无人机森林航摄影像三维点云估测林分蓄积量研究
|
摘要
森林蓄积量能够评估林地生产力的高低及经营措施的效果,为森林经营与采伐提供重要依据。目前,大多基于无人机影像的蓄积量估算,均建立在测绘标准所生成的DOM、DSM、DEM等测绘成果基础上,而未充分利用原始影像数据上的林业特征,无法从点云层面上加入林业业务逻辑产生成果数据。获取无人机影像后,利用特征点提取与匹配方法自动相对定向,结合控制点和光束法平差的迭代求解,解算出精确的相机姿态数据,并沿核线方向一维搜索特征点进行影像密集匹配生成密集点云。对原始三维点云过滤后进行树冠分割,在聚类后的林冠点云中提取了树顶点和树高因子估测了森林蓄积量。研究结果表明,冠幅的提取精度85.15%,树高的提取精度83.69%,林分蓄积量估算的精度达到了82.46%。
The amount of forest stocks can assess the productivity of forest land and the effects of management measures, so as to provide important basis for forest management and harvesting. At present, most of the forest volume estimation based on UAV images was based on DOM, DSM, DEM and other surveying and mapping results generated by surveying and mapping standards. However, the forestry characteristics of the original image data were not suf?ciently utilized, and the forestry business logic can not be added to the point cloud level platform to produce the achievement data. After obtaining the uav image, based on the method of feature point extraction and matching and automatic relative orientation, combining the iterative solution of control point and beam adjustment, the accurate attitude data of the camera was calculated, and the feature points were searched along the direction of the core line in one dimension for dense image matching and formation of dense point cloud. After ?ltering the original three-dimensional point cloud, the canopy segmentation was carried out, and the tree vertex and tree height factors were extracted from the clustering point cloud to estimate the forest accumulation. The results indicate that the extractive accuracy of the crown breadth was 85.15%, the extractive accuracy of the tree height was 83.69%, and the estimation accuracy of the forest stand volume reached 82.46%.
The amount of forest stocks can assess the productivity of forest land and the effects of management measures, so as to provide important basis for forest management and harvesting. At present, most of the forest volume estimation based on UAV images was based on DOM, DSM, DEM and other surveying and mapping results generated by surveying and mapping standards. However, the forestry characteristics of the original image data were not suf?ciently utilized, and the forestry business logic can not be added to the point cloud level platform to produce the achievement data. After obtaining the uav image, based on the method of feature point extraction and matching and automatic relative orientation, combining the iterative solution of control point and beam adjustment, the accurate attitude data of the camera was calculated, and the feature points were searched along the direction of the core line in one dimension for dense image matching and formation of dense point cloud. After ?ltering the original three-dimensional point cloud, the canopy segmentation was carried out, and the tree vertex and tree height factors were extracted from the clustering point cloud to estimate the forest accumulation. The results indicate that the extractive accuracy of the crown breadth was 85.15%, the extractive accuracy of the tree height was 83.69%, and the estimation accuracy of the forest stand volume reached 82.46%.
引文
[1]杨柳,陈延辉,岳德鹏,等.无人机遥感影像的城市绿地信息提取[J].测绘科学,2016,40(7):1-9.
[2]何海清.低空摄影测量航带重构及数据处理方法研究[J].测绘学报,2014,43(4):440.
[3]曹明兰,张力小,王强.无人机遥感影像中行道树信息快速提取[J].中南林业科技大学学报,2016,36(10):89-93.
[4]何原荣,陈鉴知,林泉,等.航拍影像与点云数据在矿区生态修复中的应用[J].中南林业科技大学学报,2017,37(4):79-85.
[5]李长青,曹明兰,李亚东,等.低空无人机航测露天矿三维重构方法与试验[J].测绘通报,2018(3):89-92.
[6]程武学,杨存建,周介铭,等.森林蓄积量遥感定量估测研究综述[J].安徽农业科学,2009,37(16):7746-7750.
[7]王佳,尹华丽,王晓莹,等.基于资源三号影像的森林蓄积量估测遥感因子选择[J].中南林业科技大学学报,2015,35(12):29-33.
[8]李亚东,冯仲科,明海军,等.无人机航测技术在森林蓄积量估测中的应用[J].测绘通报,2017(4):63-66.
[9]张园,陶萍,梁世祥,等.无人机遥感在森林资源调查中的应用[J].西南林学院学报,2011,31(3):49-53
[10]SHENG Y,GONG P,BLGLNG G S.Model-Based ConiferCrown Surface Reconstruct ion from High-Resolut ion Aerial Images[J].Photogrammetric Engineering&Remote Sensing,2001,67(8):957-965.
[11]蒲莹,曾伟生.基于树高和树冠因子的立木材积模型研究[J].林业科学研究,2016,29(2):196-201.
[2]何海清.低空摄影测量航带重构及数据处理方法研究[J].测绘学报,2014,43(4):440.
[3]曹明兰,张力小,王强.无人机遥感影像中行道树信息快速提取[J].中南林业科技大学学报,2016,36(10):89-93.
[4]何原荣,陈鉴知,林泉,等.航拍影像与点云数据在矿区生态修复中的应用[J].中南林业科技大学学报,2017,37(4):79-85.
[5]李长青,曹明兰,李亚东,等.低空无人机航测露天矿三维重构方法与试验[J].测绘通报,2018(3):89-92.
[6]程武学,杨存建,周介铭,等.森林蓄积量遥感定量估测研究综述[J].安徽农业科学,2009,37(16):7746-7750.
[7]王佳,尹华丽,王晓莹,等.基于资源三号影像的森林蓄积量估测遥感因子选择[J].中南林业科技大学学报,2015,35(12):29-33.
[8]李亚东,冯仲科,明海军,等.无人机航测技术在森林蓄积量估测中的应用[J].测绘通报,2017(4):63-66.
[9]张园,陶萍,梁世祥,等.无人机遥感在森林资源调查中的应用[J].西南林学院学报,2011,31(3):49-53
[10]SHENG Y,GONG P,BLGLNG G S.Model-Based ConiferCrown Surface Reconstruct ion from High-Resolut ion Aerial Images[J].Photogrammetric Engineering&Remote Sensing,2001,67(8):957-965.
[11]蒲莹,曾伟生.基于树高和树冠因子的立木材积模型研究[J].林业科学研究,2016,29(2):196-201.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |