[1]向飞宇,张秀伟.基于卷积神经网络的人群计数算法研究[J].计算机技术与发展,2021,31(07):42-46.[doi:10. 3969 / j. issn. 1673-629X. 2021. 07. 007]
　XIANG Fei-yu,ZHANG Xiu-wei.Research on Crowd Counting Algorithm Based on Convolution Neural Network[J].,2021,31(07):42-46.[doi:10. 3969 / j. issn. 1673-629X. 2021. 07. 007]
点击复制

基于卷积神经网络的人群计数算法研究()

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
31
期数:
2021年07期
页码:
42-46
栏目:
图形与图像
出版日期:
2021-07-10

文章信息/Info

Title:
Research on Crowd Counting Algorithm Based on Convolution Neural Network
文章编号:
1673-629X(2021)07-0042-05
作者:
向飞宇; 张秀伟
西北工业大学 计算机学院,陕西 西安 710129
Author(s):
XIANG Fei-yu; ZHANG Xiu-wei
School of Computer Science,Northwestern Polytechnical University,Xi’an 710129,China
关键词:
人群计数; 全卷积神经网络; 注意力模型; 扩张卷积; 特征提取
Keywords:
crowd counting; fully convolutional network; attention model; expansion convolution; feature extraction
分类号:
TP183
DOI:
10. 3969 / j. issn. 1673-629X. 2021. 07. 007
摘要:
随着城市监控网络的完善,对人群图像的计数处理正产生巨大价值。 传统人群计数方法存在准确度低,无法处理高遮挡图像,受光影影响大等问题。 卷积神经网络在人群计数上表现出色,但仍存在精确度较低,无法排除背景图像干扰等问题。 为提高对复杂人群图像的感知能力,减少背景区域对统计的影响,并同时生成人群密度特征图像,在卷积神经网络的基础上增加空间与通道注意力模型,对不同通道和不同位置的图像赋予不同的权重以增加目标区域的影响力,同时更换全连接层为上采样层,输出与输入图像大小相同的人群密度特征图像。 实验中使用 ShanghaiTech 数据集以及 NWPU-Crowd 数据集进行训练与测试,在与 MCNN、CSRNet 等网络的比较结果中显示,使用了注意力模型与全卷积神经网络的算法在平均绝对值误差与均方误差两项数据上有较好的结果,表示该算法在高密度高遮挡的人群图像计数上有着更高的精确度。
Abstract:
As the growing of urban monitoring network,the counting of crowd image is proved to be of great value. Traditional methods of crowd counting have problems like low accuracy,inability to deal with high occlusion images and being greatly affected by light and shadow. Convolution neural network performs well in crowd counting,but problems like rather low accuracy and being influenced by background area still exist. To improve the perception of complex crowd image,decrease the influence of background area and create crowd density image in the same time,spatial-wise and channel-wise attention model are added to convolution neural network in order to give more weight to target area. Full connect layers are replaced by up-sampling layers to output crowd density image by the same sizeof the original input image. ShanghaiTech data set and NWPU - Crowd data set are used to train and validate the network, and the comparison among networks like MCNN and CSRNet shows that the proposed algorithm with attention model and fully convolutional network has better results in the mean absolute error and mean square error data,indicating that it has a higher accuracy in the high density and high-occlusion crowd image counting.

相似文献/References:

[1]施泽浩,赵启军.基于全卷积网络的目标检测算法[J].计算机技术与发展,2018,28(05):55.[doi:10.3969/j.issn.1673－629X.2018.05.013]
　SHI Ze-hao,ZHAO Qi-jun.Object Detection Algorithm Based on Fully Convolutional Neural Network[J].,2018,28(07):55.[doi:10.3969/j.issn.1673－629X.2018.05.013]
[2]朱 鹏,陈 虎*,李 科,等.一种轻量级的多尺度特征人脸检测方法[J].计算机技术与发展,2020,30(04):1.[doi:10. 3969 / j. issn. 1673-629X. 2020. 04. 001]
　ZHU Peng,CHEN Hu*,LI Ke,et al.A Face Detection Method with Lightweight and Multi-scale Feature[J].,2020,30(07):1.[doi:10. 3969 / j. issn. 1673-629X. 2020. 04. 001]
[3]翟正利,孙 霞,周 炜,等.基于全卷积神经网络的多目标显著性检测[J].计算机技术与发展,2020,30(08):34.[doi:10. 3969 / j. issn. 1673-629X. 2020. 08. 006]
　ZHAI Zheng-li,SUN Xia,ZHOU Wei,et al.Multi-objective Saliency Detection Based on Full Convolution Neural Network[J].,2020,30(07):34.[doi:10. 3969 / j. issn. 1673-629X. 2020. 08. 006]

更新日期/Last Update: 2021-07-10