《计算机应用》唯一官方网站 ›› 2020, Vol. 40 ›› Issue (2): 553-560.DOI: 10.11772/j.issn.1001-9081.2019101795

• 第七届CCF大数据学术会议 • 上一篇    下一篇

面向特定目标自识别的交通图像语义检索方法

赵一1,2, 段兴1(), 谢仕义1, 梁春林1,2   

  1. 1.广东海洋大学 数学与计算机学院,广东 湛江 524000
    2.湛江湾实验室 南海渔业大数据中心,广东 湛江 524000
  • 收稿日期:2019-08-30 修回日期:2019-10-24 接受日期:2019-10-24 发布日期:2019-10-31 出版日期:2020-02-10
  • 通讯作者: 段兴
  • 作者简介:赵一(1984—),男,湖北武汉人,讲师,博士,CCF会员,主要研究方向:软件工程、人工智能
    谢仕义(l963—),男,广东湛江人,教授,硕士,主要研究方向:软件工程
    梁春林(l975—),男,广东湛江人,副教授,硕士,主要研究方向:软件工程。
  • 基金资助:
    南方海洋科学与工程广东省实验室自主立项重大项目(ZJW-2019-08);广东海洋大学创新强校重大科研项目(GDOU2017052501)

Traffic image semantic retrieval method based on specific object self-recognition

Yi ZHAO1,2, Xing DUAN1(), Shiyi XIE1, Chunlin LIANG1,2   

  1. 1.College of Mathematics and Computer Science,Guangdong Ocean University,Zhanjiang Guangdong 524000,China
    2.Guangdong Zhanjiang Bay Laboratory,Fisheries Big Data Center of South China Sea,Zhanjiang Guangdong 524000,China
  • Received:2019-08-30 Revised:2019-10-24 Accepted:2019-10-24 Online:2019-10-31 Published:2020-02-10
  • Contact: Xing DUAN
  • About author:ZHAO Yi, born in 1984, Ph. D., lecturer. His research interests include software engineering, artificial intelligence.
    XIE Shiyi, born in 1963, M. S., professor. His research interests include software engineering.
    LIANG Chunlin, born in 1975, M. S., associate professor. His research interests include software engineering.
  • Supported by:
    the Guangdong Laboratory Autonomous Set Up Major Project for Southern Marine Science and Engineering(ZJW-2019-08);the Major Scientific Research Projects of Innovative and Strong University of Guangdong Ocean University(GDOU2017052501)

摘要:

为了从海量的道路交通图像中检索出违反交通法规的图像,提出了一种特定目标自识别的语义图像检索方法。首先,通过交通领域专家建立交通领域本体及道路交通规则描述;然后,通过卷积神经网络(CNN)对交通图像的特征进行提取,并结合改进的支持向量机决策树(SVM-DT)算法对图像特征进行分类的策略,对交通图像中的特定目标及目标间空间位置关系进行自动识别,并映射成为相应的本体实例及其对象之间的关联关系(规则实例);最后,利用本体实例和规则实例,通过推理得到语义检索结果。实验结果表明,相比关键字和本体交通图像语义检索方法,所提方法具有更高的准确率、召回率和检索效率。

关键词: 交通领域本体, 图像语义检索, 语义推理, 支持向量机决策树分类, 目标识别

Abstract:

In order to retrieve images of traffic violations from a large number of road traffic images, a semantic retrieval method based on specific object self-recognition was proposed. Firstly, road traffic domain ontology as well as road traffic rule description were established by experts in traffic domain. Secondly, traffic image features were extracted by Convolutional Neural Network (CNN), and combined with the strategy for classifying image features which is based on the proposed improved Support Vector Machine based Decision Tree (SVM-DT) algorithm, the specific objects and the spatial positional relationship between the objects in the traffic images were automatically recognized and mapped into the association relationship (rule instance) between the corresponding ontology instance and its objects. Finally, the image semantic retrieval result was obtained by reasoning based on ontology instances and rule instances. Experimental results show that the proposed method has higher accuracy, recall and retrieval efficiency compared to keyword and ontology traffic image semantic retrieval methods.

Key words: traffic ontology, image semantic retrieval, semantic reasoning, Support Vector Machine Decision Tree (SVM-DT) classification, object recognition

中图分类号: