The content of this article comes from the "Surveying and Mapping Bulletin" No. 3 of 2024, drawing review number: GS Jing (2024) No. 0499
A height measurement method for a single high-rise building based on convolutional neural network and UAV oblique photographic image
Gao Gui-tang1, Xi Lian-xia2, Zhong Xiao-long1
1. Survey Branch of State Nuclear Power Planning and Design Institute Co., Ltd., Beijing 100095, China; 2. Beijing SuperMap Software Co., Ltd., Beijing 100600, China
Keywords: convolutional neural network, UAV oblique photography, image enhancement, shadow extraction, height measurement
Citation format: Gao Guitang, Xi Lianxia, Zhong Xiaolong. Height Measurement Method of a Single High-rise Building Based on Convolutional Neural Network and UAV Oblique Photographic Image[J]. Bulletin of Surveying and Mapping, 2024(3): 118-122. DOI: 10.13474/j.cnki.11-2246.2024.0320
Abstract:In order to improve the height measurement effect of a single high-rise building, this paper proposes a height measurement method for a single high-rise building based on convolutional neural network (UAV) oblique photography image. Firstly, UAV oblique photography technology was used to collect images of high-rise buildings. Then, the image of the high-rise building was input into the convolutional neural network, and the enhancement process was carried out to improve the clarity of the image. Finally, the shadow length is calculated by the band conversion index and the shadow index, and the height of a single high-rise building is calculated based on this. The experimental results show that the proposed method has better image enhancement processing effect, better height measurement accuracy and better efficiency.
About authorAbout author:GAO Guitang (1982—), male, master, senior engineer, mainly engaged in the application of neural network and image recognition technology in surveying and mapping. E-mail:[email protected]
First trial: Ji Yinxiao review: Song Qifan
Final Judge: Jin Jun
information