| 599 | 12 | 28 |
| 下载次数 | 被引频次 | 阅读次数 |
近年来,随着遥感技术的不断发展,遥感图像数据集的数量和质量在大幅地提升。相比于传统的基于统计学习的目标检测方法,深度学习凭借其能够自主学习数据特征的优势,省去了繁琐的人工提取特征过程,适合处理海量数据。许多基于深度学习的目标检测模型在面对遥感图像中的小型目标(如船舶)时,不能准确地检测船舶的具体位置。针对该问题,使用基于FPN框架的特征提取方法,结合金字塔式的特征信息和可旋转的候选框设计,在不同场景下的光学遥感图像上进行了实验。验证结果表明,基于FPN特征提取方法的目标检测模型,可较好地完成多尺度、密集并倾斜排列目标的检测任务。
Abstract：With the continuous development of remote sensing technology in recent years, both the quantity and the quality of remote sensing datasets are greatly improved.Deep learning beats the traditional statistics-based learning method by its advantage of autonomous learning data features, and it helps to eliminate the tedious manual feature extraction process and is suitable for dealing with mass data.Many deep learning-based object detection models cannot accurately detect the specific position of small objects like ships in remote sensing images.To solve this problem, the feature extraction method based on FPN framework, combined with the pyramid feature information and the rotatable candidate frame design is used and the experiments are carried out on optical remote sensing images in different scenes.The experimental results show that the target detection model based on FPN feature extraction method can better complete the detection task of multi-scale dense and obliquely arranged targets.
[1] 王桥，刘思含.国家环境遥感监测体系研究与实现[J].遥感学报，2016,20(5):1161-1169.
[2] 徐冠华，柳钦火，陈良富，等.遥感与中国可持续发展：机遇和挑战[J].遥感学报，2016,20(5):679-688.
[3] LIU Y,CUI H Y,ZHENG K,et al.Ship Detection and Classification on Optical Remote Sensing Images Using Deep Learning[J].Itm Web of Conferences,2017,12:05012.
[4] GALLEGO A J,PERTUSA A,GIL P.Automatic Ship Classification from Optical Aerial Images with Convolutional Neural Networks[J].Remote Sensing,2018,10(4):511.
[5] ETTENA V.You Only Look Twice:Rapid Multi-Scale Object Detection in Satellite Imagery[J].arXiv preprint arXiv:1805.09512,2018.
[6] NIE G H,PENG Z,XIN N,et al.Ship Detection Using Transfer Learned Single Shot Multi Box Detector[J].Itm Web Conferences,2017,12:01006.
[7] WEI S H,CHEN H M,ZHU X J et al.Remote Sensing Image Ship Detection Based on Faster R-CNN with Dilated Convolution[C]//2020 39th Chinese Control Conference (CCC).Shenyang:IEEE,2020:7148-7153.
[8] LIN T,DOLLáR P,GIRSHICK R,et al.Feature Pyramid Networks for Object Detection[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Honolulu:HI,2017:936-944.
[9] KANG M,JI K,LENG X,et al.Contextual Region-Based Convolutional Neural Network with Multilayer Fusion for SAR Ship Detection[J].Remote Sensing,2017,9(8):860.
[10] DENG C,WANG M,LIU L,et al.Extended Feature Pyramid Network for Small Object Detection[J].IEEE Transaction on Multimedia,2021(99):1.
[11] GHIASI G,LIN T Y,LE Q V .NAS-FPN:Learning Scalable Feature Pyramid Architecture for Object Detection[C]//2019 IEEE/CVF Conference on Computer Visionand Pattern Recognition (CVPR).Long Beach:IEEE,2019:7029-7038.
[12] RUMELHART D,HINTON G E,WILLIAMS R J.Learning Representations by Back Propagating Errors[J].Nature,1986,323(6088):533-536.
[13] HINTON G E,SALAKHUTDINOV R R.Reducing the Dimensionality of Data with Neural Networks[J].Science,2006,313(5786):504-507.
[14] KRIZHEVSKY A,SUTSKEVER I,HINTON G.ImageNet Classification with Deep Convolutional Neural Networks[J].Advances in Neural Information Processing Systems,2017,60(6):84-90.
[15] GIRSHICK R.Fast R-CNN[C]//2015 IEEE International Conference on Computer Vision (ICCV).Santiago:IEEE,2015:1440-1448.
[16] REN S,HE K,GIRSHICK R,et al.Faster R-CNN:Towards Real-Time Object Detection with Region Proposal Networks[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2017,39(6):1137-1149.
[17] LIU L,PAN Z,LEI B.Learning a Rotation Invariant Detector with Rotatable Bounding Box[J].arXiv preprint arXiv:1711.09405,2017.
[18] HE K M,GKIOXARI G,DOLLáR P,et al.Mask R-CNN[C]//International Conference on Computer Vision (ICCV).Venice:IEEE,2017:2980-2988.
[19] YANG X,SUN H,SUN X,et al.Position Detection and Direction Prediction for Arbitrary-Oriented Ships via Multitask Rotation Region Convolutional Neural Network[J].IEEE Access,2018(6):50839-50849.
[20] LIU W,ANGUELOV D,ERHAN D,et al.SSD:Single Shot MultiBox Detector[C]//European Conference on Computer Vision.Cham:Springer,2016:21-37.
[21] REDMON J,DIVVALA S,GIRSHICK R,et al.You Only Look Once:Unified,Real-Time Object Detection[C]//Computer Vision & Pattern Recognition.Singapore:IEEE,2016:1020-1026.
基本信息:
DOI：
中图分类号:U675.79;TP751
引用信息:
[1]牛瑞欣,赵正健,张时源等.FPN特征提取的光学遥感图像船舶检测[J].无线电工程,2021,51(11):1296-1302.
基金信息:
国家自然科学基金资助项目(61762025)~~