文献类型：期刊文献

英文题名：Individual Tree Crown Segmentation and Crown Width Extraction From a Heightmap Derived From Aerial Laser Scanning Data Using a Deep Learning Framework

作者：Sun, Chenxin[1] Huang, Chengwei[1] Zhang, Huaiqing[2] Chen, Bangqian[3] An, Feng[3] Wang, Liwen[1] Yun, Ting[1,4]

第一作者：Sun, Chenxin

通信作者：Yun, T[1];Yun, T[2]

机构：[1]Nanjing Forestry Univ, Sch Informat Sci & Technol, Nanjing, Peoples R China;[2]Chinese Acad Forestry, Res Inst Forest Resource Informat Tech, Beijing, Peoples R China;[3]Chinese Acad Trop Agr Sci, Rubber Res Inst, Danzhou Invest & Expt Stn Trop Crops, Minist Agr, Danzhou, Peoples R China;[4]Nanjing Forestry Univ, Coll Forestry, Nanjing, Peoples R China

年份：2022

卷号：13

外文期刊名：FRONTIERS IN PLANT SCIENCE

收录：;Scopus(收录号：2-s2.0-85133527210);WOS:【SCI-EXPANDED(收录号:WOS:000817151800001)】；

基金：Funding This research was financially supported by the Foundation Research Funds of IFRIT (CAFYBB2019SZ004) and the National Natural Science Foundation of China (grant numbers: 31770591, 32071681, and 42071418).

语种：英文

外文关键词：airborne LiDAR; deep learning; heightmap; individual tree crown segmentation; forest parameter retrieval

摘要：Deriving individual tree crown (ITC) information from light detection and ranging (LiDAR) data is of great significance to forest resource assessment and smart management. After proof-of-concept studies, advanced deep learning methods have been shown to have high efficiency and accuracy in remote sensing data analysis and geoscience problem solving. This study proposes a novel concept for synergetic use of the YOLO-v4 deep learning network based on heightmaps directly generated from airborne LiDAR data for ITC segmentation and a computer graphics algorithm for refinement of the segmentation results involving overlapping tree crowns. This concept overcomes the limitations experienced by existing ITC segmentation methods that use aerial photographs to obtain texture and crown appearance information and commonly encounter interference due to heterogeneous solar illumination intensities or interlacing branches and leaves. Three generative adversarial networks (WGAN, CycleGAN, and SinGAN) were employed to generate synthetic images. These images were coupled with manually labeled training samples to train the network. Three forest plots, namely, a tree nursery, forest landscape and mixed tree plantation, were used to verify the effectiveness of our approach. The results showed that the overall recall of our method for detecting ITCs in the three forest plot types reached 83.6%, with an overall precision of 81.4%. Compared with reference field measurement data, the coefficient of determination R-2 was >= 79.93% for tree crown width estimation, and the accuracy of our deep learning method was not influenced by the values of key parameters, yielding 3.9% greater accuracy than the traditional watershed method. The results demonstrate an enhancement of tree crown segmentation in the form of a heightmap for different forest plot types using the concept of deep learning, and our method bypasses the visual complications arising from aerial images featuring diverse textures and unordered scanned points with irregular geometrical properties.