基于高分影像和改进YOLOv7模型在输电线路走廊的建筑物识别

doi:10.13474/j.cnki.11-2246.2025.0414

摘要/Abstract

摘要： 输电线廊道内的灾害预警与安全评估是智能电网建设工作的重点之一,因此掌握输电线廊道区域内聚落的所在位置及其分布情况等信息对于做好山区防灾减灾工作非常重要。近年来,随着目标检测技术的不断发展,所应用到的领域也越来越广泛,而遥感目标检测作为其中的一个应用场景,因为其覆盖范围广、涵盖目标多的特点,被广泛应用于建筑物信息提取。现有的深度学习模型在建筑物识别分割上的识别精度和检测速度都存在局限。针对此类问题,本文以GF-2影像作为数据基础,首先对山区建筑物标记,建立样本数据集并按照9∶1的比例划分为训练集和验证集;然后对标准版YOLOv7网络进行改进,在颈部部分添加GAM-CBAM合成的双重注意力模块,减少建筑物的特征丢失,从而提高网络的检测能力。结果表明,改进后的YOLOv7网络对山区建筑物的分割识别平均精度达到了88.74%,在精确度和召回率上也高于其他深度学习模型。因此,该方法可以快速高效地获取山区聚落信息,并为山区防灾减灾过程中应急预案的制定提供数据支撑。

关键词: 建筑物识别, 机器学习, YOLOv7, 输电线路走廊, 目标检测

Abstract: Early disaster warning and safety assessment in transmission line corridors are among the priorities of smart grid construction. Therefore, it is very important to grasp the location and distribution of settlements in the transmission line corridor area to do a good job of disaster prevention and mitigation in mountainous areas. In recent years, with the continuous development of target detection technology, the fields it is applied to are becoming more and more extensive, and remote sensing target detection, as one of the application scenarios, is widely used in building information because of its wide coverage and the characteristics of covering many targets. The existing deep learning models have limitations regarding both recognition accuracy and detection speed in building identification and segmentation. Aiming at such problems, this study takes the GF-2 image as the database, labels the buildings in mountainous areas, establishes the sample dataset and divides it into the training set and the test set according to the ratio of 9∶1. Secondly, the standard version of YOLOv7 network is improved by adding a dual-attention module with GAM-CBAM synthesis in the neck part to reduce the feature loss of buildings, which improves the detection ability of the network. The results show that the improved YOLOv7 network achieves an average precision of 88.74% for segmentation and recognition of buildings in mountainous areas, which is also higher than other deep learning models in terms of precision and recall. Therefore, this method can provide data support for rapid and efficient acquisition of mountainous area settlement information, geographic information analysis in the process of disaster prevention and mitigation in mountainous areas, and the development of emergency plans.

Key words: building recognition, machine learning, YOLOv7, transmission line corridor, target detection

中图分类号:

P237

杨国柱, 孙诗睿, 田茂杰, 孙华敏, 胡伟, 李俊磊. 基于高分影像和改进YOLOv7模型在输电线路走廊的建筑物识别[J]. 测绘通报, 2025, 0(4): 82-89.

YANG Guozhu, SUN Shirui, TIAN Maojie, SUN Huamin, HU Wei, LI Junlei. Building recognition in transmission line corridors based on high-resolution images and improved YOLOv7 model[J]. Bulletin of Surveying and Mapping, 2025, 0(4): 82-89.

参考文献

[1] 史培军.中国自然灾害风险地图集[M].北京:科学出版社,2011.
[2] 应急管理部网站.国家防灾减灾救灾委员会办公室应急管理部发布2024年7月全国自然灾害情况[J].中国减灾,2024(16):6-7.
[3] 章可怡,石咏,郭海湘,等.基于卡车-无人机协同的山区自然灾害应急物资调度优化决策研究[J/OL].中国管理科学.[2024-07-30].https://doi.org/10.16381/j.cnki.issn1003-207x.2023.1278.
[4] 苏林雪.汶川地震前后四川省主要地质灾害时空发育规律研究[D].焦作:河南理工大学,2021.
[5] 李成范,孟令奎,刘学锋.基于深度学习的高分遥感图像建筑物识别[J].应用科学学报,2024,42(3):375-387.
[6] 张艳.高分辨率遥感影像建筑物检测及倒塌建筑物识别方法研究[D].南京:南京信息工程大学,2022.
[7] YU Bailang,LIU Hongxing,WU Jianping,et al.Automated derivation of urban building density information using airborne LiDAR data and object-based method[J].Landscape and Urban Planning,2010,98(3/4): 210-219.
[8] GUO Li,CHEHATA N,MALLET C,et al.Relevance of airborne lidar and multispectral image data for urban scene classification using Random Forests[J].ISPRS Journal of Photogrammetry and Remote Sensing,2011,66(1): 56-66.
[9] HUANG Xin,ZHANG Liangpei.An SVM ensemble approach combining spectral,structural,and semantic features for the classification of high-resolution remotely sensed imagery[J].IEEE Transactions on Geoscience and Remote Sensing,2013,51(1): 257-272.
[10] WANG C,SHEN Y,SHEN Y,et al.Building extraction from high-resolution remote sensing images by adaptive morphological attribute profile under object boundary constraint[J].Sensors (Basel),2019,19(17): E3737.
[11] 汪泉.基于卷积神经网络的机载高光谱影像针叶树种分类研究[D].哈尔滨:东北林业大学,2022.
[12] KRIZHEVSKY A,SUTSKEVER I,HINTON G E.ImageNet classification with deep convolutional neural networks[J].Communications of the ACM,2017,60(6): 84-90.
[13] SIMONYAN K,ZISSERMAN A.Very deep convolutional networks for large-scale image recognition[C]//Proceedings of International Conference on Learning Representations.[S.l.]:ICLR,2015.
[14] SZEGEDY C,LIU Wei,JIA Yangqing,et al.Going deeper with convolutions[C]//Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition.[S.l.]:IEEE,2015.
[15] 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.
[16] ALIDOOST F,AREFI H.A CNN-based approach for automatic building detection and recognition of roof types using a single aerial image[J].Journal of Photogrammetry,Remote Sensing and Geoinformation Science,2018,86(5): 235-248.
[17] 刘媛.高分辨率遥感影像建筑物屋顶识别方法研究[D].北京:北京建筑大学,2021.
[18] 左俊皓,赵聪,朱晓龙,等.Faster-RCNN和Level-Set结合的高分遥感影像建筑物提取[J].液晶与显示,2019,34(4):439.
[19] 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.
[20] 蒯宇.基于深度学习的无人机遥感植被识别方法研究[D].合肥:安徽大学,2022.
[21] 奚祥书,夏凯,杨垠晖,等.结合多光谱影像降维与深度学习的城市单木树冠检测[J].遥感学报,2022,26(4):711-721.
[22] 朱伟东,何月顺,陈杰,等.基于改进YOLOv5的海洋生物检测算法[J].计算机与数字工程,2022,50(8):1631-1636.
[23] 王梦园.基于深度学习的遥感图像目标检测技术研究[D].天津:天津大学,2022.
[24] ZHANG Zhi,HE Tong,ZHANG Hang,et al.Bag of freebies for training object detection neural networks[EB/OL].[2024-10-13].http://doi.org/10.48550/arxiv.1902.04103.
[25] LECUN Y,BENGIO Y,HINTON G.Deep learning[J].Nature,2015,521(7553): 436-444.
[26] 吴志高,陈明.基于改进YOLO v7的微藻轻量级检测方法[J].大连海洋大学学报,2023,38(1):129-139.
[27] 苏鹏创.大地背景弱小目标识别技术研究[D].西安:西安工业大学,2023.
[28] 候瑞环,杨喜旺,王智超,等.一种基于YOLOv4-TIA的林业害虫实时检测方法[J].计算机工程,2022,48(4):255-261.
[29] 马永康,刘华,凌成星,等.基于改进YOLOv5的红树林单木目标检测研究[J].激光与光电子学进展,2022,59(18):436-446.