基于YOLOV3改进的算法在对地目标检测中的应用

doi:10.3969/j.issn.2095-2198.2024.02.012

摘要
参考文献
相关文章
推荐阅读
Metrics

下载:

PDF (1618KB)
输出: BibTeX | EndNote (RIS)

摘要

对地目标检测因其视野旷阔，在交通安全、无人机侦察等领域应用广泛.对地目标具有数量多、尺度小的特点，导致检测精度不高、召回率低.针对上述问题，提出了一种基于YOLOV3改进的对地目标检测算法.首先，对数据集进行维度聚类，设计新的锚框尺寸，将先验数据融入模型，增强检测模型的有效性；其次，改进原有的网络模型，优化YOLOV3的目标预测框损失函数，使用CIoU损失代替原有的和方差损失，提高了目标预测框的回归稳定性.实验结果表明：改进的算法在VisDrone2018数据集上相对YOLOV3算法的召回率提高了11.2%，平均准确率均值提高了3.36%，改进的算法对对地目标检测的结果优于原本的YOLOV3算法.

	服务
	把本文推荐给朋友
	加入引用管理器
	E-mail Alert
	RSS
	（）
	作者相关文章

关键词: 对地目标')" href="#">

对地目标目标检测维度聚类 YOLOV3 CIoU

Abstract:

Ground target detection is widely used in fields such as traffic safety and drone reconnaissance due to its broad field of view.Due to the large number and small scale of ground targets,the detection accuracy is not high and the recall rate is low.In order to solve the above problems,an improved algorithm based on YOLOV3 is proposed.Firstly,the data set is clustered by dimension,and a new anchor box size is designed.The prior data is integrated into the model to enhance the effectiveness of the detection model.Secondly,the original network model is improved and the target prediction frame loss function of YOLOV3 is optimized.The original sum variance loss is replaced by CIoU loss,which improves the stability of the regression of target prediction box.The experimental results show that the recall rate of the improved algorithm is 11.2% higher than that of YOLOV3 algorithm,and the average accuracy rate(map)of the improved algorithm is increased by 3.36%.The improved algorithm effectively improves the recall rate and average accuracy of the detection algorithm,and is better than the original YOLOV3 algorithm in the performance of ground target detection.

Key words: ground target')" href="#">

ground target object detection dimensional clustering YOLOV3 CIoU

出版日期: 2024-04-30 发布日期: 2025-01-02 整期出版日期: 2024-04-30

基金资助:

国家重点研发计划（2018YFB1700200）

通讯作者: 王国刚

作者简介: 王奕然（1994—），男，江苏宿迁人，硕士研究生在读，主要从事图像处理方面的研究.

引用本文:

王奕然1, 王国刚1, 刘云鹏2.

基于YOLOV3改进的算法在对地目标检测中的应用 [J]. 沈阳化工大学学报, 2024, 38(2): 167-172.
WANG Yiran1, WANG Guogang1, LIU Yunpeng2.

Application of Improved YOLOV3 Algorithm in Ground Object Detection . Journal of Shenyang University of Chemical Technology, 2024, 38(2): 167-172.

链接本文:

https://xuebao.syuct.edu.cn/CN/10.3969/j.issn.2095-2198.2024.02.012 或 https://xuebao.syuct.edu.cn/CN/Y2024/V38/I2/167

［1］LOWE D G.Object.Recognition from Local Scale-Invariant Features［C］// /Proceedings of the Seventh IEEE International Conference on Computer Vision.Los Alamitos：IEEE Computer Society,1999:1150-1157.

［2］徐渊,许晓亮,李才年,等.结合SVM分类器与HOG特征提取的行人检测［J］.计算机工程,2016,42(1):56-60,65.

［3］金志浩,迟展,于宝刚.基于声发射技术的减速顶故障诊断［J］.沈阳化工大学学报,2020,34(2):147-152.

［4］张荣，李伟平，莫同．深度学习研究综述［J］.信息与控制,2018,47(4):385-397,410

［5］GIRSHICK R,DONAHUE J,DARRELL T,et al.Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation［C］// 2014 IEEE Conference on Computer Vision and Pattern Recognition．Los Alamitos：IEEE Computer Society,2014:580-587.

［6］GIRSHICK R.Fast R-CNN［C］// 2015 IEEE International Conference on Computer Vision（ICCV）．Los Alamitos：IEEE Computer Society,2015:1440-1448.

［7］REN S Q,HE K M,GIRSHICK R,et al.Faster R-CNN:Towards Real-Time Object Detection with Region Proposal Networks［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(6):1137-1149.

［8］HE K M,GKIOXARI G,DOLLR P,et al.Mask R-CNN［C］/2017 IEEE International Conference on Computer Vision（ICCV）．Los Alamitos：IEEE Computer Society,2017:2980-2988.

［9］DAI J,LI Y,HE K M,et al.R-FCN:Object Detection Via Region-Based Fully Convolutional Networks［C］// Proceedings of the 30th International Conference on Neural Information Processing Systems．New York：Curran Associates Inc,2016:379-387.

［10］LIU W,ANGUELOV D,ERHAN D,et al.SSD:Single Shot Multibox Detector［C］//Computer Vision–ECCV 2016．Cham：Springer，2016：21-37．

［11］REDMON J,DIVVALA S,GIRSHICK R,et al.You Only Look Once:Unified,Real-Time Object Detection［C］//2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).Los Alamitos：IEEE Computer Society,2016:779-788.

［12］REDMON J,FARHADI A.YOLOV3:an Incremental Improvement［EB/OL］.（2018-04-08）．https://arxiv.org/abs/1804.02767．

［13］REDMON J,FARHADI A.YOLOV3:an Incremental Improvement［EB/OL］.（2018-04-08）．https://arxiv.org/abs/1804.02767．

［13］He K M,Zhang X Y,Ren S Q,et al.Deep Residual Learning for Image Recognition［C］//2016 IEEE Conference on Computer Vision and Pattern Recognition（CVPR）．Los Alamitos：IEEE Computer Society,2016:770-778.

［14］LIN T Y,DOLL?R P,GIRSHICK R,et al.Feature Pyramid Networks for Object Detection［C］//2017 IEEE Conference on Computer Vision and Pattern Recognition（CVPR）.Los Alamitos：IEEE Computer Society,2017:936-944.

［15］JAIN A K.Data Clustering:50 Years Beyond K-Means［J］.Pattern Recognition Letters,2010,31(8):651-666.

［16］ARTHUR,D,VASSILVITSKII,S.K-Means++:the Advantages of CarefullSeeding［C］.Proceedings of the eighteenth annual ACM-SIAM symposiumon Discrete algorithms,Philadelphia：Society for Industrial and Applied Mathematics,2007:1027-1035.

［17］JIANG B R,LUO R X,MAO J Y,et al.Acquisition of Localization Confidence for Accurate Object Detection［C］//Computer Vision – ECCV 2018,Cham：Springer，2018:816-832．

［18］杨树国,和文静,刘银玲等..基于形态学的高斯模型和八邻域帧差法混合运动目标检测算法［J］.计算机与现代化,2019(7):32-36,42.

［19］YU X H,CHEN G A,CHENG S X.Dynamic Learning Rate Optimization of the Backpropagation Algorithm［J］.IEEE Transactions on Neural Networks,1995,6(3):669-677.

No related articles found!

No Suggested Reading articles found!

Viewed

Full text

Abstract

Cited

Shared

Discussed