中国电建集团华东勘测设计研究院有限公司,浙江 杭州 310000
何明杰(1973—),男,教授级高工,学士。主要从事水工结构设计和边坡灾害治理研究。E-mail: he_mj@hdec.com
刘德方(1994—),男,博士(后)。主要从事边坡工程防灾减灾方面的研究。E-mail: liudf0303@163.com
收稿:2023-01-05,
修回:2023-03-13,
纸质出版:2023-12-15
移动端阅览
何明杰,刘德方,张猛等.融合YOLOX和ASFF的高原山地灾害检测模型[J].防灾减灾工程学报,2023,43(06):1215-1223.
HE Mingjie,LIU Defang,ZHANG Meng,et al.A Plateau Mountain Disaster Detection Model by Integrating YOLOX and ASFF[J].Journal of Disaster Prevention and Mitigation Engineering,2023,43(06):1215-1223.
何明杰,刘德方,张猛等.融合YOLOX和ASFF的高原山地灾害检测模型[J].防灾减灾工程学报,2023,43(06):1215-1223. DOI: 10.13409/j.cnki.jdpme.20230105002.
HE Mingjie,LIU Defang,ZHANG Meng,et al.A Plateau Mountain Disaster Detection Model by Integrating YOLOX and ASFF[J].Journal of Disaster Prevention and Mitigation Engineering,2023,43(06):1215-1223. DOI: 10.13409/j.cnki.jdpme.20230105002.
YOLOX是全球首个一阶无锚框目标检测模型,超越了YOLO‑V(3‑5)和SSD等传统锚框模型,极大提高了山地灾害检测识别精度。然而,该模型存在不同尺度之间特征不一致的问题,融合后的特征图质量有待提升。以西藏高原山地灾害重灾区为试验区,在建立了西藏高原山地灾害数据集的基础上,通过融合Adaptively Spatial Feature Fusion(ASFF)注意力机制和骨干网络尺寸调整机制,设计了一个优化的、不同适用性的高原山地灾害检测模型(Plateau mountain disaster detection model,PMDDM)。为验证PMDDM模型的优越性,将其与传统YOLOX模型、不同注意力机制、不同目标检测模型进行了对比分析,并且对不同尺度模型的检测性能和可视化结果也进行了对比分析研究。结果表明:ASFF注意力机制可以有效的解决传统YOLOX模型中存在的不同尺度特征间的特征不一致问题,且对模型检测性能提升明显优于SE、CBAM、ECA、GAM和Coord等注意力机制;PMDDM模型对山地灾害的检测精度优于Faster‑RCNN、SSD和YOLO‑V3模型,可以满足不同工作场景对硬件配置、检测速度和精度的需求,且模型尺度越大,识别目标的准确率越高。
YOLOX is the world's earliest first-order anchor-free frame target detection model, which surpasses traditional anchor frame models such as YOLO-V (3-5) and SSD, and greatly improves the accuracy of mountain disaster detection and recognition. However, the model suffers from feature inconsistency between features at different scales, and the quality of the fused feature map needs to be improved. This paper presents the mountain disaster-stricken areas on the Tibetan plateau as the experimental site, based on the basis of the established Tibetan plateau mountain disaster dataset, introduces the Adaptively Spatial Feature Fusion (ASFF) attention mechanism and backbone network size adjustment mechanism to design an optimized system with different applicability. The new plateau mountain disaster detection model (PMDDM) improves the detection accuracy of plateau mountain hazards and enhances the convenience of plateau mountain hazard detection. Meanwhile, the PMDDM model was also compared with traditional YOLOX and other models with different attention mechanisms and different object detection to verify it superiority. The detection performance and visualization results of different scale models are also comparatively analyzed and studied in this paper. The results show that the ASFF attention mechanism can effectively solve the problem of feature inconsistency among different scale features in the traditional YOLOX model; the ASFF attention mechanism improves the model detection performance much better than the attention mechanisms of SE, CBAM, ECA, GAM and Coord; the detection accuracy of the PMDDM model for mountain hazards is better than that of Faster-RCNN, SSD and YOLO-V3 models; the PMDDM model can meet requirements of hardware configuration, detection speed and accuracy for different working scenarios; the larger the scale of PMDDM model, the higher the accuracy rate of identifying targets.
Zhang L M , Xiao T , He J , et al . Erosion-based analysis of breaching of Baige landslide dams on the Jinsha River, China, in 2018 [J]. Landslides , 2019 , 16 ( 10 ): 1965 - 1979 .
Jia H C , Chen F , Pan D H . Disaster chain analysis of avalanche and landslide and the river blocking dam of the Yarlung Zangbo River in Milin County of Tibet on 17 and 29 October 2018 [J]. International Journal of Environmental Research and Public Health , 2019 , 16 ( 23 ): 4707 .
Liu G , Li J . Research on the effect of tree barriers on rockfall using a three-dimensional discontinuous deformation analysis method [J]. International Journal of Computational Methods , 2020 , 17 ( 8 ): 1950046 .
李睿祺 , 胡桂胜 , 陈宁生 , 等 . 川藏铁路沟谷型泥石流形成运动特征与工程危害研究 [J]. 防灾减灾工程学报 , 2021 , 41 ( 3 ): 432 - 440,462 .
Li R Q , Hu G S , Chen N S , et al . Study on formation and motion characteristics and engineering hazards of gully type debris flow in Sichuan Tibet Railway [J]. Journal of Disaster Prevention and Mitigation Engineering , 2021 , 41 ( 3 ): 432 - 440,462 . (in Chinese)
张庆海 . 1/5万县(市)地质灾害调查野外工作方法概述 [J]. 科技资讯 , 2015 , 13 ( 22 ): 220 - 221 .
Zhang Q H . 1 / 5 Overview of fieldwork methods for geohazard surveys in 10,000 counties (cities) [J]. Science & Technology Information, 2015 , 13 ( 22 ): 220 - 221 . (in Chinese)
雷添杰 , 李长春 , 何孝莹 . 无人机航空遥感系统在灾害应急救援中的应用 [J]. 自然灾害学报 , 2011 , 20 ( 1 ): 178 - 183 .
Lei T J , Li C C , He X Y . Application of aerial remote sensing of pilotless aircraft to disaster emergency rescue [J]. Journal of Natural Disasters , 2011 , 20 ( 1 ): 178 - 183 . (in Chinese)
杨婷婷 . 基于遥感综合分析的灾害地质研究 [D]. 北京 : 中国地质大学(北京) , 2020 .
Yang T T . Research on disaster geology based on comprehensive analysis of remote sensing [D]. Beijing : China University of Geosciences (Beijing) , 2020 . (in Chinese)
叶唐进 , 谢强 , 王鹰 . 国道G318巴塘至拉萨段碎屑斜坡灾害与稳定性分析 [J]. 安全与环境学报 , 2018 , 18 ( 5 ): 1917 - 1922 .
Ye T J , Xie Q , Wang Y . Slope disaster and stability analysis of the G318 Highway in the section from Batang to Lhasa [J]. Journal of Safety and Environment , 2018 , 18 ( 5 ): 1917 - 1922 . (in Chinese)
叶唐进 , 曹景轩 , 张宇 , 等 . 基于三维地形仿真的高原落石运动特征分析 [J]. 高原科学研究 , 2022 , 6 ( 2 ): 60 - 68 .
Ye T J , Cao J X , Zhang Y , et al . Analysis of rockfalls movement characteristics on the Tibetan Plateau based on 3D terrain simulation [J]. Plateau Science Research , 2022 , 6 ( 2 ): 60 - 68 . (in Chinese)
Okwuashi O A , Ndehedehe C E N . Deep support vector machine for hyperspectral image classification [J]. Pattern Recognition , 2023 , DOI: 10.1016/j.patcog. 2020.107298 http://dx.doi.org/10.1016/j.patcog.2020.107298 .
Song Y Y , Ying L . Decision tree methods: applications for classification and prediction [J]. Shanghai Archives of Psychiatry , 2015 , 27 ( 2 ): 130 .
刘汉龙 , 马彦彬 , 仉文岗 . 大数据技术在地质灾害防治中的应用综述 [J]. 防灾减灾工程学报 , 2021 , 41 ( 4 ): 710 - 722 .
Liu H L , Ma Y B , Zhang W G . Application of big data techniques in geological disaster analysis and prevention: a systematic review [J]. Journal of Disaster Prevention and Mitigation Engineering , 2021 , 41 ( 4 ): 710 - 722 . (in Chinese)
裴振伟 , 年廷凯 , 吴昊 , 等 . 滑坡地质灾害应急处置技术研究进展 [J]. 防灾减灾工程学报 , 2021 , 41 ( 6 ): 1382 - 1394 .
Pei Z W , Nian T K , Wu H , et al . Research progress on emergency treatment techniques for landslide geological hazards [J]. Journal of Disaster Prevention and Mitigation Engineering , 2021 , 41 ( 6 ): 1382 - 1394 . (in Chinese)
Hacıefendioğlu K , Başağa H B , Demir G . Automatic detection of earthquake-induced ground failure effects through Faster R-CNN deep learning-based object detection using satellite images [J]. Natural Hazards , 2021 , 105 ( 1 ): 383 - 403 .
Cheng L , Li J , Duan P , et al . A small attentional YOLO model for landslide detection from satellite remote sensing images [J]. Landslides , 2021 , 18 ( 8 ): 1 - 15 .
Su Z Y , Chow J K , Tan P S , et al . Deep convolutional neural network-based pixel-wise landslide inventory mapping [J]. Landslides , 2021 , 18 ( 4 ): 1421 ‑ 1443 .
Ji S P , Yu D W , Shen C Y , 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 .
Ge Z , Liu S , Wang F , et al . YOLOX: exceeding YOLO series in 2021 [J]. ArXiv E-Prints , 2021 , DOI: 10.48550/arXiv.2017.08430 http://dx.doi.org/10.48550/arXiv.2017.08430 .
崔鹏 , 苏凤环 , 邹强 , 等 . 青藏高原山地灾害与气象灾害风险评估与减灾对策 [J]. 科学通报 , 2015 , 60 : 3067 ‑ 3077 .
Cui P , Su F H , Zou Q , et al . Risk assessment and disaster reduction strategies for mountainous and meteorological hazards in Tibetan Plateau [J]. Chinese Science Bulletin , 2015 , 60 : 3067 ‑ 3077 . (in Chinese)
Hungr O , Leroueil S , Picarelli L . The Varnes classification of landslide types, an update [J]. Landslides , 2014 , 11 ( 2 ): 167 - 194 .
Hu J , Shen L , Sun G . Squeeze-and-excitation networks [C]∥ Proceedings of the IEEE Conference on ComPuter Vision and Pattern Recognition . Seattle, WA, USA : IEEE , 2018 : 7132 - 7141 .
Woo S , Park J , Lee J-Y , et al . Cbam: Convolutional block attention module [C]∥ Proceedings of the European Conference on Computer Vision . Seattle, WA, USA : IEEE , 2018 : 3 - 19 .
Wang Q , Wu B , Zhu P , et al . Supplementary material for ‘ECA-Net: Efficient channel attention for deep convolutional neural networks [C]∥ Proceedings of the Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition . Seattle : IEEE , 2020 : 13 - 19 .
Liu Y , Shao Z , Hoffmann N . Global attention mechanism: retain information to enhance channel-spatial interactions [J]. ArXiv E‑Prints , 2021 , DOI: 10.48550/arXiv.2112.05561 http://dx.doi.org/10.48550/arXiv.2112.05561 .
Hou Q , Zhou D , Feng J . Coordinate attention for efficient mobile network design [C]∥ Proceedings of the Proceedings of the IEEE/CVF conference on Computer Vision and Pattern Recognition . Seattle, WA, USA : IEEE , 2021 : 13713 - 13722 .
0
浏览量
0
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
苏公网安备32010202012147号
