基于U2-Net 的广域InSAR 开采沉陷区自动识别
方法研究

金属矿山 ›› 2024, Vol. 53 ›› Issue (4): 177-.

基于U2-Net 的广域InSAR 开采沉陷区自动识别方法研究

吝　涛1　范洪冬1,2　孙　叶1　李向伟3　庄会富1

1. 中国矿业大学环境与测绘学院,江苏徐州 221116;2. 自然资源部国土环境与灾害监测重点实验室,江苏徐州 221116; 3. 山东省煤田地质局物探测量队,山东济南 250104

出版日期:2024-04-15 发布日期:2024-05-19

Study on Automatic Identification Method of Wide-area InSAR Mining Subsidence Area Based on U2-Net

LIN Tao1　FAN Hongdong1,2　SUN Ye1　LI Xiangwei3　ZHUANG Huifu1

1. School of Environment and Spatial Informatics,China University of Mining and Technology,Xuzhou 221116,China; 2. Key Laboratory of Land Environment and Disaster Monitoring,Ministry of Natural Resources,Xuzhou 221116,China; 3. Geophysical Prospecting and Surveying Team of Shandong Bureau of Coal Geology,Jinan 250104,China

Online:2024-04-15 Published:2024-05-19

摘要/Abstract

摘要： 我国井工煤矿量大面广,地下开采隐蔽性强,现有的人工调查、遥感探测、现场实测等方式难以满足大范围开采沉陷区自动识别实现要求,不利于实现高效监管、动态监测。为此,提出了一种基于U2-Net 的广域合成孔径雷达干涉测量(Interferometric Synthetic Aperture Radar,InSAR)开采沉陷区自动识别方法,该方法通过各种形变梯度和噪声水平的模拟数据集训练卷积神经网络(Convolutional Neural Network,CNN),使其能够实现由差分干涉图一步输出包含开采沉陷位置信息的二值矩阵。试验表明:U2-Net 的平均像素准确率(Mean Pixel Accuracy,MPA)和平均交并比 (Mean Intersection Over Union,MIoU)分别达到了0. 916 3、0. 911 9,均高于试验中的其他2 种模型,能够更好地抑制噪声,突出形变信号。在覆盖神东矿区不同时间间隔的InSAR 干涉图上,U2-Net 自动识别了覆盖面积超过54 600 km2 的干涉图,检测出了多处边界信息清晰平滑的沉陷区,识别的平均准确率达到92. 45%。结果表明:对比其他网络, U2-Net 通过2 级嵌套的“U”形结构能够以较小的计算量融合多尺度和多层次特征,在噪声抑制和形变区域识别方面具有显著优势。由此可见,联合深度学习可服务于精细化开采沉陷区详细调查,促进InSAR 技术的应用,为广域开采沉陷区智能识别提供了一种新的技术方法。

关键词: 开采沉陷　深度学习　InSAR　U2-Net　语义分割

Abstract: China′s underground coal mines are large and wide,and underground mining is highly concealed. The existing methods of manual investigation,remote sensing detection and field measurement are difficult to meet the requirements of automatic identification of large-scale mining subsidence areas,which is not conducive to the efficient supervision and dynamic monitoring. Therefore,this paper proposes an automatic identification method of mining subsidence area based on U2-Net wide-area synthetic aperture radar interferometry (InSAR). This method trains convolutional neural network (CNN)through simulation data sets of various deformation gradients and noise levels,so that it can output a binary matrix containing mining subsidence location information in one step from the differential interferogram. The test results show that the mean pixel accuracy (MPA) and mean intersection over union (MIoU) of U2-Net reach 0. 916 3 and 0. 911 9,respectively,which are higher than the other two models in the experiment. It can better suppress noise and highlight deformation signals. On the InSAR interferograms covering the Shendong mining area at different time intervals,U2-Net automatically identified interferograms covering an area of more than 54 600 km2,and detected multiple subsidence areas with clear and smooth boundary information. The average accuracy of recognition reached 92. 45%. The results show that compared with other networks,U2-Net can fuse multi-scale and multi-level features with less computation through a two-stage nested U-shaped structure,which has significant advantages in noise suppression and deformation region recognition. It can be indicated that joint deep learning can serve the detailed investi-gation of refined mining subsidence areas,promote the application of InSAR technique,and provide a new technical method for intelligent identification of wide-area mining subsidence areas.

Key words: mining subsidence,deep learning,InSAR,U2-Net,semantic segmentation

吝　涛, 范洪冬, 孙　叶, 李向伟, 庄会富. 基于U2-Net 的广域InSAR 开采沉陷区自动识别方法研究[J]. 金属矿山, 2024, 53(4): 177-.

LIN Tao, FAN Hongdong, SUN Ye, LI Xiangwei, ZHUANG Huifu. Study on Automatic Identification Method of Wide-area InSAR Mining Subsidence Area Based on U2-Net[J]. Metal Mine, 2024, 53(4): 177-.

扫码分享