基于改进Faster R-CNN的码头自动识别

常莉莉; 王贤敏; 王春胜

doi:10.11834/jrs.20220424

智能遥感处理与分析 | 浏览量 : 0 下载量: 452 CSCD: 1 更多指标

R-PDF
PDF
导出
分享
收藏
专辑

基于改进Faster R-CNN的码头自动识别
Automatic dock identification based on improved Faster R-CNN
2022年26卷第4期页码：752-765
纸质出版日期： 2022-04-07 ，
DOI： 10.11834/jrs.20220424

扫描看全文

常莉莉，王贤敏，王春胜.2022.基于改进Faster R-CNN的码头自动识别.遥感学报，26（4）： 752-765

C L L，W X M and W C S. 2022. Automatic Dock Identification Based on Improved Faster R-CNN. National Remote Sensing Bulletin， 26（4）：752-765
常莉莉，王贤敏，王春胜.2022.基于改进Faster R-CNN的码头自动识别.遥感学报，26（4）： 752-765 DOI： 10.11834/jrs.20220424.

C L L，W X M and W C S. 2022. Automatic Dock Identification Based on Improved Faster R-CNN. National Remote Sensing Bulletin， 26（4）：752-765 DOI： 10.11834/jrs.20220424.

摘要

码头自动识别能够为港口的建设与开发、海岸带地理信息的获取及海上军事实力的评估提供重要依据。然而由于码头普遍尺寸小、数量多、分布散乱，且受周围船舶、建筑等环境干扰严重，传统算法难以满足对高速发展的码头进行准确监测的需求，如何对码头目标进行准确识别成为亟需解决的问题。本文基于公开遥感数据集及Google Earth高分遥感影像构建了3种码头类型的数据集，并针对码头的尺寸特征和空间分布特征对Faster R-CNN算法进行了如下改进：（1）采用K-Means算法对候选框进行预设，使其大小更适应码头尺寸；（2）采用Soft-NMS算法代替NMS算法，以降低分布密集地区码头的误删率和漏检率。实验结果表明，本文改进的Faster R-CNN算法FKSN（Faster R-CNN+K-Means+Soft-NMS）识别精度达到92.6%，相较Faster R-CNN算法精度提高了8.3%。将码头目标识别结果和传统分类方法ISODATA、SSD及Faster R-CNN、Faster R-CNN+K-Means等目标提取模型的识别结果相对比，本文方法在虚警率和漏检率的评价指标表现最好，分别为3.2%和7.6%，说明本文方法对于各类码头目标识别具有更好的效果。基于改进Faster R-CNN算法的码头自动识别研究可以为码头的合理建设、规划及治理提供技术支持，为港口高效利用和军事实力分析提供有效途径。

Abstract

Automatic identification of docks can provide an important basis for the construction and development of ports

acquisition of coastal geographic information

and evaluation of maritime military strength. However

docks are characterized by small sizes

large quantities

and scattered distribution. Docks are also negatively affected by serious information interference of the surrounding environment

including ships and buildings. Traditional algorithms cannot easily meet the needs of accurate monitoring of rapidly developing docks. Accurate identification of dock targets has become an urgent problem to be solved. Based on the open remote sensing data sets and Google Earth high-resolution remote sensing images

the data sets of three types of docks are constructed

and the following improvements are made to the Faster R-CNN algorithm according to the size and spatial distribution characteristics of docks. (1) The K-means algorithm is used to preset the anchors

making the anchor sizes more suitable for the actual dock sizes. (2) Soft-NMS is used instead of NMS to reduce the rates of mistaken deletion and missed detection of dock borders in densely distributed areas. The experimental results show that the accuracy of the improved FKSN algorithm reached 92.6%

which is 6% higher than that of the Faster R-CNN algorithm. The final result of dock target recognition is compared with the ones of the traditional classification methods such as ISODATA

SSD

Faster R-CNN

and Faster R-CNN+K-Means. Among these approaches

the method suggested in this paper performs best in the evaluation indices of false alarm rate and omission rate

which are 3.2% and 7.6%

respectively. Thus

the proposed method has a better effect on the identification of various dock targets. The automatic dock identification algorithm based on the improved Faster R-CNN can provide technical support for reasonable construction

planning

and management of docks and provide effective approaches for efficient utilization and military strength analysis of docks.

关键词

Faster R-CNN码头自动识别K-means算法Soft-NMS算法高分遥感

Keywords

Faster R-CNNautomatic dock identificationK-means algorithmSoft-NMS algorithmhigh resolution remote sensing

references

Ball G H and Hall D J. 1965. ISODATA, A Novel Method of Data Analysis and Pattern Classification. California: Stanford Research Inst Menlo Park CA

Bengio Y. 2012. Practical recommendations for gradient-based training of deep architectures//Montavon G, Orr G B and Müller K R, eds. Neural Networks: Tricks of the Trade. Berlin, Heidelberg: Springer: 437-478 [DOI: 10.1007/978-3-642-35289-8_26http://dx.doi.org/10.1007/978-3-642-35289-8_26]

Bhagavathy S, Newsam S and Manjunath B S. 2002. Modeling object classes in aerial images using texture motifs//Proceedings of 2002 International Conference on Pattern Recognition. Quebec City: IEEE: 981-984 [DOI: 10.1109/ICPR.2002.1048470http://dx.doi.org/10.1109/ICPR.2002.1048470]

Bi Q, Tong X, Zhang J Y, Xu K, Zhang H and Qin K. 2019. Small harbor detection based on PLSA and BoW in high resolution remotely sensed imagery. Journal of Applied Sciences, 37(3): 301-312

毕奇, 童心, 张济勇, 许凯, 张涵, 秦昆. 2019. 基于PLSA和BoW的高分遥感影像小型港口检测. 应用科学学报, 37(3): 301-312 [DOI: 10.3969/j.issn.0255-8297.2019.03.001http://dx.doi.org/10.3969/j.issn.0255-8297.2019.03.001]

Bodla N, Singh B, Chellappa R and Davis L S. 2017. Soft-NMS—improving object detection with one line of code//Proceedings of 2017 IEEE International Conference on Computer Vision. Venice: IEEE: 5562-5570 [DOI: 10.1109/ICCV.2017.593http://dx.doi.org/10.1109/ICCV.2017.593]

Bovolo F, Marin C and Bruzzone L. 2013. A hierarchical approach to change detection in very high resolution SAR images for surveillance applications. IEEE Transactions on Geoscience and Remote Sensing, 51(4): 2042-2054 [DOI: 10.1109/TGRS.2012.2223219http://dx.doi.org/10.1109/TGRS.2012.2223219]

Burns J B, Hanson A R and Riseman E M. 1986. Extracting straight lines. IEEE Transactions on Pattern Analysis and Machine Intelligence, PAMI-8(4): 425-455 [DOI: 10.1109/TPAMI.1986.4767808http://dx.doi.org/10.1109/TPAMI.1986.4767808]

Girshick R, Donahue J, Darrell T and Malik J. 2014. Rich feature hierarchies for accurate object detection and semantic segmentation//Proceedings of 2014 IEEE Conference on Computer Vision and Pattern Recognition. Columbus: IEEE: 580-587 [DOI: 10.1109/CVPR.2014.81http://dx.doi.org/10.1109/CVPR.2014.81]

Girshick R. 2015. Fast R-CNN//Proceedings of 2015 IEEE International Conference on Computer Vision. Santiago: IEEE: 1440-1448 [DOI: 10.1109/ICCV.2015.169http://dx.doi.org/10.1109/ICCV.2015.169]

Han Y S, Ma S P, Zhang F and Li C H. 2020. Object detection of remote sensing airport image based on improved Faster R-CNN. Journal of Physics: Conference Series, 1601: 032010 [DOI: 10.1088/1742-6596/1601/3/032010http://dx.doi.org/10.1088/1742-6596/1601/3/032010]

Hartigan J A and Wong M A. 1979. Algorithm AS 136: A K-Means clustering algorithm. Journal of the Royal Statistical Society, 28(1): 100-108 [DOI: 10.2307/2346830http://dx.doi.org/10.2307/2346830]

He K M, Zhang X Y, Ren S Q and Sun J. 2016. Deep residual learning for image recognition//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE: 770-778 [DOI: 10.1109/CVPR.2016.90http://dx.doi.org/10.1109/CVPR.2016.90]

Kharchenko V and Chyrka I. 2018. Detection of airplanes on the ground using YOLO neural network//Proceedings of the 2018 IEEE 17th International Conference on Mathematical Methods in Electromagnetic Theory (MMET). Kyiv: IEEE: 294-297 [DOI: 10.1109/MMET.2018.8460392http://dx.doi.org/10.1109/MMET.2018.8460392]

Krizhevsky A, Sutskever I and Hinton G E. 2012. ImageNet classification with deep convolutional neural networks//Proceedings of Advances in Neural Information Processing Systems 25: 26th Annual Conference on Neural Information Processing Systems 2012. Lake Tahoe: [s.n.]: 1106-1114

Lechgar H, Bekkar H and Rhinane H. 2019. Detection of cities vehicle fleet using YOLO V2 and aerial images. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, XLII-4/W12: 121-126 [DOI: 10.5194/isprs-archives-XLII-4-W12-121-2019http://dx.doi.org/10.5194/isprs-archives-XLII-4-W12-121-2019]

Li J Y, Li X R and Zhao L Y. 2019. Docked ship detection based on edge line analysis and aggregation channel features. Acta Optica Sinica, 39(8): 0815004

黎经元, 厉小润, 赵辽英. 2019. 基于边缘线分析与聚合通道特征的港口舰船检测. 光学学报, 39(8): 0815004 [DOI: 10.3788/AOS201939.0815004http://dx.doi.org/10.3788/AOS201939.0815004]

Li K, Wan G, Cheng G, Meng L Q and Han J W. 2020. Object detection in optical remote sensing images: a survey and a new benchmark. ISPRS Journal of Photogrammetry and Remote Sensing, 159: 296-307 [DOI: 10.1016/j.isprsjprs.2019.11.023http://dx.doi.org/10.1016/j.isprsjprs.2019.11.023]

Li Z W, Guo H T, Shi L, Yu J T, Wu Z Y and Fang S L. 2018. A segmentation method of the coastal wharfs in remote sensing image based on structural features. Hydrographic Surveying and Charting, 38(5): 63-669

李正威, 郭海涛, 石朗, 喻金桃, 吴祯优, 方绍磊. 2018. 基于结构特征的遥感影像海岸码头分割方法. 海洋测绘, 38(5): 63-66 [DOI: 10.3969/j.issn.1671-3044.2018.05.015http://dx.doi.org/10.3969/j.issn.1671-3044.2018.05.015]

Liu C, Xiao Y Y, Yang J and Yin J J. 2016a. Harbor detection in Polarimetric SAR images based on the characteristics of parallel curves. IEEE Geoscience and Remote Sensing Letters, 13(10): 1400-1404 [DOI: 10.1109/LGRS.2016.2560944http://dx.doi.org/10.1109/LGRS.2016.2560944]

Liu T F. 2018. Current situation and countermeasures of China’s maritime rights and interests under the new situation. Journal of Hainan Tropical Ocean University, 25(3): 31-37

刘腾飞. 2018. 新形势下中国海洋权益现状与维护对策. 海南热带海洋学院学报, 25(3): 31-37 [DOI: 10.13307/j.issn.2096-3122.2018.03.05http://dx.doi.org/10.13307/j.issn.2096-3122.2018.03.05]

Liu W, Anguelov D, Erhan D, Szegedy C, Reed S, Fu C Y and Berg A C. 2016b. SSD: single shot multibox detector//Proceedings of the 14th European Conference on Computer Vision. Amsterdam: Springer: 21-37 [DOI: 10.1007/978-3-319-46448-0_2http://dx.doi.org/10.1007/978-3-319-46448-0_2]

Liu Y F, Pan H J, Wang G W and Qi C S. 2014. Artificially intellectual collection of island shoreline and dock based on IDL. Geomatics and Spatial Information Technology, 37(2): 96-98

刘亚飞, 潘洪军, 王广伟, 亓常松. 2014. 基于IDL的岛屿岸线及港口码头的人工智能提取. 测绘与空间地理信息, 37(2): 96-98 [DOI: 10.3969/j.issn.1672-5867.2014.02.027http://dx.doi.org/10.3969/j.issn.1672-5867.2014.02.027]

Long J, Shelhamer E and Darrell T. 2015. Fully convolutional networks for semantic segmentation//Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston: IEEE: 3431-3440 [DOI: 10.1109/CVPR.2015.7298965http://dx.doi.org/10.1109/CVPR.2015.7298965]

Ma C W and Bai X Z. 2015. Analysis of false alarm and missing alarm in conjunction assessment of space objects//Shen R J and Qian W P, eds. Proceedings of the 27th Conference of Spacecraft TTandC Technology in China. Berlin, Heidelberg: Springer: 477-487 [DOI: 10.1007/978-3-662-44687-4_43http://dx.doi.org/10.1007/978-3-662-44687-4_43]

Mandal D P, Murthy C A and Pal S K. 1996. Analysis of IRS imagery for detecting man-made objects with a multivalued recognition system. IEEE Transactions on Systems, Man, and Cybernetics-Part A: Systems and Humans, 26(2): 241-247 [DOI: 10.1109/3468.485750http://dx.doi.org/10.1109/3468.485750]

Neubeck A and Van Gool L. 2006. Efficient non-maximum suppression//Proceedings of the 18th International Conference on Pattern Recognition (ICPR'06). Hong Kong, China: IEEE: 850-855 [DOI: 10.1109/ICPR.2006.479http://dx.doi.org/10.1109/ICPR.2006.479]

Qu J S, Su C, Zhang Z W and Razi A. 2020. Dilated convolution and feature fusion SSD network for small object detection in remote sensing images. IEEE Access, 8: 82832-82843 [DOI: 10.1109/ACCESS.2020.2991439http://dx.doi.org/10.1109/ACCESS.2020.2991439]

Redmon J, Divvala S, Girshick R and Farhadi A. 2016. You only look once: unified, real-time object detection//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE: 779-788 [DOI: 10.1109/CVPR.2016.91http://dx.doi.org/10.1109/CVPR.2016.91]

Ren S Q, He K M, Girshick R B and Sun J. 2015. Faster R-CNN: towards real-time object detection with region proposal networks//Proceedings of Advances in Neural Information Processing Systems 28: Annual Conference on Neural Information Processing Systems 2015. Montreal: [s.n.]: 91-99 [DOI: 10.1109/TPAMI.2016.2577031http://dx.doi.org/10.1109/TPAMI.2016.2577031]

Rezatofighi H, Tsoi N, Gwak J Y, Sadeghian A, Reid I and Savarese S. 2019. Generalized intersection over union: A metric and a loss for bounding box regression//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach: IEEE: 658-666 [DOI: 10.1109/CVPR.2019.00075http://dx.doi.org/10.1109/CVPR.2019.00075]

Salakhutdinov R, Torralba A and Tenenbaum J. 2011. Learning to share visual appearance for multiclass object detection//Proceedings of CVPR 2011. Colorado Springs: IEEE: 1481-1488 [DOI: 10.1109/CVPR.2011.5995720http://dx.doi.org/10.1109/CVPR.2011.5995720]

Saputra D M, Saputra D and Oswari L D. 2020. Effect of distance metrics in determining K-Value in K-Means clustering using Elbow and Silhouette method//Proceedings of Sriwijaya International Conference on Information Technology and Its Applications (SICONIAN 2019). [s.l.]: Atlantis Press: 341-346 [DOI: 10.2991/aisr.k.200424.051http://dx.doi.org/10.2991/aisr.k.200424.051]

Smith L N. 2017. Cyclical learning rates for training neural networks//Proceedings of 2017 IEEE Winter Conference on Applications of Computer Vision (WACV). Santa Rosa: IEEE: 464-472 [DOI: 10.1109/WACV.2017.58http://dx.doi.org/10.1109/WACV.2017.58]

Taylor L and Nitschke G. 2018. Improving deep learning with generic data augmentation//Proceedings of 2018 IEEE Symposium Series on Computational Intelligence (SSCI). Bangalore: IEEE: 1542-1547 [DOI: 10.1109/SSCI.2018.8628742http://dx.doi.org/10.1109/SSCI.2018.8628742]

Uijlings J R R, Van De Sande K E A, Gevers T and Smeulders A W M. 2013. Selective search for object recognition. International Journal of Computer Vision, 104(2): 154-171 [DOI: 10.1007/s11263-013-0620-5http://dx.doi.org/10.1007/s11263-013-0620-5]

Wang J, Huang J, Wang M and Ming D P. 2019. Dock extraction from China’‍s Gaofen-2 multispectral imagery based on region-line primitive association analyses. International Journal of Remote Sensing, 40(10): 3878-3899 [DOI: 10.1080/01431161.2018.1553321http://dx.doi.org/10.1080/01431161.2018.1553321]

Wei J W. 2007. Research and implement of harbor detection in remote sensing image. Xi’an: Xidian University

魏军伟. 2007. 遥感图像中港口目标检测研究与实现. 西安: 西安电子科技大学 [DOI: 10.7666/d.y1246776http://dx.doi.org/10.7666/d.y1246776]

Wei S H, Chen H M, Zhu X J and Zhang H S. 2020. Ship detection in remote sensing image based on Faster R-CNN with dilated convolution//Proceedings of the 2020 39th Chinese Control Conference (CCC). Shenyang: IEEE: 7148-7153 [DOI: 10.23919/CCC50068.2020.9189467http://dx.doi.org/10.23919/CCC50068.2020.9189467]

Ye Q K, Huo H, Zhu T H and Fang T. 2017. Harbor detection in large-scale remote sensing images using both deep-learned and topological structure features//Proceedings of the 2017 10th International Symposium on Computational Intelligence and Design (ISCID). Hangzhou: IEEE: 218-222 [DOI: 10.1109/ISCID.2017.90http://dx.doi.org/10.1109/ISCID.2017.90]

Yu J T, Guo H T, Li C G and Li J. 2016. Coast dock extraction method based on waterline and perceptual organization//Proceedings of 2016 IEEE International Geoscience and Remote Sensing Symposium (IGARSS). Beijing: IEEE: 6201-6204 [DOI: 10.1109/IGARSS.2016.7730620http://dx.doi.org/10.1109/IGARSS.2016.7730620]

Zalpour M, Akbarizadeh G and Alaei-Sheini N. 2020. A new approach for oil tank detection using deep learning features with control false alarm rate in high-resolution satellite imagery. International Journal of Remote Sensing, 41(6): 2239-2262[DOI: 10.1080/01431161.2019.1685720http://dx.doi.org/10.1080/01431161.2019.1685720]

Zhang S M, Wu R Z, Xu K Y, Wang J M and Sun W W. 2019. R-CNN-based ship detection from high resolution remote sensing imagery. Remote Sensing, 11(6): 631 [DOI: 10.3390/rs11060631http://dx.doi.org/10.3390/rs11060631]

Zhou T Y. 2019. Research on object detection based on deep convolutional neural network. Harbin: Harbin Institute of Technology

周天怡. 2019. 基于深度卷积神经网络的目标检测算法研究. 哈尔滨: 哈尔滨工业大学

Zhu M. 2004. Recall, precision and average precision. Waterloo: University of Waterloo: 30

Zhu T H. 2018. Detection of compound object and core elements of airport in high resolution remote sensing images based on deep learning. Shanghai: Shanghai Jiaotong University

朱廷贺. 2018. 基于深度学习的高分辨率遥感影像复合目标及机场核心要素检测. 上海: 上海交通大学 [DOI: 10.27307/d.cnki.gsjtu.2018.001257http://dx.doi.org/10.27307/d.cnki.gsjtu.2018.001257]

Zitnick C L and Dollár P. 2014. Edge boxes: locating object proposals from edges//Proceedings of the 13th European Conference on Computer Vision. Zurich: Springer: 391-405 [DOI: 10.1007/978-3-319-10602-1_26http://dx.doi.org/10.1007/978-3-319-10602-1_26]

文章被引用时，请邮件提醒。

提交

改进Faster R-CNN的遥感图像多尺度飞机目标检测

中国林业遥感发展历程