基于级联卷积神经网络的港口多方向舰船检测与分类

doi:10.3969/j.issn.1001-506X.2020.09.04

系统工程与电子技术 ›› 2020, Vol. 42 ›› Issue (9): 1903-1910.doi: 10.3969/j.issn.1001-506X.2020.09.04

基于级联卷积神经网络的港口多方向舰船检测与分类

孙嘉赤(), 邹焕新*(), 邓志鹏(), 李美霖(), 曹旭(), 马倩()

国防科技大学电子科学学院, 湖南长沙 410073

收稿日期:2019-09-20 出版日期:2020-08-26 发布日期:2020-08-26
通讯作者: 邹焕新 E-mail:s.jcsome@gmail.com;hxzou2008@163.com;dzp_whu@163.com;summit_mll@qq.com;1135459767@qq.com;2233809618@qq.com
作者简介:孙嘉赤(1996-),男,硕士研究生,主要研究方向为高分辨率光学遥感图像舰船目标检测。E-mail:s.jcsome@gmail.com|邓志鹏(1990-),男,博士研究生,主要研究方向为计算机视觉、图像处理、模式识别等。E-mail:dzp_whu@163.com|李美霖(1995-),女,硕士研究生,主要研究方向为极化SAR图像地物分类、模式识别。E-mail:summit_mll@qq.com|曹旭(1996-),男,硕士研究生,主要研究方向为SAR图像和光学图像目标检测分类与识别。E-mail:1135459767@qq.com|马倩(1996-),女,硕士研究生,主要研究方向为多源遥感数据变化检测。E-mail:2233809618@qq.com
基金资助:
国家自然科学基金(61331015)

Oriented inshore ship detection and classification based on cascade RCNN

Jiachi SUN(), Huanxin ZOU*(), Zhipeng DENG(), Meilin LI(), Xu CAO(), Qian MA()

College of Electronic Science and Technology, National University of Defense Technology, Changsha 410073, China

Received:2019-09-20 Online:2020-08-26 Published:2020-08-26
Contact: Huanxin ZOU E-mail:s.jcsome@gmail.com;hxzou2008@163.com;dzp_whu@163.com;summit_mll@qq.com;1135459767@qq.com;2233809618@qq.com

摘要/Abstract

摘要：

港口舰船目标自动检测的定位和类型分类是一个重要而具有挑战性的问题。针对高分辨率光学遥感影像中多方向性排列密集的近岸舰船目标定位和识别困难的问题,提出基于级联区域卷积神经网络和手工提取特征相结合的近岸舰船检测识别框架。首先,使用级联的区域卷积神经网络对舰船位置进行粗定位并对类别进行估计，得到一系列粗定位的垂直预测框。然后,设计一个可以准确定位舰船的斜框旋转回归器,其将第一阶段所得粗定位垂直矩形框转变为带方向的斜矩形框。最后,使用非极大值抑制的方法去除冗余的预测框。实验采用谷歌地球上采集的数据集进行训练和预测,实验结果表明所提算法在精准率和召回率上均具有较大优势。

关键词: 港口舰船检测, 斜框标注, 舰船分类, Canny边缘检测, Hough直线检测

Abstract:

Automatic inshore ship recognition, including target localization and type classification, is an important and challenging problem. However, arbitrarily rotated ships are always moored inshore densely. This makes it very difficult to recognize and locate ship targets. To resolve this problem, a multiclass oriented ship localization and recognition framework is proposed based on a cascade region convolutional neural network (RCNN) and feature designed manually. Firstly, cascade RCNN is adopted to localize and classify the positive regions of ships-a set of bounding boxes (BBox). Secondly, a novel procedure which transforms a bounding box to a rotated bounding box is designed and applied to each BBox. Finally, non-maximum suppression (NMS) is adopted to remove the redundant rotated BBoxes (RBoxes). Extensive experimental results conducted on the dataset collected from Google Earth demonstrate the effectiveness of the proposed approach, compared to the other approaches.

Key words: inshore ship detection, rotated bounding box annotation, ship classification, Canny edge detection, Hough line detection

中图分类号:

TN957.52

孙嘉赤, 邹焕新, 邓志鹏, 李美霖, 曹旭, 马倩. 基于级联卷积神经网络的港口多方向舰船检测与分类[J]. 系统工程与电子技术, 2020, 42(9): 1903-1910.

Jiachi SUN, Huanxin ZOU, Zhipeng DENG, Meilin LI, Xu CAO, Qian MA. Oriented inshore ship detection and classification based on cascade RCNN[J]. Systems Engineering and Electronics, 2020, 42(9): 1903-1910.

图/表 12

图1

图2

图3

图4

图5

表1

图6

图7

图8

图9

图10

表2

参考文献 30

1	李晓斌, 江碧涛, 杨渊博, 等. 光学遥感图像目标检测技术综述[J]. 航天返回与遥感, 2019, 40 (4): 95- 104.
	LI X B , JIANG B T , YANG Y B , et al. A survey on object detection technology in optical remote sensing images[J]. Spacecraft Recovery & Remote Sensing, 2019, 40 (4): 95- 104.
2	LIU Z K , WANG H Z , WENG L B , et al. Ship rotated bound-ing box space for ship extraction from high-resolution optical satellite images with complex backgrounds[J]. IEEE Geoscience and Remote Sensing Letters, 2016, 13 (8): 1074- 1078. doi: 10.1109/LGRS.2016.2565705
3	MA L, GUO J, WANG Y Q, et al. Ship detection by salient convex boundaries[C]//Proc.of the International Congress on Image and Signal Processing, 2010: 202-205.
4	LIU G , ZHANG Y S , ZHENG X W , et al. A new method on inshore ship detection in high-resolution satellite images using shape and context information[J]. IEEE Geoscience and Remote Sensing Letters, 2014, 11 (3): 617- 621. doi: 10.1109/LGRS.2013.2272492
5	LI S , ZHOU Z Q , WANG B , et al. A novel inshore ship detection via ship head classification and body boundary determination[J]. IEEE Geoscience and Remote Sensing Letters, 2016, 13 (12): 1920- 1924. doi: 10.1109/LGRS.2016.2618385
6	刘松涛, 姜康辉, 刘振兴. 基于区域协方差和目标度的航空侦察图像舰船目标检测[J]. 系统工程与电子技术, 2019, 41 (5): 972- 980.
	LIU S T , JIANG K H , LIU Z X , et al. Ship target detection of aerial reconnaissance image based on region covariance and objectness[J]. Systems Engineering and Electronics, 2019, 41 (5): 972- 980.
7	YANG F , XU Q Z , LI B , et al. Ship detection from optical sate-llite images based on saliency segmentation and structure-LBP feature[J]. IEEE Geoscience and Remote Sensing Letters, 2017, 14 (5): 602- 606. doi: 10.1109/LGRS.2017.2664118
8	黎经元, 厉小润, 赵辽英. 基于边缘线分析与聚合通道特征的港口舰船检测[J]. 光学学报, 2019, 39 (8): 225- 234.
	LI J Y , LI X R , ZHAO L Y , et al. Docked ship detection based on edge line analysis and aggregation channel features[J]. Acta Optical Sinica, 2019, 39 (8): 225- 234.
9	REN S Q , HE K M , ROSS G , et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE Tran.on Pattern Analysis and Machine Intelligence, 2017, 39 (6): 1137- 1149. doi: 10.1109/TPAMI.2016.2577031
10	ZOU Z X, SHI Z W. Ship detection in spaceborne optical image with SVD networks[C]//Proc.of the IEEE Trans.on Geoscience and Remote Sensing, 2016, 54(10): 5832-5845.
11	TANG T Y , ZHOU S L , DENG Z P , et al. Vehicle detection in aerial images based on region convolutional neural networks and hard negative example mining[J]. Sensors, 2017, 17 (2): 336- 348.
12	DENG Z P , SUN H , ZHOU S L , et al. Toward fast and accurate vehicle detection in aerial images using coupled region-based convolutional neural networks[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sens-ing, 2017, 10 (8): 3652- 3664. doi: 10.1109/JSTARS.2017.2694890
13	DENG Z P , SUN H , ZHOU S L , et al. Multi-scale object detection in remote sensing imagery with convolutional neural networks[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2018, 145, 3- 22. doi: 10.1016/j.isprsjprs.2018.04.003
14	LIN H N , SHI Z W , ZOU Z X , et al. Fully convolutional network with task partitioning for inshore ship detection in optical remote sensing images[J]. IEEE Geoscience and Remote Sens-ing Letters, 2017, 14 (10): 1665- 1669. doi: 10.1109/LGRS.2017.2727515
15	LIU Z K, HU J G, WENG L B, et al. Rotated region based CNN for ship detection[C]//Proc.of the IEEE International Conference Image Process, 2017: 900-904.
16	XIA G S, BAI X, DING J, et al. DOTA: a large-scale dataset for object detection in aerial images[C]//Proc.of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018: 3974-3983.
17	余东行, 郭海涛, 赵传, 等. 直线特征辅助的靠岸舰船检测[J]. 测绘科学技术学报, 2019, 36 (3): 275- 280, 286.
	YU D X , GUO H T , ZHAO C , et al. Detection of ship docked in harbor assisted with line feature[J]. Journal of Geomatics Science and Technology, 2019, 36 (3): 275- 280, 286.
18	JIAN D, XUE N, LONG Y, et al. Learning ROI transformer for detecting oriented objects in aerial images[C]//Proc.of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019: 2844-2853.
19	XIAO X W, ZHOU Z Q, WANG B, et al. Accurate ship detection via paired semantic segmentation[C]//Proc.of the Chinese Control and Decision Conference, 2019: 5990-5994.
20	ZHANG Z H , GUO W W , ZHU S N , et al. Toward arbitrary-oriented ship detection with rotated region proposal and discrimination networks[J]. IEEE Geoscience and Remote Sensing Letters, 2018, 15 (11): 1745- 1749. doi: 10.1109/LGRS.2018.2856921
21	LIU W C , MA L , CHEN H , et al. Arbitrary-Oriented ship detection framework in optical remote-sensing images[J]. IEEE Geoscience and Remote Sensing Letters, 2018, 15 (6): 937- 941. doi: 10.1109/LGRS.2018.2813094
22	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. doi: 10.1109/ACCESS.2018.2869884
23	陈慧元, 刘泽宇, 郭炜炜, 等. 基于级联卷积神经网络的大场景遥感图像舰船目标快速检测方法[J]. 雷达学报, 2019, 8 (3): 413- 424.
	CHEN H Y , LIU Z Y , GUO W W , et al. Fast detection of ship targets for large-scale remote sensing image based on a cascade convolutional neural network[J]. Journal of Radars, 2019, 8 (3): 413- 424.
24	仲伟峰, 郭峰, 向世明, 等. 旋转矩形区域的遥感图像舰船目标检测模型[J]. 计算机辅助设计与图形学学报, 2019, 31 (11): 1935- 1945.
	ZHONG W F , GUO F , XIANG S M , et al. Ship detection in remote sensing based with rotated rectangular region[J]. Journal of Computer-Aided Design & Computer Graphics, 2019, 31 (11): 1935- 1945.
25	ZHAO W C, NUNO V. Cascade R-CNN: delving into high quality object detection[C]//Proc.of the IEEE Conference Compute Vision Pattern Recognition, 2018: 6154-6162.
26	ALEXANDER N, LUC V G. Efficient non-maximum suppression[C]//Proc.of the International Conference on Pattern Recog-nition, 2006: 850-855.
27	HE K M, ZHANG X Y, REN S Q, et al. Deep residual learn-ing for image recognition[C]//Proc.of the IEEE Conference Compute Vision Pattern Recognition, 2016: 770-778.
28	CANNY J . A computational approach to edge detection[J]. IEEE Trans.on Pattern Analysis and Machine Intelligence, 1986, 8 (6): 679- 698.
29	RICHARD O D , PETER E H . Use of the Hough transform to detect lines and curves in pictures[J]. Communications of the ACM, 1972, 15 (1): 11- 15.
30	BALLARD D H . Generalizing the Hough transform to detect arbitrary shapes[J]. Pattern Recognition, 1981, 13 (2): 111- 122. doi: 10.1016/0031-3203(81)90009-1

类型	1	2	3	4	5	6	7
LWR	4.24	5.10	8.32	8.24	8.45	12.07	5.10
数量	26	65	178	35	205	123	316

目标类型	方法
目标类型	faster RCNN OBB	faster RCNN C-H	Cascade RCNN C-H
第1类	41.18	73.39	88.30
第2类	33.01	74.67	80.89
第3类	30.87	63.15	65.66
第4类	22.31	55.16	60.85
第5类	37.75	49.99	64.30
第6类	73.60	28.94	38.18
第7类	13.66	46.10	44.36
mAP	36.5	55.91	63.22

[1]	吴志鹏, 张平, 李震, 黄磊, 刘畅, 高硕. 基于轻小型无人机雷达的植被高度反演方法[J]. 系统工程与电子技术, 2022, 44(12): 3667-3675.
[2]	李彦君, 刘佳, 徐秋锋. 基于切比雪夫拟合的BP自聚焦算法[J]. 系统工程与电子技术, 2022, 44(10): 3020-3028.
[3]	王彩云, 吴钇达, 王佳宁, 马璐, 赵焕玥. 基于改进的CNN和数据增强的SAR目标识别[J]. 系统工程与电子技术, 2022, 44(8): 2483-2487.
[4]	刘斌, 范翔宇, 张自伟, 张烨. 基于自适应尺度变换-双稳态随机共振模型的BPSK信号检测算法[J]. 系统工程与电子技术, 2022, 44(7): 2084-2095.
[5]	廖金玲, 廖桂生, 许京伟, 兰岚. 基于EPC-MIMO编码设计的解距离模糊性能分析[J]. 系统工程与电子技术, 2022, 44(7): 2166-2174.
[6]	孙晶明, 虞盛康, 孙俊. 基于元学习的雷达小样本目标识别方法及改进[J]. 系统工程与电子技术, 2022, 44(6): 1839-1845.
[7]	孙晶明, 虞盛康, 孙俊. 基于深度学习的HRRP识别姿态敏感性分析[J]. 系统工程与电子技术, 2022, 44(3): 802-807.
[8]	吴涛, 王伦文, 朱敬成. 基于迁移学习和注意力机制的伪装图像分割[J]. 系统工程与电子技术, 2022, 44(2): 376-384.
[9]	方澄, 李慧娟, 路稳, 宋玉蒙, 杨磊. 基于形态学自适应分块的高分辨SAR多特征增强算法[J]. 系统工程与电子技术, 2022, 44(2): 470-479.
[10]	杨兴家, 段克清, 李想, 祁炜. 无人机集群协同探测距离解模糊方法[J]. 系统工程与电子技术, 2022, 44(2): 480-489.
[11]	阮锋, 郭亮, 李亚超, 许然. 基于关联计算的相控阵雷达前视成像技术[J]. 系统工程与电子技术, 2021, 43(9): 2457-2462.
[12]	方文, 全英汇, 沙明辉, 刘智星, 高霞, 邢孟道. 捷变频联合波形熵的密集假目标干扰抑制算法[J]. 系统工程与电子技术, 2021, 43(6): 1506-1514.
[13]	薛东方, 朱晓秀, 胡文华, 郭宝锋, 曾慧燕. 基于加权l₁范数优化的双基地ISAR稀疏成像算法[J]. 系统工程与电子技术, 2021, 43(4): 944-953.
[14]	杨磊, 李慧娟, 黄博, 刘伟, 李埔丞. 双层稀疏组Lasso高分辨SAR结构特征增强成像[J]. 系统工程与电子技术, 2021, 43(2): 351-362.
[15]	董淑仙, 全英汇, 陈侠达, 高霞, 李亚超, 邢孟道. 基于捷变频联合数学形态学的干扰抑制算法[J]. 系统工程与电子技术, 2020, 42(7): 1491-1498.

基于级联卷积神经网络的港口多方向舰船检测与分类

Oriented inshore ship detection and classification based on cascade RCNN

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献 30

相关文章 15

编辑推荐

Metrics

本文评价