基于DETR的高分辨率遥感影像滑坡体识别与检测

doi:10.13474/j.cnki.11-2246.2023.0129

测绘通报 ›› 2023, Vol. 0 ›› Issue (5): 16-20.doi: 10.13474/j.cnki.11-2246.2023.0129

基于DETR的高分辨率遥感影像滑坡体识别与检测

杜宇峰¹, 黄亮^1,2, 赵子龙³, 李国柱^1,3

1. 昆明理工大学国土资源工程学院, 云南昆明 650093;
2. 云南省高校高原山区空间信息测绘技术应用工程研究中心, 云南昆明 650093;
3. 云南海钜地理信息技术有限公司, 云南昆明 650093

收稿日期:2022-09-21 发布日期:2023-05-31
通讯作者: 黄亮。E-mail:kmhuangliang@163.com
作者简介:杜宇峰(1996-),男,硕士,主要研究方向为遥感影像目标检测。E-mail:1164151228@qq.com
基金资助:
云南省基础研究计划（202201AT070164）；国家自然科学基金（41961039）；云南省基础研究计划（202101AT070102）

Landslide body identification and detection of high-resolution remote sensing image based on DETR

DU Yufeng¹, HUANG Liang^1,2, ZHAO Zilong³, LI Guozhu^1,3

1. Faculty of Land Resource Engineering, Kunming University of Science and Technology, Kunming 650093, China;
2. Surveying and Mapping Geo-Informatics Technology Research Center on Plateau Mountains of Yunnan Higher Education, Kunming 650093, China;
3. Yunnan Haiju Geographic Information Technology Co., Ltd., Kunming 650093, China

Received:2022-09-21 Published:2023-05-31

摘要/Abstract

摘要： 滑坡灾害因其极大的破坏性而引起高度重视，如何快速、高精度地自动检测滑坡体成为主要研究问题。针对滑坡体检测数据不足、精度低、检测滑坡体不完全等问题，本文结合卷积神经网络（CNN）和Transformer的优点，以Transformer为主体，采用DETR网络实现滑坡体的自动检测。首先，对于数据集数据不足的问题，采用离线数据增强的方式实现滑坡体数据增广；然后，采用编码器-解码器结构的DETR网络结构对增广数据集进行多尺度训练和预测；最后，对试验结果进行定量评价。试验结果表明，采用DETR网络对滑坡体检测的平均准确率（AP）达0.997，可准确识别和检测滑坡体。此外，试验结果还验证了数据增强可有效提升DETR网络对滑坡体的检测精度。

关键词: 滑坡, 目标检测, 卷积神经网络, DETR, 注意力机制

Abstract: Landslide disasters have attracted great attention because of their great destructiveness, and how to quickly and accurately detect landslides has become a major problem. Aiming at the problems of insufficient landslide detection dataset, low accuracy, and incomplete detection of landslide body, this paper combines the advantages of convolutional neural networks (CNN) and Transformer, and adopts the DETR network to realize the automatic detection of landslide body with Transformer as the main body. First of all, in order to solve the problem of insufficient data in the data set, the offline data enhancement method is used to achieve landslide data augmentation; Secondly, the DETR network structure using the encoder-decoder structure performs multi-scale training and prediction of the augmented dataset; Finally, the experimental results are quantitatively evaluated. Experimental results show that the average accuracy(AP)of landslide detection is 0.997,which can accurately identify and detect landslide bodies.In addition,the experimental results also verify that data enhancement can effectively improve the detection accuracy of landslide bodies in the DETR network.

Key words: landslide, object detection, convolutional neural network, DETR, attention mechanism

中图分类号:

P237

杜宇峰, 黄亮, 赵子龙, 李国柱. 基于DETR的高分辨率遥感影像滑坡体识别与检测[J]. 测绘通报, 2023, 0(5): 16-20.

DU Yufeng, HUANG Liang, ZHAO Zilong, LI Guozhu. Landslide body identification and detection of high-resolution remote sensing image based on DETR[J]. Bulletin of Surveying and Mapping, 2023, 0(5): 16-20.

参考文献

[1] 张毅, 谭龙, 陈冠, 等. 基于面向对象分类法的高分辨率遥感滑坡信息提取[J]. 兰州大学学报(自然科学版), 2014, 50(5):745-750.
[2] 彭令, 徐素宁, 梅军军, 等. 地震滑坡高分辨率遥感影像识别[J]. 遥感学报, 2017, 21(4):509-518.
[3] 赵福军, 樊雅婧.深度卷积神经网络的遥感影像滑坡灾害识别[J].黑龙江科技大学学报, 2020, 30(5):556-561.
[4] CHENG Libo, LI Jia, DUAN Ping, et al. A small attentional YOLO model for landslide detection from satellite remote sensing images[J].Landslides, 2021, 18(8):2751-2765.
[5] JI Shunping, YU Dawen, SHEN Chaoyong, et al. Landslide detection from an open satellite imagery and digital elevation model dataset using attention boosted convolutional neural networks[J].Landslides, 2020, 17(6):1337-1352.
[6] CARION N, MASSA F, SYNNAEVE G, et al. End-to-end object detection with transformers[M]//Computer Vision-ECCV 2020. Cham:Springer International Publishing, 2020:213-229.
[7] 师芸, 石龙龙, 牛敏杰, 等.基于单阶段实例分割网络的黄土滑坡多任务自动识别[J].测绘通报, 2022(4):26-31.
[8] HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Deep residual learning for image recognition[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas:IEEE, 2016:770-778.
[9] 钱茹雪. 基于深度神经网络的花朵图像识别及分类[J]. 信息技术与信息化, 2022(3):210-213.
[10] HASSANIN M, ANWAR S, RADWAN I, et al. Visual attention methods in deep learning:an In-depth survey"[EB/OL].[2022-08-21].https://arxiv.org/abs/2204.07756.
[11] ZHOU Xingyi, WANG Dequan, KRÄHENBVHL P. Objects as points"[EB/OL].[2022-08-21]. https://arxiv.org/abs/1904.07850.
[12] 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 and Machine Intelligence, 2017, 39(6):1137-1149.
[13] TAN Mingxing, PANG Ruoming, LE Q V. EfficientDet:scalable and efficient object detection[C]//Proceedings of 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle:IEEE, 2020:10778-10787.
[14] REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once:unified, real-time object detection[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas:IEEE, 2016:779-788.
[15] BOCHKOVSKIY A, WANG C Y, LIAO H Y M. YOLOv4:optimal speed and accuracy of object detection"[EB/OL].[2022-08-21]. https://arxiv.org/abs/2004.10934.

基于DETR的高分辨率遥感影像滑坡体识别与检测

Landslide body identification and detection of high-resolution remote sensing image based on DETR

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	孟永东, 袁昌纬, 田斌, 蔡征龙, 张伟杰. 滑坡表面位移的无人机航测点云比对监测方法[J]. 测绘通报, 2023, 0(5): 1-8.
[2]	王之栋, 唐伟, 马志刚, 李雨宸, 杨本勇, 李维庆, 李永鑫. 九寨沟地区高位滑坡隐患InSAR-LiDAR早期识别[J]. 测绘通报, 2023, 0(5): 9-15.
[3]	胡祥祥, 柯福阳, 张志山, 姚永顺, 宋宝, 明璐璐, 尹继鑫, 张海欢. 顾及多动态环境因子的滑坡演化规律研究——以西宁市9大滑坡区为例[J]. 测绘通报, 2023, 0(5): 21-26,43.
[4]	曹小燕, 满新耀, 汪继平, 麦荣章, 郭云开. 基于鸟群优化BP神经网络的滑坡处治后变形预测[J]. 测绘通报, 2023, 0(5): 27-31.
[5]	齐建伟, 王伟峰, 张乐, 王光彦. 基于改进DeepLabV3+算法的遥感影像建筑物变化检测[J]. 测绘通报, 2023, 0(4): 145-149.
[6]	何亚东, 徐许雄, 李林, 孙铭, 黄逢. 联合北斗与地基InSAR多源数据分析四川茂县特大滑坡[J]. 测绘通报, 2023, 0(4): 159-162.
[7]	雷洋洋, 李礼, 孙飞, 姚剑. 一种基于域适应的动态商品视觉识别方法[J]. 测绘通报, 2023, 0(3): 10-15.
[8]	丁乐乐, 刘艳飞, 孟凡效, 张涛, 王震, 潘宇明. 深度学习倾斜三维模型车辆去除算法[J]. 测绘通报, 2023, 0(3): 94-98.
[9]	李子彧, 周维勋, 耿万轩. 联合类别筛选与重排序的交叉视角图像地理定位[J]. 测绘通报, 2023, 0(2): 40-45.
[10]	张永彬, 刘雅辉, 刘明月, 满卫东, 宋唐雷, 李春雨. 基于Relief F与卷积神经网络的湿地植物群落精细分类[J]. 测绘通报, 2023, 0(2): 58-64.
[11]	张领旗, 谢酬, 陈蜜, 田帮森, 杨知, 朱玉. 时序InSAR技术在三峡库区特高压输电通道滑坡隐患识别的应用[J]. 测绘通报, 2023, 0(2): 97-103.
[12]	施仲添, 沈正伟, 杨四海. 基于前景感知的遥感影像建筑物提取方法[J]. 测绘通报, 2023, 0(2): 134-138.
[13]	方留杨, 刘天逸, 赵鑫, 吴昊, 贾志文, 曾珍. 基于增强生成对抗网络的滑坡影像集超分辨率重建[J]. 测绘通报, 2023, 0(1): 45-51.
[14]	宋伟东, 毕春杨, 赵奉书. 改进Mask R-CNN公路病害检测算法[J]. 测绘通报, 2023, 0(1): 58-64.
[15]	陈巧, 袁飞云, 付霞, 张文居, 董秀军. 无人机摄影测量技术在阿娘寨滑坡应急调查中的应用[J]. 测绘通报, 2023, 0(1): 77-83,119.