Please wait a minute...
img

官方微信

遥感技术与应用  2019, Vol. 34 Issue (4): 727-735    DOI: 10.11873/j.issn.1004-0323.2019.4.0727
CNN 专栏     
基于深度卷积神经网络的油罐目标检测研究
王颖洁(),张荞(),张艳梅,蒙印,郭文
国家测绘地理信息局第三航测遥感院,四川 成都 610100
Oil Tank Detection from Remote Sensing Images based on Deep Convolutional Neural Network
Yingjie Wang(),Qiao Zhang(),Yanmei Zhang,Yin Meng,Wen Guo
The Third Remote Sensing Geomatics Institute of National Administration of Surveying, Mapping and Geoinformation, Chengdu 610100, China
 全文: PDF(16251 KB)   HTML
摘要:

油罐是用于储存油品的工业设施,常用在炼油厂等工业园中,通过卫星或航空遥感图像实现油罐目标的快速检测,可以实现对侵占生态保护红线的疑似工业园区的快速查找,为自然资源监管和生态环境保护提供科学技术支持。探讨了基于深度卷积神经网络在高分辨率遥感影像目标检测中的有效性,基于深度学习目标检测算法中具有代表性的Faster R-CNN(Convolutional Neural Network)和R-FCN(Region-based Fully Convolutional Network)框架,通过对ZF、VGG16、ResNet-50 3种网络模型进行训练和测试,实现了遥感影像上油罐目标的快速检测;通过修改锚点尺度和数量,丰富了候选框类型和数量,提升了油罐的目标检测精度,最优召回率接近80%。研究表明:深度卷积神经网络能够实现对高分辨率遥感影像中油罐目标的快速检测,为深度学习技术在遥感小目标的快速检测提供了实例和新的思路。

关键词: 深度学习卷积神经网络遥感目标检测油罐    
Abstract:

Oil tanks are industrial facilities for storing oil products, which are commonly used in industrial parks such as oil refineries. The rapid detection of oil tank target through satellite or aerial remote sensing images can quickly find suspected industrial parks, providing scientific and technical support for natural resource regulation and ecological environment protection. This paper discussed the possibility of object detection with high-resolution remote sensing images based on deep convolutional neural network. The state-of-the-art algorithms of Faster R-CNN (Convolutional Neural Network) and R-FCN (Region-based Fully Convolutional Network) and three network models were applied for oil tank detection from high-resolution remote sensing images. To promote the detection accuracy and execution efficiency for the oil tank target, an improved approach by increasing the scales of the anchor was proposed. The optimum recall reached about 80%. The results confirm that deep learning network approach can rapid detect oil tank from high-resolution remote sensing image. This provide an example and new idea for rapid detection small target from remote sensing image by deep learning technology.

Key words: Deep learning    Convolutional neural network(CNN)    Remote sensing object detection    Oil tank
收稿日期: 2018-08-03 出版日期: 2019-10-16
ZTFLH:  TP75  
基金资助: 四川省遥感大数据应用工程技术研究中心开放基金课题(2018KFJJ01);自然资源部基础测绘科技项目(2018KJ0304);成都市科技项目(2015-CX00-00011-ZF)
通讯作者: 张荞     E-mail: 641055472@qq.com;scrs_qiaozh@163.com
作者简介: 王颖洁(1990-),女,重庆人,硕士,助理工程师,主要从事遥感数据处理应用研究。E?mail:641055472@qq.com
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
王颖洁
张荞
张艳梅
蒙印
郭文

引用本文:

王颖洁,张荞,张艳梅,蒙印,郭文. 基于深度卷积神经网络的油罐目标检测研究[J]. 遥感技术与应用, 2019, 34(4): 727-735.

Yingjie Wang,Qiao Zhang,Yanmei Zhang,Yin Meng,Wen Guo. Oil Tank Detection from Remote Sensing Images based on Deep Convolutional Neural Network. Remote Sensing Technology and Application, 2019, 34(4): 727-735.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2019.4.0727        http://www.rsta.ac.cn/CN/Y2019/V34/I4/727

图1  Faster R-CNN算法网络结构
图2  候选区域生成网络(RPN)
图3  分类回归网络结构
图4  R-FCN结构图
图5  标注的油罐样本
图6  ZF模型各个卷积层提取到的特征
网络 准确率 召回率 时间/h
端到端

交替

优化

端到端

交替

优化

端到端 交替优化
ZF 0.619 0.610 0.622 0.618 3 5
VGG16 0.618 0.612 0.664 0.669 7.5 14
ResNet-50 0.626 0.600 0.695 0.693 4 13
表1  两种训练方式油罐检测结果
网络 准确率(mAp) 召回率(recall) 时间/h
修改前 修改后 修改前 修改后 修改前 修改后
ZF 0.619 0.712 0.622 0.728 3 3
VGG16 0.618 0.714 0.664 0.764 7.2 7.5
ResNet-50 0.626 0.716 0.695 0.795 4 4
表2  不同模型修改锚点大小对油罐检测的影响
图7  不同模型油罐识别效果比较
1 Zhang Weisheng , Wang Chao , Zhang Hong , et al . An Automatic Oil Tank Detection Algorithm based on Remote Sensing Image [J]. Journal of Astronautics, 2006, 27(6): 1298-1301.
1 张维胜,王超,张红,等 . 基于遥感影像的油罐自动检测算法[J].宇航学报, 2006, 27(6): 1298-1301.
2 Yin Liang , Gao Kun , Bai Tingzhu . Research on Oilcan Segmentation in Remote Sensing Image based on Improved Otsu Algorithm [J]. Optical Technique, 2012, 38(2): 197-201.
2 殷亮,高昆,白廷柱 . 基于改进Otsu法的遥感图像油罐目标分割研究[J].光学技术, 2012, 38 (2): 197-201.
3 Wu Xiaodong , Feng Wufa , Feng Qianqian , et al . Oil Tank Extraction from Remote Sensing Images based on Visual Attention Mechanism and Hough Transform [J]. Journal of Information Engineering University, 2015, 16(4): 503-506.
3 吴晓东,冯伍法,冯倩倩,等 . 基于视觉注意机制和Hough变换融合的遥感影像油罐提取[J].信息工程大学学报, 2015, 16 (4): 503-506.
4 Song Huansheng , Zhang Xiangqing , Zheng Baofeng , et al .Vehicle Detection based on Deep Learning in Complex scene [J]. Application Research Computers, 2018, 35 (4): 1270-1273.
4 宋焕生,张向清,郑宝峰,等 . 基于深度学习方法的复杂场景下车辆目标检测[J]. 计算机应用研究, 2018, 35 (4): 1270-1273.
5 Hinton G E , Osindero S , Teh Y W . A Fast Learning Algorithm for Deep Belief Nets [J]. Neural Computation, 2006, 18: 1527-1554.
6 Krizhevsky A , Sutskever I , Hinton G E . ImageNet Classification with Deep Convolutional Neural Networks [C]∥International Conference on Neural Information Processing Systems, 2012, 25(2): 1097-1105.
7 He K M , Zhang X Y , Ren S Q , et al . Deep Residual Learning for Image Recognition [C]∥IEEE Conference on Computer Vision and Pattern Recognition. IEEE Computer Society, 2006: 770-778.
8 Girshick R , Donahue J , Darrell T , et al . Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation [C]∥Computer Vision and Pattern Recognition, 2014: 580-587.
9 Ren Shaoqing , He Kaiming , 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.
10 Redmon J , Divvala S , Girshick R , et al . You only Look Once: Unified, Real-time Object Detection[C]∥IEEE Conference on Computer Vision and Pattern Recognition. IEEE Computer Society, 2016: 779-788.
11 Liu Wei , Anguelov D , Erhan D , et al . SSD: Single Shot MultiBox Detector [M]. Computer Vision-ECCV 2016. Springer International Publishing, 2016: 21-37.
12 Dai Chenka , Li Yi . Aeroplane Detection in Static Aerodrome based on Faster RCNN and Multi-part Model [J]. Journal of Computer Applications, 2017, 37(Sup.2: 85-88.
12 戴陈卡, 李毅 . 基于Faster RCNN以及多部件结合的机场场面静态飞机检测[J]. 计算机应用, 2017, 37(增刊2): 85-88.
13 Wang Wanguo , Tian Bing , Liu Yue , et al . Study on the Electrical Devices Detection in UAV Images based on Region Based Convolutional Neural Networks [J]. Journal of Geo-information Science, 2017, 19(2): 256-263.
13 王万国,田兵,刘越,等 . 基于RCNN的无人机巡检图像电力小部件识别研究[J]. 地球信息科学学报, 2017, 19(2): 256-263.
14 Hu Yan , Shan Zili , Gao Feng , et al . Ship Detection based on Faster-RCNN and Multiresolution SAR [J]. Radio Engineering, 2018, 48(2): 96-100.
14 胡炎,单子力,高峰 . 基于Faster-RCNN和多分辨率SAR的海上舰船目标检测[J]. 无线电工程, 2018, 48(2): 96-100.
15 Wu Weiming . Research on Object Detection Algorithm Based on Faster R-CNN [D]. Guangzhou: South China University of Technology, 2017.
15 伍伟明 . 基于Faster R-CNN的目标检测算法的研究 [D]. 广州:华南理工大学,2017.
16 Yosinski J , Clune J , Bengio Y , et al . How Transferable are Features in Deep Neural Networks [C]∥International Conference on Neural Information Processing Systems. MIT Press, 2014: 3320-3328.
17 Russakovsky O , Deng Jia , Su Hao , et al . ImageNet Large Scale Visual Recognition Challenge [J]. International Journal of Computer Vision, 2015, 115(3): 211-252.
18 Zhang Wei . Land Cover Classification with Extracted Deep Features of Deep Convolutional Neural Network [D]. Beijing: Institute of Remote Sensing and Digital Earth Chinese Academy of Sciences, 2017.
18 张伟 . 基于深度卷积神经网络自学习特征的地表覆盖分类研究 [D]. 北京:中国科学院大学,2017.
19 Dai J F , Li Y , He K M , et al . R-FCN: Object Detection Via Region-based Fully Convolutional Networks [M]. Advances in Neural Information Processing Systems, 2016: 379-387.
20 Jiang Sheng , Huang Min , Zhu Qibing , et al . Pedestrain Detection Method based on R-FCN [J]. Computer Engineering and Application, 2018(1):180-183,262.
20 蒋胜,黄敏,朱启兵, 等 . 基于R-FCN的行人检测方法研究[J]. 计算机工程与应用, 2018(1):180-183,262.
21 Xu Yizhi , Yao Xiaojing , Li Xiang , et al . Object Detection in High Resolution Remote Sensing Images based on Fully Convolution Networks, 2018(1):77-82.
21 徐逸之,姚晓婧,李祥,等 . 基于全卷积网络的高分辨遥感影像目标检测[J]. 测绘通报, 2018(1):77-82.
22 Lecun Y , Bengio Y , Hinton G . Deep Learning [J]. Nature, 2015, 521(7553): 436-444.
23 Zeiler M D , Fergus R . Visualizing and Understanding Convolutional Networks [M]. Computer Vision-ECCV 2014. Springer International Publishing, 2014: 818-833.
24 Simonyan K , Zisserman A . Very Deep Convolutional NetWorks for Large-scale Image Recognition [J]. Computer Science, 2014: abs/1409.1556.
25 Dong Yunya , Zhang Qian . A Survey of Depth Semantic Feature Extraction of High-resolution Remote Sensing Images based on CNN [J]. Remote Sensing Technology and Application, 2019, 34(1): 1-11.
25 董蕴雅, 张倩 . 基于CNN的高分遥感影像语义特征提取研究综述[J]. 遥感技术与应用, 2019, 34(1): 1-11.
[1] 林志玮,丁启禄,黄嘉航,涂伟豪,胡典,刘金福. 基于DenseNet的无人机光学图像树种分类研究[J]. 遥感技术与应用, 2019, 34(4): 704-711.
[2] 刘天福,陈学泓,董琪,曹鑫,陈晋. 深度学习在GlobeLand30-2010产品分类精度优化中应用研究[J]. 遥感技术与应用, 2019, 34(4): 685-693.
[3] 徐梦竹, 徐佳, 邓鸿儒, 袁春琦. 基于全极化SAR影像的海岛地物分类[J]. 遥感技术与应用, 2019, 34(3): 647-654.
[4] 李淑敏, 冯权泷, 梁其椿, 张学庆. 基于深度学习的国产高分遥感影像飞机目标自动检测[J]. 遥感技术与应用, 2018, 33(6): 1095-1102.
[5] 田德宇,张耀南,赵国辉,韩立钦. 基于卷积神经网络的遥感沙漠绿地提取方法[J]. 遥感技术与应用, 2018, 33(1): 151-157.
[6] 何海清,庞燕,陈晓勇. 面向遥感影像场景的深度卷积神经网络递归识别模型[J]. 遥感技术与应用, 2017, 32(6): 1078-1082.