检索规则说明:AND代表“并且”;OR代表“或者”;NOT代表“不包含”;(注意必须大写,运算符两边需空一格)
检 索 范 例 :范例一: (K=图书馆学 OR K=情报学) AND A=范并思 范例二:J=计算机应用与软件 AND (U=C++ OR U=Basic) NOT M=Visual
作 者:何继 崔瑞豪 李虎民 王磊 马飞 王培俊 HE Ji;CUI Ruihao;LI Humin;WANG Lei;MA Fei;WANG Peijun(Inner Mongolia Mengtai Buliangou Coal Industry Co.,Ltd.,Ordos,Inner Mongolia 010300,China;School of Environment and Spatial Informatics,China University of Mining and Technology,Xuzhou,Jiangsu 221116,China;Huadian Coal Industry Group Digital Intelligence Technology Co.,Ltd.,Fangshan,Beijing 102400,China;School of Public Administration,China University of Mining and Technology,Xuzhou 221116,China)
机构地区:[1]内蒙古蒙泰不连沟煤业有限责任公司,内蒙古自治区鄂尔多斯市010300 [2]中国矿业大学环境与测绘学院,江苏省徐州市221116 [3]华电煤业集团数智技术有限公司,北京市房山区102400 [4]中国矿业大学公共管理学院,江苏省徐州市221116
出 处:《中国煤炭》2024年第11期142-152,共11页China Coal
基 金:内蒙古自治区科技计划项目(2022YFHH0071)。
摘 要:人工植被重建是矿区土地复垦与生态修复的基础性工作,分析人工植被重建效果是评价生态修复质量的重要环节。以内蒙古蒙泰不连沟煤矿土地复垦示范区为主要研究区,采用无人机搭载多光谱相机进行数据采集,并应用3种常见的机器学习算法,综合地物光谱、纹理的差异提取人工植被。在此基础上,选取植被生物量指标,分析研究区坡顶人工植被生长状况,并评价其植被重建效果。研究结果表明,对机器学习分类方法进行超参数调优能够有效提升模型分类精度,超参数调优后随机森林分类模型分类精度最高,总体精度为82.11%,Kappa系数为0.77;使用随机森林模型获得的反演结果精度最高,其R^(2)为0.93,均方根误差为13.97。示范区坡顶的草本植被重建效果欠佳,有接近66.6%面积区域生物量等级处于0~50 g/m^(2)的范围;灌木重建效果较好,有接近61.4%面积区域生物量等级处于100~150 g/m^(2)的范围。Artificial vegetation reconstruction is the foundational work for land reclamation and ecological restoration in mining areas,and analyzing the effects of artificial vegetation reconstruction is an important link of ecological restoration quality evaluation.Taking the land reclamation demonstration area of Inner Mongolia Mengtai Buliangou Coal Mine as the main study area,Using UAV with multispectral cameras for data acquisition and three common machine learning algorithms for application,artificial vegetation data was extracted based on the differences in spectral and texture characteristics of land and land cover.On this basis,selecting the vegetation biomass index to analyze the growth status of artificial vegetation on the ridges of the study area,its artificial vegetation reconstruction effect was evaluated.The results showed that hyperparameter tuning for machine learning classification methods could effectively improve the classification accuracy of the model,the random forest classification model had the highest classification accuracy after hyperparameter tuning,with an overall accuracy of 82.11%and a Kappa coefficient of 0.77;the inversion results obtained by the random forest model had the highest accuracy,with an R^(2) of 0.93 and a root mean square error of 13.97;the reconstruction effect of herbaceous vegetation on the ridges of the demonstration area was poor,with nearly 66.6%of the area having a biomass grade in the range of 0-50 g/m^(2);the reconstruction effect of shrubs was better,with nearly 61.4%of the area having a biomass grade in the range of 100-150 g/m^(2).
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在链接到云南高校图书馆文献保障联盟下载...
云南高校图书馆联盟文献共享服务平台 版权所有©
您的IP:216.73.216.38