摘要
针对景区人流科学监测、精细化管理的需求,文中对人流密度的估计方法进行了研究。该方法利用景区现有的视频监控设备获取视频图像,基于纹理分析的方法提取图像的灰度共生矩阵,借助能量、对比度、熵值、均匀度和相关性5个指标计算图像中的人流密度。利用不同时刻得出的人流密度数据建立时间序列,用自回归移动平均(ARIMA)建立人流密度估计模型,并用置信区间法评估模型的准确性。采用20个时刻的人流密度样本图像进行方法验证,结果表明该方法在估计人流密度值的平均误差为4.70%,估计出的人流密度变化曲线与实际曲线具有良好的一致性。
In view of the demand of scientific monitoring and refined management of human flow in scenic spots,this paper studies the estimation method of human flow density.Based on texture analysis,the gray level co-occurrence matrix of the image is extracted,and the human flow density in the image is calculated with the help of energy,contrast,entropy,uniformity and correlation.The time series is established by using the data of human flow density at different times,and the estimation model of human flow density is established by ARIMA,and the accuracy of the model is evaluated by confidence interval method.The results show that the average error of the method is 4.70%in the estimated value of human flow density,and the estimated curve of human flow density is in good agreement with the actual curve.
作者
马骞
MA Qian(Xi’an Vocational and Technical College of Aeronautics and Astronautics,Xi’an 710089,China)
出处
《信息技术》
2020年第8期34-38,共5页
Information Technology
基金
陕西省教育科学规划课题(SGH17V012)
西安航空职业技术学院教改课题(18XHJG-033)。
关键词
灰度共生矩阵
图像处理
时间序列
人流估计
ARIMA
gray level co-occurrence matrix
image processing
time series
human flow estimation
ARIMA
