沈哲1,
黄永平1,
王玉2,,
1.吉林大学计算机科学与技术学院 长春 130012
2.吉林大学应用技术学院 长春 130012
基金项目:智慧法院智能化服务技术研究及支撑平台开发(2018YFC0830100)，国家自然科学基金(61672259, 61876070)，国家自然科学基金青年科学基金(61602203)，吉林省科技发展计划重点科技研发项目(20180201064SF)，吉林省优秀青年人才基金(20180520020JH)
详细信息
作者简介:申铉京：男，1958年生，博士，教授，研究方向为图像处理与模式识别、多媒体信息安全、智能控制技术
沈哲：男，1995年生，硕士生，研究方向为图像处理与模式识别
黄永平：男，1964年生，博士，副教授，研究方向为图像处理与模式识别、智能控制与嵌入式系统
王玉：男，1983年生，博士，副教授，研究方向为图像处理与模式识别、多媒体信息技术
通讯作者:王玉 wangyu001@jlu.edu.cn
中图分类号:TN911.73; TP391计量
文章访问数:997
HTML全文浏览量:411
PDF下载量:74
被引次数:0
出版历程
收稿日期:2019-05-17
修回日期:2020-01-04
网络出版日期:2020-07-01
刊出日期:2020-09-27
Deep Convolutional Neural Network for Parking Space Occupancy Detection Based on Non-local Operation
Xuanjing SHEN1,Zhe SHEN1,
Yongping HUANG1,
Yu WANG2,,
1. College of Computer Science and Technology, Jilin University, Changchun 130012, China
2. College of Applied Technology, Jilin University, Changchun 130012, China
Funds:The Intelligent Court Intelligent Service Technology Research and Support Platform Development (2018YFC0830100), The National Natural Science Foundation of China (61672259, 61876070), The National Natural Science Foundation of China Youth Science Foundation (61602203), The Key Scientific and Technological R & D Projects of Jilin Province Science and Technology Development Plan(20180201064SF), Jilin Province Outstanding Young Talent Fund Project (20180520020JH)
摘要
摘要:随着城市交通智能化发展，准确高效地获取可用车位对于解决日益严峻的停车难问题至关重要。该文提出一种基于非局部操作的深度卷积神经网络车位占用检测算法。针对停车位图像特性，引入非局部操作，度量远距离像素间的相似性，直接获取边缘高频特征；使用小卷积核获取局部细节特征；以端到端的方式训练网络。实验中，通过设置不同卷积核尺寸和非局部模块层数，优化网络结构。实验结果表明，该文所提算法与传统的基于纹理特征的车位占用检测算法相比，无论在预测精度还是模型的泛化性能，均具有显著的优势。与当前广泛应用的基于局部特征提取的卷积神经网络相比，该算法具有较大的优势。在真实场景中，该算法同样具有较高精度，具备实际应用价值。
关键词:车位占用检测/
纹理特征/
卷积神经网络/
非局部操作
Abstract:With the intelligent development of urban traffic, accurate and efficient access to available parking spaces is essential to solve the increasingly difficult problem of parking difficulties. Therefore, this paper proposes a deep convolutional neural network parking occupancy detection algorithm based on non-local operation. For the image characteristics of parking spaces, non-local operations are introduced, the similarity between distant pixels is measured, and the high-frequency features of the edges are directly obtained. The local details are obtained by using small convolution kernels, and the network is trained in an end-to-end manner. In the experiment, the network structure is optimized by setting different convolution kernel sizes and non-local module layers. The experimental results show that compared with the traditional texture feature-based parking space occupancy detection algorithm, the proposed algorithm has significant advantages in both prediction accuracy and generalization performance of the model. At the same time, compared with the currently widely used convolutional neural network based on local feature extraction, the algorithm also has great advantages. In real scenes, the algorithm also has high precision and has practical application value.
Key words:Parking space occupancy detection/
Texture feature/
Convolutional neural network/
Non-local operation
PDF全文下载地址:
https://jeit.ac.cn/article/exportPdf?id=bf89e0be-a61a-4f13-93ca-04f6e771f9a2
