original image, and the iterative times represent the gray scale of the image. In this paper, the geometric information and the gray scale information of the image are integrated for binding features, and the separation features of the image are represented together. We conduct several experiments which demonstrate that spiking cortical model can provide a method for binding features such as segments, edges and intensity of the image.
Key words: image processing; spiking cortical model; feature binding; Gamma band oscillation
发表期数: 2015年9月第18期
引用格式: 绽琨,滕骥才,王明英,等. 脉冲发放皮层模型实现特征捆绑[J]. 中国科技论文在线精品论文,2015,8(18):1917-1923.
 
0 评论数 0
暂无评论
友情链接