首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于结构平衡网络的图像识别新方法
引用本文:高沛涛,王银河.基于结构平衡网络的图像识别新方法[J].教育技术导刊,2020,19(3):220-224.
作者姓名:高沛涛  王银河
作者单位:广东工业大学 自动化学院,广东 广州 510006
基金项目:国家自然科学基金项目(61673120)
摘    要:针对机器视觉中的图像识别问题,提出一种新的基于结构平衡网络的图像识别方法。从一种新的结构平衡网络(特殊的复杂网络)视角重新审视平面灰度图像,将图像灰度矩阵视为结构平衡网络的连接关系(权值)矩阵(像素点可不作为节点),通过Hadamard乘积变换得到像结构平衡矩阵,再利用像结构平衡矩阵的拓扑结构,产生新的图像识别特征参量对图像进行描述。该方法具有高速率、高识别率、尺度不变性及旋转不变性等特点。最后的实验仿真验证了该方法的有效性。

关 键 词:图像识别  结构平衡网络  Hadamard乘积变换  
收稿时间:2019-05-07

A Structural Balance Network-based New Methodology for Image Recognition
GAO Pei-tao,WANG Yin-he.A Structural Balance Network-based New Methodology for Image Recognition[J].Introduction of Educational Technology,2020,19(3):220-224.
Authors:GAO Pei-tao  WANG Yin-he
Institution:Faculty of Automation, Guangdong University of Technology, Guangzhou 510006, China
Abstract:Aiming at the problem of image recognition in machine vision, a new image recognition method based on structural balance network is proposed. The planar grayscale image is re-examined from the perspective of a new structural balanced network (special complex network), and the image gray matrix is regarded as the connection relation (weight) matrix of the structural balanced network (pixels may not be used as nodes). The image structural balance matrix is obtained by Hadamard product transformation, and then the topology structure of the structural balance matrix is used to generate a new image recognition feature parameter to describe the image. The method has the characteristics of high speed, high recognition rate, scale invariance and rotation invariance. Finally, the simulation example is given to show the validity of the method in this paper.
Keywords:face recognition    structural balance networks    Hadamard product transformation  
点击此处可从《教育技术导刊》浏览原始摘要信息
点击此处可从《教育技术导刊》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号