Post-earthquake building damage detection method based on unmanned aerial vehicle video

Abstract

The invention provides a post-earthquake building damage detection method based on an unmanned aerial vehicle video, and the method comprises the steps: step 1, carrying out preprocessing of collectedpost-earthquake unmanned aerial vehicle video data, and obtaining a video key frame sequence; step 2, generating building three-dimensional point cloud data by utilizing the video key frame sequence,performing building structure damage detection on the three-dimensional point cloud data by adopting a method based on point cloud structure feature analysis and deep learning feature analysis, and if the detected result is structure damage, entering step 4; if detecting that the structure is not damaged, executing the third step; step 3, obtaining a two-dimensional key frame image of a to-be-detected building, carrying out building facade damage detection by adopting a method based on deep learning feature analysis, then carrying out super-pixel segmentation on a two-dimensional key frame image of a to-be-detected building, and carrying out fusion post-processing optimization on an obtained building facade damage detection result by utilizing a super-pixel segmentation result; and 4, outputting a damage detection result of the to-be-detected building.

Description

technical field [0001] The present invention relates to the field of remote sensing application technology and the field of disaster assessment, in particular to the post-earthquake building damage detection technology based on unmanned aerial vehicle video, in particular to video key frame selection based on ambiguity and overlap analysis, and building damage detection technology based on computer vision technology. 3D point cloud reconstruction, building structural damage detection based on 3D point cloud deep learning and structural feature analysis, and building facade damage detection based on 2D key frame image deep learning and superpixel segmentation algorithm. Background technique [0002] After the earthquake, accurate acquisition of building damage information at the first time can provide important technical support and decision-making basis for emergency rescue, decision-making command, and post-earthquake reconstruction. Traditional satellite remote sensing ima...

AI Technical Summary

Problems solved by technology

On the one hand, similar to airborne images, this method is limited by factors such as the shooting height and the terrain of the disaster area, and inevitably has defects such as occlusion of ground objects and dead angles of photography; on the other hand, pre-disaster / post-disaster remote sensing image change detection methods often Based on the remote sensing images of the same sensor in different time phases, it is quite difficult to obtain the remote sensing data of the same sensor in the emergency monitoring and evaluation of actual earthquake disasters

Images