文献类型：期刊文献

中文题名：高精度DEM支持下的多时期航片杉木人工林树高生长监测

英文题名：Forest Height Growth Monitoring of Cunninghamia lanceolata Plantation Using Multi-Temporal Aerial Photography with the Support of High Accuracy DEM

作者：夏永杰[1] 庞勇[1] 刘鲁霞[1,2] 陈博伟[1] 董斌[3] 黄庆丰[2]

第一作者：夏永杰

通信作者：Pang, Yong

机构：[1]中国林业科学研究院资源信息研究所;[2]安徽农业大学林学与园林学院;[3]安徽农业大学理学院

年份：2019

卷号：55

期号：4

起止页码：108-121

中文期刊名：林业科学

外文期刊名：Scientia Silvae Sinicae

收录：CSTPCD;;EI(收录号：20193607400803);Scopus;北大核心:【北大核心2017】；CSCD:【CSCD2019_2020】；

基金：国家林业和草原局林业行业公益性科研项目"我国主要林区林地立地质量和生产力评价研究"(201504303);国家自然科学基金项目"基于高分辨率遥感数据的森林生物多样性监测"(31570546);国家科技部863课题"全球森林生物量和碳储量遥感估测关键技术(2012AA12A306)

语种：中文

中文关键词：历史航片;密集匹配;激光雷达;数字高程模型;数字表面模型;冠层高度模型;生长监测

外文关键词：historical aerial photographs;dense matching;LiDAR;DEM;DSM;CHM;growth monitoring

分类号：S757

摘要：【目的】集成多时期航片数据和由机载激光雷达数据获取的密集林区数字高程模型,估测多时期杉木人工林冠层高度,并对其生长情况进行定量监测,为多时期航片监测森林生长趋势和评价林地生产力提供可能。【方法】首先基于分类后的激光雷达点云数据获得林下高精度数字高程模型和森林数字表面模型,利用航片数据构建立体像对,通过自动立体匹配算法生成森林冠层的摄影测量数字表面模型,然后借助数字高程模型将2种数字表面模型进行高度归一化,提取研究区多时期森林冠层高度。利用1996、2004年历史航片和2014年数字航片以及激光雷达数据,构建18年内皖南杉木人工林3期森林冠层高度,并对其精度进行分析。【结果】1)由2014年数字航片和激光雷达数据获取的森林冠层高度的R^2为0. 52,RMSE为1. 79 m; 2)由2014年数字航片处理得到的森林冠层高度与对应样地实测上层木的平均高验证精度较高,平均绝对误差1. 59 m,平均相对误差15%,最大绝对误差3. 45 m,最大相对误差30. 80%,测量精度85. 00%; 3)由1996、2004、2014年航片得到3期杉木人工林冠层高度,其增长趋势与树高生长曲线预测趋势一致。【结论】在多山复杂地形条件下,利用航片可准确定量反映山脊向阳面的森林冠层高度变化,但对于山谷阴影处,则会出现冠层高度被低估情况,利用多期航片结合高精度DEM数据可定量反映上层木的冠层高度变化。
【Objective】 This study integrated multi-temporal aerial photographs and DEM derived from airborne LiDAR data to calculate the forest canopy height of Cunninghamia lanceolata and monitor the variation of growth quantitatively.【Method】 First of all, high accuracy digital elevation model beneath canopy and forestry digital surface model were constructed based on classified LiDAR point cloud data.Digital surface models were then created by applying an automated stereo-matching algorithm to the scanning copy of aerial photographs.These multi-temporal canopy heights were obtained by subtracting the LiDAR ground elevations from the two kinds of DSM.Using historical aerial photographs of 1996, 2004 and digital aerial photographs, LiDAR data of 2014, multi-temporal CHMs were reconstructed within a period of 18 years, and the accuracy was evaluated and analyzed.【Result】 1) The R 2 between the canopy height models acquired by LiDAR data and corresponding digital aerial photographs in 2014 is 0.52, and the root mean square error is 1.79m.2) Compared with the measurements from field plots, our data showed an accuracy of 85.00% with mean absolute errorand mean relative error of 1.59 m and 15.00%, and the maximum absolute error and maximum relative error of 3.45m and 30.80% respectively.3) Combined with the aerial photos of year 1996, 2004 and 2014, these multi-temporal canopy height models of Cunninghamia lanceolata plantation have a similar growth trend to the predicted growth curve.【Conclusion】 Based on the results, utilizing aerial photographs can characterize the variation of canopy height in the sunny slope of mountainous terrain.However, for forests located in the valley bottom, the canopy height would be under estimated with aerial photographs.Multi-temporal aerial photographs combining with the high accuracy DEM can reflect the variation of overstory’s height, which provide the possibility for monitoring the forest’s growth trend and access the forest’s productivity.