文献类型：期刊文献

英文题名：Very High Resolution Images and Superpixel-Enhanced Deep Neural Forest Promote Urban Tree Canopy Detection

作者：Liu, Yang[1,2,3] Zhang, Huaiqing[1,2,3] Cui, Zeyu[1,2,3] Lei, Kexin[1,2,3] Zuo, Yuanqing[1,2,3] Wang, Jiansen[1,2,3] Hu, Xingtao[1,2,3,4] Qiu, Hanqing[1,2,3]

第一作者：Liu, Yang;刘燕

通信作者：Zhang, HQ[1];Zhang, HQ[2];Zhang, HQ[3]

机构：[1]Chinese Acad Forestry, Inst Forest Resource Informat Tech, Beijing 100091, Peoples R China;[2]Natl Forestry & Grassland Adm, Key Lab Forestry Remote Sensing & Informat Syst, Beijing 100091, Peoples R China;[3]Dongting Lake Remote Sensing Prod Validat Stn, Beijing 100091, Peoples R China;[4]Guizhou Normal Univ, Sch Geog & Environm Sci, Guiyang 550025, Peoples R China

年份：2023

卷号：15

期号：2

外文期刊名：REMOTE SENSING

收录：;EI(收录号：20230413427880);Scopus(收录号：2-s2.0-85146548224);WOS:【SCI-EXPANDED(收录号:WOS:000927246000001)】；

基金：This research was funded by the Foundation Research Funds of Institute of Forest Resource Information Techniques (IFRIT), grant number CAFYBB2019SZ004 and the National Natural Science Foundation of China, grant number 32071681.

语种：英文

外文关键词：VHR; urban tree canopy; superpixel-enhanced deep neural forest; remote sensing

摘要：Urban tree canopy (UTC) area is an important index for evaluating the urban ecological environment; the very high resolution (VHR) images are essential for improving urban tree canopy survey efficiency. However, the traditional image classification methods often show low robustness when extracting complex objects from VHR images, with insufficient feature learning, object edge blur and noise. Our objective was to develop a repeatable method-superpixel-enhanced deep neural forests (SDNF)-to detect the UTC distribution from VHR images. Eight data expansion methods was used to construct the UTC training sample sets, four sample size gradients were set to test the optimal sample size selection of SDNF method, and the best training times with the shortest model convergence and time-consumption was selected. The accuracy performance of SDNF was tested by three indexes: F1 score (F1), intersection over union (IoU) and overall accuracy (OA). To compare the detection accuracy of SDNF, the random forest (RF) was used to conduct a control experiment with synchronization. Compared with the RF model, SDNF always performed better in OA under the same training sample size. SDNF had more epoch times than RF, converged at the 200 and 160 epoch, respectively. When SDNF and RF are kept in a convergence state, the training accuracy is 95.16% and 83.16%, and the verification accuracy is 94.87% and 87.73%, respectively. The OA of SDNF improved 10.00%, reaching 89.00% compared with the RF model. This study proves the effectiveness of SDNF in UTC detection based on VHR images. It can provide a more accurate solution for UTC detection in urban environmental monitoring, urban forest resource survey, and national forest city assessment.