
浏览全部资源
扫码关注微信
1.中国科学院、水利部 成都山地灾害与环境研究所, 山地自然灾害与;工程安全重点实验室, 四川 成都 610213
2.中国科学院大学, 北京 100049
Received:27 June 2025,
Revised:2025-08-18,
Published:10 December 2025
移动端阅览
刘海涛, 陈剑刚, 陶紫琴, 等.基于人工智能模型的小流域沟道漂木识别方法[J].水土保持通报,2025,45(6):158-168.
Liu Haitao, Chen Jiangang, Tao Ziqin, et al. Identification method for large wood in small watershed channels based on segment anything model [J]. Bulletin of Soil and Water Conservation,2025,45(6):158-168.
刘海涛, 陈剑刚, 陶紫琴, 等.基于人工智能模型的小流域沟道漂木识别方法[J].水土保持通报,2025,45(6):158-168. DOI: 10.13961/j.cnki.stbctb.2025.06.026. CSTR: 32312.14.stbctb.2025.06.026.
Liu Haitao, Chen Jiangang, Tao Ziqin, et al. Identification method for large wood in small watershed channels based on segment anything model [J]. Bulletin of Soil and Water Conservation,2025,45(6):158-168. DOI: 10.13961/j.cnki.stbctb.2025.06.026. CSTR: 32312.14.stbctb.2025.06.026.
目的
2
介绍一种基于人工智能模型的漂木图像分割方法,为该模型在漂木灾害调查与评估方面的应用提供理论依据。
方法
2
选取西藏自治区昌都市贡觉县则巴沟为研究区,基于人工智能图像分割大模型(segment anything model,SAM),通过引入轻量级适配器、简化掩码解码器、设计多任务损失函数以及添加辅助分类器,构建一种针对漂木图像的分割方法(large wood SAM,LWSAM)。训练时冻结原始图像编码器和提示编码器的参数,以低训练成本提升漂木分割性能,在构建的漂木相机(LW_CAM_dataset)和无人机(LW_UAV_dataset)两个数据集上对模型进行训练与测试,并与当前先进图像分割模型进行对比。
结果
2
①多任务损失函数能从不同角度优化分割质量,有效解决了漂木识别中前景稀疏和类别不平衡的问题,提高了模型对多种漂木形态的适应能力;②相较于SAM方法,在采用点提示的情况下,LWSAM在LW_CAM_dataset数据集上的MDice,MIoU和F
1
分数分别提升15.9%,15.9%和10.0%,在LW_UAV_dataset数据集上的MDice,MIoU和F
1
分数分别提升21.6%,29.6%和16.7%;③漂木分割效果受数据集质量影响,高质量数据集模型分割结果更好。
结论
2
采用LWSAM对漂木图像进行分割是可行的,且在实际运用中表现出较高的精度和较强的鲁棒性,能够准确分割漂木图像,可应用于小流域漂木灾害调查。
Objective
2
An image segmentation method for large wood based on the segment anything model (SAM) was introduced, in order to provide theoretical support for its application in investigating and assessing large wood disasters.
Methods
2
The Zebagou area in Gongjue County, Chamdo City, Xizang Autonomous Region was selected as the study area, based on SAM, a segmentation method for large wood images—large wood SAM (LWSAM)-was developed by introducing a lightweight adapter, simplifying the mask decoder, designing a multi-task loss function, and adding an auxiliary classifier. During training, the parameters of the original image encoder and the prompt encoder were frozen to improve large wood segmentation performance at a low training cost. The model was trained and tested on two datasets, LW_CAM_dataset and LW_UAV_dataset, and compared with current state-of-the-art image segmentation models.
Results
2
① The proposed multi-task loss function could optimize segmentation quality from different perspectives, effectively address the issues of sparse foreground and class imbalance in large wood recognition, and enhance the model’s adaptability to various large wood morphologies. ② Compared with the SAM method, under point prompt conditions, LWSAM achieved improvements of 15.9%, 15.9%, and 10.0% in MDice, MIoU, and F1 score, respectively, on the LW_CAM_dataset, and improvements of 21.6%, 29.6%, and 16.7% on the LW_UAV_dataset, respectively. ③ The performance of large wood segmentation was influenced by dataset quality, with models trained on higher-quality datasets achieving better segmentation results.
Conclusion
2
Using LWSAM for large wood image segmentation is feasible, and it demonstrates high accuracy and strong robustness in practical applications, enabling accurate segmentation of large wood images. This approach can be applied to large wood disaster investigations in small watersheds.
Comiti F , Lucía A , Rickenmann D . Large wood recruitment and transport during large floods:A review [J]. Geomorphology , 2016 , 269 : 23 - 39 .
陈剑刚 , 费高高 , 王喜安 , 等 . 漂木对山洪泥石流运动致灾影响研究进展 [J]. 水利水电科技进展 , 2022 , 42 ( 3 ): 104 - 111 .
Chen Jiangang , Fei Gaogao , Wang Xi’an , et al . Advances on disaster effects of drift wood in flash flood debris flows [J]. Advances in Science and Technology of Water Resources , 2022 , 42 ( 3 ): 104 - 111 .
Fei Gaogao , Wang Xiekang . A review of large wood dynamics relevant to hazard characteristics for built structures [J]. Geomorphology , 2024 , 453 : 109152 .
Chen Jiangang , Liu Wenrun , Zhao Wanyu , et al . Magnitude amplification of flash floods caused by large woody in Keze gully in Jiuzhaigou National Park, China [J]. Geomatics, Natural Hazards and Risk , 2021 , 12 ( 1 ): 2277 - 2299 .
Schalko I , Follett E , Nepf H . Impact of lateral gap on flow distribution, backwater rise, and turbulence generated by a logjam [J]. Water Resources Research , 2023 , 59 ( 10 ): e2023WR034689 .
Schalko I , Lageder C , Schmocker L , et al . Laboratory flume experiments on the formation of spanwise large wood accumulations: Part Ⅱ. Effect on local scour [J]. Water Resources Research , 2019 , 55 ( 6 ): 4871 - 4885 .
Ruiz-Villanueva V , Piégay H , Gaertner V , et al . Wood density and moisture sorption and its influence on large wood mobility in rivers [J]. Catena , 2016 , 140 : 182 - 194 .
De Cicco P N , Paris E , Solari L , et al . Bridge pier shape influence on wood accumulation:Outcomes from flume experiments and numerical modelling [J]. Journal of Flood Risk Management , 2020 , 13 ( 2 ): e12599 .
杨华铨 , 柳金峰 , 孙昊 , 等 . 四川木里县项脚沟“7·5”特大型泥石流特征及发展趋势分析 [J]. 中国地质灾害与防治学报 , 2024 , 35 ( 1 ): 100 - 107 .
Yang Huaquan , Liu Jinfeng , Sun Hao , et al . Analysis of the characteristics and development trends of the “7 · 5”catastrophic debris flow in Xiangjiao gully, Muli County, Sichuan [J]. The Chinese Journal of Geological Hazard and Control , 2024 , 35 ( 1 ): 100 - 107 .
May C L , Gresswell R E . Processes and rates of sediment and wood accumulation in headwater streams of the Oregon Coast Range, USA [J]. Earth Surface Processes and Landforms , 2003 , 28 ( 4 ): 409 - 424 .
MacVicar B , Piégay H . Implementation and validation of video monitoring for wood budgeting in a wandering piedmont river, the Ain River (France) [J]. Earth Surface Processes and Landforms , 2012 , 37 ( 12 ): 1272 - 1289 .
He Kaiming , Gkioxari G , Dollár P , et al . Mask R-CNN [C]∥ 2017 IEEE International Conference on Computer Vision (ICCV) . October 22-29, 2017 , Venice, Italy . IEEE , 2017 : 2980 - 2988 .
Han Kai , Wang Yunhe , Chen Hanting , et al . A survey on vision transformer [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2023 , 45 ( 1 ): 87 - 110 .
Kirillov A , Mintun E , Ravi N , et al . Segment anything [C]∥ 2023 IEEE/CVF International Conference on Computer Vision (ICCV) . 2023 , Paris, France . IEEE , 2024 : 3992 - 4003 .
周洁 , 方振宇 . 基于SAM多尺度标签优化的半监督学习遥感目标检测 [J/OL]. 微电子学与计算机 , 2024 : 1 - 10 .( 2024-12-24 ) [ 2025-06-27 ]. https:∥kns.cnki.net/kcms/detail/61.1123.TN.20241224.0839.002.html https://kns.cnki.net/kcms/detail/61.1123.TN.20241224.0839.002.html .
Zhou Jie , Fang Zhenyu . Semi-supervised learning remote sensing target detection based on SAM multi-scale label optimization [J/OL]. Microelectronics & Computer , 2024 : 1 - 10 .( 2024-12-24 ) [ 2025-06-27 ]. https:∥kns.cnki.net/kcms/detail/61.1123.TN.20241 224.0839.002.html https://kns.cnki.net/kcms/detail/61.1123.TN.20241224.0839.002.html .
张振伟 , 蔡可天 , 高轩 , 等 . 基于SAM图像处理的堆石料级配计算方法及验证 [J]. 水力发电 , 2025 , 51 ( 2 ): 80 - 86 .
Zhang Zhenwei , Cai Ketian , Gao Xuan , et al . Calculation method and verification of rockfill gradation based on SAM image processing [J]. Water Power , 2025 , 51 ( 2 ): 80 - 86 .
张鸿 , 杨俊雅 , 刘可心 , 等 . 基于Stone-SAM的便携式粗集料级配智能检测 [J]. 建筑材料学报 , 2025 , 28 ( 6 ): 581 - 590 .
Zhang Hong , Yang Junya , Liu Kexin , et al . Portable intelligent detection of coarse aggregate gradation based on stone-SAM [J]. Journal of Building Materials , 2025 , 28 ( 6 ): 581 - 590 .
付立群 , 金峰 , 张喜喜 , 等 . 结合Mask R-CNN和SAM获取堆石混凝土坝堆石级配曲线 [J]. 水电能源科学 , 2024 , 42 ( 11 ): 7 - 11 .
Fu Liqun , Jin Feng , Zhang Xixi , et al . Obtaining particle size distribution curves for rock-filled concrete dams by combining mask R-CNN and SAM [J]. Water Resources and Power , 2024 , 42 ( 11 ): 7 - 11 .
马小川 , 付佳 , 王李廷煜 , 等 . SAM特征引导的主动学习在缺陷检测中的应用 [J]. 电子机械工程 , 2025 , 41 ( 3 ): 80 - 86 .
Ma Xiaochuan , Fu Jia , Wang L , et al . Application of active learning to defect detection guided by SAM feature [J]. Electro-Mechanical Engineering , 2025 , 41 ( 3 ): 80 - 86 .
陶攀 , 方宇 , 王欣 , 等 . 基于改进SAM模型的多任务轨道缺陷检测方法 [J]. 南京大学学报(自然科学) , 2024 , 60 ( 5 ): 776 - 784 .
Tao Pan , Fang Yu , Wang Xin , et al . Multi-task track defect detection method based on improved SAM model [J]. Journal of Nanjing University (Natural Sciences) , 2024 , 60 ( 5 ): 776 - 784 .
刘娜 , 封筠 , 霍一儒 , 等 . SAMCP:一种轻量级微调SAM的结肠息肉分割方法 [J/OL]. 计算机应用 , 2025 : 1 - 14 .( 2025-02-27 ). https:∥kns.cnki.net/kcms/detail/51.1307.TP.20250227.1119.002.html https://kns.cnki.net/kcms/detail/51.1307.TP.20250227.1119.002.html .
Liu Na , Feng Jun , Huo Yiru , et al . SAMCP: Lightweight SAM fine-tuning method for colon polyp segmentation [J/OL]. Journal of Computer Applications , 2025 : 1 - 14 .( 2025-02-27 ). https:∥kns.cnki.net/kcms/detail/51.1307.TP.20250227.1119.002.html https://kns.cnki.net/kcms/detail/51.1307.TP.20250227.1119.002.html .
刘复昌 , 蔡煜晨 , 缪永伟 , 等 . 基于预训练SAM的提示式三维牙齿分割方法 [J]. 浙江大学学报(理学版) , 2025 , 52 ( 1 ): 59 - 69 .
Liu Fuchang , Cai Yuchen , Miao Yongwei , et al . Prompt-based three-dimensional tooth segmentation method based on pre-trained SAM [J]. Journal of Zhejiang University (Science Edition) , 2025 , 52 ( 1 ): 59 - 69 .
Li Xiaoya , Sun Xiaofei , Meng Yuxian , et al . Dice loss for data-imbalanced NLP tasks [C]∥ Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics . Online . Stroudsburg, PA, USA : ACL , 2020 : 465 - 476 .
Rahman M A , Wang Yang . Optimizing intersection-over-union in deep neural networks for image segmentation [C]∥ Advances in Visual Computing . Cham : Springer , 2016 : 234 - 244 .
Mannor S , Peleg D , Rubinstein R . The cross entropy method for classification [C]∥ Proceedings of the 22nd International Conference on Machine Learning . 2005 , Bonn, Germany . ACM , 2005: 561 - 568 .
Ronneberger O , Fischer P , Brox T . U-Net:Convolutional networks for biomedical image segmentation [C]∥ Medical Image Computing and Computer-Assisted Intervention-MICCAI 2015 . Cham : Springer , 2015 : 234 - 241 .
Castro R , Ramos L , Román S , et al . U-net vs. transunet: performance comparison in medical image segmentation [C] ∥ International Conference on Applied Technologies . Cham : Springer Nature Switzerland , 2022 : 212 - 226 .
Chen L C , Zhu Yukun , Papandreou G , et al . Encoder-decoder with atrous separable convolution for semantic image segmentation [C]∥ Computer Vision-ECCV 2018 . Cham : Springer , 2018 : 833 - 851 .
Vaka I R , Sundharakumar K B . Comparative Analysis for SAM, FastSAM, EfficientSAM, Detectron 2 for Semantic Segmentation in Self Driving Cars [C] ∥ International Conference on Computer Vision and Image Processing . Cham : Springer Nature Switzerland , 2024 : 281 - 294 .
0
Views
2
下载量
0
CSCD
Publicity Resources
Related Articles
Related Author
Related Institution
京公网安备11010802024621