Please wait a minute...
 
国土资源遥感  2020, Vol. 32 Issue (4): 68-73    DOI: 10.6046/gtzyyg.2020.04.10
  技术方法 本期目录 | 过刊浏览 | 高级检索 |
基于改进RetinaNet的冷却塔目标检测
卫虹宇(), 赵银娣(), 董霁红
中国矿业大学环境与测绘学院,徐州 221116
Cooling tower detection based on the improved RetinaNet
WEI Hongyu(), ZHAO Yindi(), DONG Jihong
School of Environment Science and Spatial Informatics, China University of Mining and Technology, Xuzhou 221116, China
全文: PDF(4521 KB)   HTML  
输出: BibTeX | EndNote (RIS)      
摘要 

冷却塔排放容易造成大气污染,利用高分辨率遥感影像对冷却塔进行检测,可以为废气排放治理提供决策数据。针对传统算法在高分辨率遥感影像目标检测中检测精度不高、检测速度慢等问题,采用无采样机制改进RetinaNet目标检测框架从而提取冷却塔。首先,将数据集标注为工作中的冷却塔和非工作中的冷却塔; 然后,根据数据集中目标类别数、训练中正样本的比例等特点对分类子网络最后一层的偏置项进行初始化并确定类别自适应阈值,此外,通过回归损失来设置分类损失的调整比例以避免损失函数被众多负样本所支配; 最后,采用ResNet50提取图像特征,利用特征金字塔网络(feature pyramid networks,FPN)模块生成多尺度卷积特征金字塔,对每层特征进行检测框回归以及类别置信度计算。实验结果表明: 对于高分辨率遥感影像冷却塔目标检测,该算法相比原始RetinaNet模型在保证检测速度的同时提高了检测精度,证明算法的有效性。

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
卫虹宇
赵银娣
董霁红
关键词 卷积神经网络目标检测高分辨率遥感影像无采样机制    
Abstract

Cooling tower emissions pollute the atmosphere. Using high-resolution remote sensing images to detect cooling towers can provide decision-making data for the treatment of exhaust emissions. Aiming at the problems such as low detection accuracy and slow detection speed of traditional algorithms in high-resolution remote sensing image object detection, the authors improved the RetinaNet by adopting a sampling-free mechanism to detect the cooling towers. First, Images in dataset were labeled as working cooling towers and resting cooling towers. Then, based on the number of object categories in the dataset and the proportion of positive samples in training, the bias term of the last layer in the classification subnetwork and class-adaptive threshold were determined. In addition, the regression loss was used to determine the adjustment ratio of the classification loss to avoid loss functions to be dominated by numerous background examples. Finally, ResNet50 was used to extract image features, and the FPN module was used to generate a multi-scale convolution feature pyramid. Detection boxes regression and category confidence calculations were performed for each layer of features. The results show that, for cooling tower detection on high-resolution remote sensing images, the proposed algorithm can improve the detection accuracy while ensuring the detection speed compared with RetinaNet, which proves the effectiveness of the proposed algorithm.

Key wordsconvolutional neural network    object detection    high-resolution remote sensing image    sampling-free mechanism
收稿日期: 2020-01-15      出版日期: 2020-12-23
:  TP751.1  
基金资助:国家重点研发计划项目“大型煤电基地土地整治关键技术”(2016YFC0501105);中央高校基本科研业务费专项资金项目“点云与序列影像融合的实景三维模型构建”(2015XKMS050);自然资源部退化及未利用土地整治工程重点实验室开放基金课题“基于深度学习的城市高分遥感变化检测研究”(SXDJ2019-4)
通讯作者: 赵银娣
作者简介: 卫虹宇(1996-),女,硕士研究生,主要研究方向为遥感目标检测。Email:925074644@qq.com
引用本文:   
卫虹宇, 赵银娣, 董霁红. 基于改进RetinaNet的冷却塔目标检测[J]. 国土资源遥感, 2020, 32(4): 68-73.
WEI Hongyu, ZHAO Yindi, DONG Jihong. Cooling tower detection based on the improved RetinaNet. Remote Sensing for Land & Resources, 2020, 32(4): 68-73.
链接本文:  
https://www.gtzyyg.com/CN/10.6046/gtzyyg.2020.04.10      或      https://www.gtzyyg.com/CN/Y2020/V32/I4/68
Fig.1  RetinaNet 框架
Fig.2  数据集示例
方法 AP/% mAP/% 速度/
(s·张-1)
CTWW CTWR
BoW 26.51 21.12 23.82 22.402
HOG 58.51 50.46 54.49 17.180
Faster R-CNN 97.74 90.48 94.11 0.109
SSD 90.73 89.66 90.19 0.046
RetinaNet 97.85 92.77 95.31 0.054
本文算法 98.69 93.55 96.12 0.049
Tab.1  冷却塔目标检测结果对比
Fig.3  冷却塔测试集上的目标检测结果
Fig.4-1  遥感影像冷却塔目标检测结果
Fig.4-2  遥感影像冷却塔目标检测结果
[1] 白永清, 祁海霞, 刘琳, 等. 武汉大气能见度与PM2.5浓度及相对湿度关系的非线性分析及能见度预报[J]. 气象学报, 2016,74(2):189-199.
Bai Y Q, Qi H X, Liu L, et al. Study on the nonlinear relationship among the visibility,PM2.5 concentration and relative humidity in Wuhan and the visibility prediction[J]. Acta Meteorologica Sinica, 2016,74(2):189-199.
[2] 孙俊英, 张璐, 沈小静, 等. 大气气溶胶散射吸湿增长特性研究进展[J]. 气象学报, 2016,74(5):672-682.
Sun J Y, Zhang L, Shen X J, et al. A review of the effects of relative humidity on aerosol scattering properties[J]. Acta Meteorologica Sinica, 2016,74(5):672-682.
[3] 徐刚, 岳继光, 董延超, 等. 深度卷积网络卫星图像水泥厂目标检测[J]. 中国图象图形学报, 2019,24(4):550-561.
Xu G, Yue J G, Dong Y C, et al. Cement plant detection on satellite images using deep convolution network[J]. Journal of Image and Graphics, 2019,24(4):550-561.
[4] Krizhevsky A, Sutskever I, Hinton G E. ImageNet classification with deep convolutional neural networks[C]// 26th Annual Conference on Neural Information Processing Systems, 2012:1097-1105.
[5] 谢奇芳, 姚国清, 张猛. 基于Faster R-CNN的高分辨率图像目标检测技术[J]. 国土资源遥感, 2019,31(2):38-43.doi: 10.6046/gtzyyg.2019.02.06.
Xie Q F, Yao G Q, Zhang M. Research on high resolution image object detection technology based on Faster R-CNN[J]. Remote Sensing for Land and Resources, 2019,31(2):38-43.doi: 10.6046/gtzyyg.2019.02.06.
[6] Girshick R, Donahue J, Darrell T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]// 2014 IEEE Conference on Computer Vision and Pattern Recognition, 2014:580-587.
[7] 回天, 哈力旦·阿布都热依木, 杜晗. 结合Faster R-CNN的多类型火焰检测[J]. 中国图象图形学报, 2019,24(1):73-83.
Hui T, Halidan A, Du H. Multi-type flame detection combined with Faster R-CNN[J]. Journal of Image and Graphics, 2019,24(1):73-83.
[8] He K, Zhang X, Ren S, et al. Spatial pyramid pooling in deep convo-lutional networks for visual recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015,37(9):1904-1916.
doi: 10.1109/TPAMI.2015.2389824 pmid: 26353135
[9] Girshick R. Fast R-CNN[C]// 2015 IEEE International Conference on Computer Vision (ICCV), 2015:1440-1448.
[10] 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 and Machine Intelligence, 2017,39(6):1137-1149.
doi: 10.1109/TPAMI.2016.2577031 pmid: 27295650
[11] Redmon J, Divvala S, Girshick R, et al. You only look once:Unified,real-time object detection[C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016:779-788.
[12] Liu W, Anguelov D, Erhan D, et al. SSD:Single shot multibox detector[C]// European Conference on Computer Vision.Berlin,Springer, 2016:21-37.
[13] Lin T Y, Goyal P, Girshick R, et al. Focal loss for dense object detection[C]// Proceedings of the IEEE International Conference on Computer Vision, 2017:2980-2988.
[14] Chen J, Liu D, Xu T, et al. Is sampling heuristics necessary in training deep object detectors?[EB/OL] (2019- 11- 16). http://arxiv.org/abs/1909.04868.
[15] He K, Zhang X, Ren S, et al. Deep residual learning for image reco-gnition[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. IEEE, 2016:770-778.
[16] 刘革, 郑叶龙, 赵美蓉. 基于RetinaNet改进的车辆信息检测[J]. 计算机应用, 2020,40(3):854-858.
Liu G, Zheng Y L, Zhao M R. Vehicle information detection based on RetinaNet[J]. Journal of Computer Applications, 2020,40(3):854-858.
[17] Lowe D G. Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision, 2004,60(2):91-110.
[18] Dalal N, Triggs B. Histograms of oriented gradients for human detection[C]// IEEE Computer Society Conference on Computer Vision and Pattern. IEEE, 2005.
[19] Zhang X, Xiang J, Zhang Y. Space object detection in video satellite images using motion information[J]. International Journal of Aerospace Engineering, 2017,1024529:1-9.
[1] 薛白, 王懿哲, 刘书含, 岳明宇, 王艺颖, 赵世湖. 基于孪生注意力网络的高分辨率遥感影像变化检测[J]. 自然资源遥感, 2022, 34(1): 61-66.
[2] 于新莉, 宋妍, 杨淼, 黄磊, 张艳杰. 结合空间约束的卷积神经网络多模型多尺度船企场景识别[J]. 自然资源遥感, 2021, 33(4): 72-81.
[3] 陈静, 陈静波, 孟瑜, 邓毓弸, 节永师, 张懿. 尺度和密度约束下基于YOLOv3的风电塔架遥感检测方法[J]. 自然资源遥感, 2021, 33(3): 54-62.
[4] 刘万军, 高健康, 曲海成, 姜文涛. 多尺度特征增强的遥感图像舰船目标检测[J]. 自然资源遥感, 2021, 33(3): 97-106.
[5] 谭熊, 王晶磊, 孙一帆. 基于特征的无人机载视频运动目标快速检测方法[J]. 国土资源遥感, 2021, 33(2): 27-32.
[6] 安健健, 孟庆岩, 胡蝶, 胡新礼, 杨健, 杨天梁. 基于Faster R-CNN的火电厂冷却塔检测及工作状态判定[J]. 国土资源遥感, 2021, 33(2): 93-99.
[7] 仇一帆, 柴登峰. 无人工标注数据的Landsat影像云检测深度学习方法[J]. 国土资源遥感, 2021, 33(1): 102-107.
[8] 董天成, 杨肖, 李卉, 张志, 齐睿. 基于Faster R-CNN和MorphACWE模型的SAR图像高原湖泊提取[J]. 国土资源遥感, 2021, 33(1): 129-137.
[9] 胡苏李扬, 李辉, 顾延生, 黄咸雨, 张志麒, 汪迎春. 基于高分辨率遥感影像的神农架大九湖湿地土地利用类型变化及其驱动力分析——来自长时间尺度多源遥感信息的约束[J]. 国土资源遥感, 2021, 33(1): 221-230.
[10] 刘钊, 赵桐, 廖斐凡, 李帅, 李海洋. 基于语义分割网络的高分遥感影像城市建成区提取方法研究与对比分析[J]. 国土资源遥感, 2021, 33(1): 45-53.
[11] 刘钊, 廖斐凡, 赵桐. 基于PSPNet的遥感影像城市建成区提取及其优化方法[J]. 国土资源遥感, 2020, 32(4): 84-89.
[12] 蔡之灵, 翁谦, 叶少珍, 简彩仁. 基于Inception-V3模型的高分遥感影像场景分类[J]. 国土资源遥感, 2020, 32(3): 80-89.
[13] 吴同, 彭玲, 胡媛. 基于SU-RetinaNet的高分辨率遥感影像非正规垃圾堆检测[J]. 国土资源遥感, 2020, 32(3): 90-97.
[14] 李宇, 肖春姣, 张洪群, 李湘眷, 陈俊. 深度卷积融合条件随机场的遥感图像语义分割[J]. 国土资源遥感, 2020, 32(3): 15-22.
[15] 康晋洁, 戚浩平, 杨清华, 陈华. 道路通行障碍物遥感检测与通过性评价[J]. 国土资源遥感, 2020, 32(2): 94-102.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
京ICP备05055290号-2
版权所有 © 2015 《自然资源遥感》编辑部
地址:北京学院路31号中国国土资源航空物探遥感中心 邮编:100083
电话:010-62060291/62060292 E-mail:zrzyyg@163.com
本系统由北京玛格泰克科技发展有限公司设计开发