改进YOLOv3遥感小目标检测算法  

Improved YOLOv3 Algorithm for Remote Sensing Detection of Small Targets

作　　者：许成林 黄宇博 赵舵[1] XU Cheng-Lin;HUANG Yu-Bo;ZHAO Duo(Southwest Jiaotong University,Chengdu 611756,China)

机构地区：[1]西南交通大学,成都611756

出　　处：《计算机系统应用》2023年第7期179-187,共9页Computer Systems & Applications

摘　　要：针对目前在遥感目标检测领域广泛使用的YOLOv3算法存在对小目标物体的特征表达能力不足,检测效果不好的问题,本文提出一种改进的YOLOv3小目标检测算法.首先,引入全局信息注意力机制并改进特征提取网络和特征金字塔结构,提高模型小目标特征提取能力和检测能力;其次,对数据集进行单尺度Retinex融合特征增强,提高模型对小目标特征的学习效果;最后,使用自适应锚框优化算法对anchors进行优化,提高anchors和目标的匹配程度.选用遥感数据集RSOD进行实验,本文算法的全类平均精度为92.5%,相比经典YOLOv3算法,提高10.1%,对遥感小目标的检测效果得到明显提升.As YOLOv3,an algorithm widely used in the field of remote sensing target detection,has insufficient feature expression ability for small targets and a poor detection effect,an improved YOLOv3 algorithm for small target detection is proposed.Firstly,the global context(GC)attention mechanism is introduced,and the feature extraction network and feature pyramid networks(FPN)are improved to enhance the small-target feature extraction ability and detection ability of the model.Secondly,single-scale Retinex(SSR)fusion feature enhancement is applied to the dataset to improve the model’s learning effect of small target features.Finally,the adaptive anchor box optimization(AABO)algorithm is adopted to optimize anchors and better match anchors and targets.The experimental results on the remote sensing dataset RSOD show that the mean average precision(mAP)of the proposed algorithm is 92.5%,which is improved by 10.1%compared with that of the classic YOLOv3 algorithm,and the detection effect of small remote sensing targets is significantly improved.

关 键 词：小目标检测 YOLOv3 特征增强 锚框优化 深度学习 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置] TP183[自动化与计算机技术—控制科学与工程]