摘要
传统的YOLOv3模型通常利用ImageNet、COCO等与测试集目标场景特征差异较大的数据集进行训练,存在对高分辨率遥感影像中复杂场景目标检测精度不高的问题。为解决这一问题,提出了一种对YOLOv3网络训练过程进行优化的方法。该方法基于迁移学习的思想,在YOLOv3网络训练中,通过生成与目标域更相似的增广数据集对模型进行预训练,实现了训练过程的优化,提高了目标初始预测的精度;利用目标域训练数据对预训练模型参数进行微调,完成了对网络的训练。利用公开的RSOD和DIOR遥感图像目标检测数据集的子集对飞机、运动场、立交桥三大类目标进行模型训练和检测实验,结果表明:本文提出的训练优化后的YOLOv3模型有效地提高了复杂城区场景中上述三类目标的检测精度。与传统的YOLOv3模型相比,三类目标的平均精度均值(mAP)提高了2%以上。
The traditional YOLOv3 model uses ImageNet and COCO datasets for training,in which the scene target characteristics are significantly different from those in test datasets,and leads to low detection accuracy of complex scene targets in high-resolution remote-sensing images.This paper optimizes the training process of the traditional YOLOv3 network using the idea of transfer learning.During the training of the YOLOv3 network,the model is pre-trained by generating an augmented dataset similar to the target domain.The training-optimized method improves the accuracy of the object boundary of target prediction.Also,the parameters of the pre-training model are fine-tuned using a training dataset from the target domain,thus,completing the whole training process of the network.The experiment on the detection of three types of object,including aircraft,playground,overpass,was carried out based on a subset of RSOD&DIOR dataset for remote sensing image object detection.The results show that the proposed YOLOv3 model effectively improves the detection accuracy of the three types of targets in complex urban scenes.The mean average precision of object detection using our model improved by 2%or more,compared with the traditional YOLOv3 model.
作者
杨耘
李龙威
高思岩
柏晗
江万成
Yang Yun;Li Longwei;Gao Siyan;Bai Han;Jiang Wancheng(School of Geological Engineering and Surveying and Mapping,Chang'an University,Xi'an,Shaanxi 710054,China)
出处
《激光与光电子学进展》
CSCD
北大核心
2021年第16期147-153,共7页
Laser & Optoelectronics Progress
基金
长安大学中央高校基本科研业务费(300102269205,300102269304)
国家重点研发计划(2018YFC1504805,2019YFC1509201)。
关键词
遥感
目标检测
高分辨率遥感影像
YOLOv3
迁移学习
remote sensing
object detection
high-resolution remote-sensing image
YOLOv3 network
transfer learning
作者简介
通信作者:李龙威,1049730716@qq.com。
