基于纹理特征的人流密度监测系统设计  

Design of flow density monitoring system based on texture features

作　　者：李亚娟[1] LI Ya juan(Xi'an Aerotechnical Polytechnic College,Xi'an 710089,China)

机构地区：[1]西安航空职业技术学院,陕西西安710089

出　　处：《电子设计工程》2020年第4期60-63,68,共5页Electronic Design Engineering

基　　金：陕西省社科基金项目(2017A017)。

摘　　要：针对旅游景区人流统计、人流疏导管理的需求,文中基于计算机图像识别技术对人流密度监测的相关算法进行了研究,设计了人流密度监测系统。系统包括了图像采集、特征提取、密度判别3个算法功能模块。系统在人群图像的特征提取时,使用灰度矩阵共生法获取图像的纹理特征信息,提取了灰度图像的能量、熵、对比度、相关性等特征向量,在降低分类器输入特征向量的复杂度的同时提高了密度判别的准确度。系统经过训练和测试,识别准确率达到95.4%以上,具有广阔的应用前景。Aiming at the demand of flow management in our country,this paper studies the relevant algorithms of flow density monitoring based on image recognition technology. In this paper, the monitoring system of human flow density is designed. The system includes image acquisition,feature extraction and density discrimination. When the system extracts the features of crowd images,the gray matrix co-occurrence method is used to obtain the texture feature information of the images. It reduces the complexity of input eigenvectors of classifiers and improves the accuracy of density discrimination.After training and testing,the recognition accuracy of the system reaches more than 95.4%,which has broad application prospects.

关 键 词：纹理特征 图像识别 人流密度 灰度矩阵 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]