电子学报 ›› 2020, Vol. 48 ›› Issue (3): 590-601.DOI: 10.3969/j.issn.0372-2112.2020.03.024

• 综述评论 • 上一篇    下一篇

基于深度学习的小目标检测研究与应用综述

刘颖1,2,3, 刘红燕3, 范九伦3, 公衍超1,3, 李莹华3, 王富平1,3, 卢津1,3   

  1. 1. 电子信息现场勘验应用技术公安部重点实验室, 陕西西安 710121;
    2. 陕西省无线通信与信息处理技术国际合作研究中心, 陕西西安 710121;
    3. 西安邮电大学图像与信息处理研究所, 陕西西安 710121
  • 收稿日期:2019-09-29 修回日期:2019-11-20 出版日期:2020-03-25
    • 通讯作者:
    • 刘颖
    • 作者简介:
    • 刘红燕 女,主要研究方向为小目标检测.E-mail:hy_liu1314@163.com;范九伦 男,主要研究方向为模糊信息处理、图像处理、信息安全等.E-mail:jiulunf@xupt.edu.cn;公衍超 男,主要研究方向为视频编码及码率控制等.E-mail:gongyanchao@xupt.edu.cn;李莹华 女,主要研究方向为图像安全与图像超分辨率重建等.E-mail:liyinghua@xupt.edu.cn;王富平 男,主要研究方向为图像特征提取及图像检索等.E-mail:wfp1608@163.com;卢津 男,主要研究方向为三维痕迹图像数据采集及处理等.E-mail:lujin@xupt.edu.cn
    • 基金资助:
    • 陕西省国际合作交流项目 (No.2018KW-003); 国家自然科学基金 (No.61671377); 国家重点研发计划 (No.2017YFC0803805)

A Survey of Research and Application of Small Object Detection Based on Deep Learning

LIU Ying1,2,3, LIU Hong-yan3, FAN Jiu-lun3, GONG Yan-chao1,3, LI Ying-hua3, WANG Fu-ping1,3, LU Jin1,3   

  1. 1. Key Laboratory of Electronic Information Application Technology for Scene Investigation, Ministry of Public Security, Xi'an, Shaanxi 710121, China;
    2. International Joint Research Center for Wireless Communication and Information Processing, Shaanxi, Xi'an, Shaanxi 710121, China;
    3. Center for Image and Information Processing, Xi'an University of Posts and Telecommunications, Xi'an, Shaanxi 710121, China
  • Received:2019-09-29 Revised:2019-11-20 Online:2020-03-25 Published:2020-03-25
    • Corresponding author:
    • LIU Ying
    • Supported by:
    • Project of International Coorperation and Exchanges of Shaanxi Province (No.2018KW-003); National Natural Science Foundation of China (No.61671377); National Key Research and Development Program of China (No.2017YFC0803805)

摘要: 目标检测在基于传统手工特征及深度学习算法上已经取得较大发展,然而针对小目标检测的研究近几年才开始出现,研究成果较少,且大都是在已有目标检测算法基础上进行改进,以提高小目标检测的检测精度.小目标像素点少,本身携带的特征少,多次下采样后就更难进行特征提取,因而小目标检测面临极大挑战.小目标检测在自动驾驶、遥感图像检测、刑侦等领域都有广泛应用需求,对于小目标检测技术的研究有重要的实用价值.本文对小目标检测的现有研究成果进行了详细综述.首先,将现有算法按照检测需要的阶段数分为一阶段、二阶段、多阶段,描述了RetinaNet、CornerNet-Lite、特征金字塔网络(Feature Pyramid Network,FPN)等算法的原理并进行了对比分析.其次,本文描述了小目标检测技术在不同领域的应用情况,并汇总了MS COCO、PASCAL VOC、DOTA、KITTI等数据集及算法性能评价指标.最后,总结了小目标检测面临的挑战,并展望了未来的研究方向.

关键词: 小目标检测, 尺度变换, 特征金字塔, 深度学习, 特征提取, 卷积神经网络

Abstract: Object detection has been well studied based on the traditional manual features and deep learning algorithm. However, the research on small object detection has just begun in recent years and there are little research outcome available. Furthermore, most of the methods proposed are based on the traditional object detection algorithms with certain modifications so as to improve the accuracy of small object detection. Small object contains fewer pixels and has less features, and it is even harder to extract object features after down sampling. Hence, small object detection is a challenging task. Small object detection has a wide range of application requirements in the fields of automatic driving, remote sensing image detection, and criminal investigation. It has important practical value for the research of small object detection technology. In this paper, the existing research results of small object detection are summarized. Firstly, the existing algorithms are classified into one stage, two stages and multi-stages according to the number of stages for detection. The principles of RetinaNet、CornerNet-Lite、feature pyramid network (FPN) and other algorithms are described and compared. Secondly, this paper describes the application of small object detection technology in different fields, and summarizes the data sets such as MS COCO、PASCAL VOC、DOTA、KITTI and algorithm performance evaluation indicators. Finally, the challenges faced by small object detection are concluded, and the future research directions are prospected.

Key words: small object detection, scale variation, feature pyramid, deep learning, feature extraction, convolutional neural network

中图分类号: