2025年01月25日 星期六 首页   |    期刊介绍   |    编 委 会   |    投稿指南   |    期刊订阅   |    统合信息   |    联系我们
计量学报  2023, Vol. 44 Issue (1): 54-61    DOI: 10.3969/j.issn.1000-1158.2023.01.09
  光学计量 本期目录 | 过刊浏览 | 高级检索 |
基于改进Faster R-CNN的水母检测与识别算法
高美静1,李时雨2,刘泽昊2,张博智2,白洋2,关宁2,王萍2,常秋悦2
1.北京理工大学 集成电路与电子学院,北京 100081
2.燕山大学信息科学与工程学院河北省特种光纤与光纤传感重点实验室,河北 秦皇岛 066004
Jellyfish Detection and Recognition Algorithm Based on Improved Faster R-CNN
GAO Mei-jing1,LI Shi-yu2,LIU Ze-hao2,ZHANG Bo-zhi2,BAI Yang2,GUAN Ning2,WANG Ping2,CHANG Qiu-yue2
1. College of Integrated Circuits and Electronics, Beijing Institute of Technology, Beijing 100081, China
2. The Key Laboratory for Special Fiber and Fiber Sensor of Hebei Province, School of Information Science and Engineering, Yanshan University, Qinhuangdao, Hebei 066004, China
全文: PDF (5807 KB)   HTML (1 KB) 
输出: BibTeX | EndNote (RIS)      
摘要 提出一种基于改进Faster R-CNN水母检测与识别算法。首先,建立了包含7种水母的数据集;然后,针对ResNeXt(C=32)用于目标检测时出现计算量较大的问题,在保证精确度的前提下,将分支数C设置为8以降低计算量;最后,为解决水母检测时出现的检测精度低和小个体无法检测的问题,在残差网络中引入膨胀卷积。实验结果表明:该算法较VGG16、ResNet101、ResNeXt(C=32)和ResNeXt(C=8)方法,mAP值分别提高了3.15%、2.09%、3.01%和2.36%;F1-score分别提高了2.53%、1.99%、2.01%和2.31%;loss损失函数收敛值更优,收敛精度趋近于0。P-R曲线、可视化效果分析和水母视频检测的结果证明:该算法的水母检测准确率和水母检测数量明显优于其他算法,检测精度较高,基本可以达到实时监测的要求。
服务
把本文推荐给朋友
加入我的书架
加入引用管理器
E-mail Alert
RSS
作者相关文章
高美静
李时雨
刘泽昊
张博智
白洋
关宁
王萍
常秋悦
关键词 计量学;水母检测与识别;Faster R-CNN;ResNeXt;膨胀卷积残差网络    
Abstract:A jellyfish detection algorithm based on improved Faster R-CNN is proposed. Firstly, a data set containing 7 species of jellyfishes is established. Secondly, on the premise of ensuring the accuracy, the number of branches C is set to 8 to solve the problem that ResNeXt (C=32) has a high amount of calculation for target detection. Finally, to solve the problems of low detection accuracy and small individuals unable to be recognized, expansion convolution is introduced into the residual network. The experimental results shown that compared with VGG16, ResNet101, ResNeXt (C=32) and ResNeXt (C=8), the mAP value of the proposed algorithm increase by 3.15%, 2.09%, 3.01% and 2.36%. F1-score increase by 2.53%, 1.99%, 2.01% and 2.31%. Loss function convergence value of the proposed algorithm approach to 0. Results of P-R curve, visual analysis and video detection show that the accuracy and detection number of jellyfish by the proposed algorithm is the best, the proposed algorithm has high detection accuracy and can meet the requirements of real-time monitoring.
Key wordsmetrology    jellyfish detection and recognition    Faster R-CNN    ResNeXt    expansion convolution    residual network
收稿日期: 2022-02-21      发布日期: 2023-01-13
PACS:  TB96  
  TB973  
基金资助:国家自然科学基金(61971373);河北省自然科学基金(C2020203010);河北省博士在读研究生创新能力培养(CXZZBS2022148)
作者简介: 高美静(1977-),女,天津人,北京理工大学集成电路与电子学院教授,主要从事光电成像技术、光电检测技术、图像与视频处理技术研究。Email: gaomeijing@126.com
引用本文:   
高美静,李时雨,刘泽昊,张博智,白洋,关宁,王萍,常秋悦. 基于改进Faster R-CNN的水母检测与识别算法[J]. 计量学报, 2023, 44(1): 54-61.
GAO Mei-jing,LI Shi-yu,LIU Ze-hao,ZHANG Bo-zhi,BAI Yang,GUAN Ning,WANG Ping,CHANG Qiu-yue. Jellyfish Detection and Recognition Algorithm Based on Improved Faster R-CNN. Acta Metrologica Sinica, 2023, 44(1): 54-61.
链接本文:  
http://jlxb.china-csm.org:81/Jwk_jlxb/CN/10.3969/j.issn.1000-1158.2023.01.09     或     http://jlxb.china-csm.org:81/Jwk_jlxb/CN/Y2023/V44/I1/54
京ICP备:14006989号-1
版权所有 © 《计量学报》编辑部
地址:北三环东路18号(北京1413信箱)  邮编:100029 电话:(010)64271480
本系统由北京玛格泰克科技发展有限公司设计开发  技术支持:support@magtech.com.cn