Aviation image building damage detection method based on shadow and texture characteristics

Abstract

The invention discloses an aviation image building damage detection method based on shadow and texture characteristics, and the method comprises the steps: estimating a theoretical shadow area of a building in an image through the vector data and altitude data of the building before damages and a solar altitude angle; carrying out the detection of an actual shadow in the theoretical shadow area through employing constrained color invariance; obtaining the actual shadow area of the building; obtaining the damage level of the building according to the proportional relation of the actual shadow area and the theoretical shadow area, and dividing the damage level into three levels: complete damage, general damage, and suspected intactness; detecting the top surface of the intact building through employing a vision bag-of-word model, and further judging whether the building is damaged or not. The method integrates the shadow information (height) and top surface information (texture) of the building for detection, avoids registering difficulties in conventional data fusion, and improves the detection accuracy of the damage of the building.

Description

technical field [0001] The invention belongs to the technical field of remote sensing image applications, in particular to a building damage detection method in aerial images that combines building shadow and texture features. Background technique [0002] Natural disasters have caused huge losses of human life and property for a long time, and are a huge obstacle to human survival and development. Remote sensing technology has the characteristics of short revisit period, large detection range and high data comprehensiveness, which provides a favorable means for disaster monitoring and assessment. With the development of various monitoring methods and high-tech, the traditional disaster detection and evaluation gradually develops from qualitative statistical evaluation to quantitative fine evaluation. As the core elements of people's production and life, buildings are of great significance to detect and extract damage information after natural disasters, which can provide i...

AI Technical Summary

Problems solved by technology

The above-mentioned change detection methods using high-resolution remote sensing images have achieved good results. However, since these methods are mainly based on two-dimensional data change detection, it is difficult to detect building height change information. Buildings with a change in height such as the collapse of the lower part or the collapse of the middle storey have congenital defects

3) Using different time phase LIDAR (Light Detection and Ranging, lidar) data or stereo image pairs for building damage detection, both LIDAR data and stereo image pairs contain three-dimensional information of ground objects, through the extraction of DSM (Digital Surface Model, digital surface model) The comparison analysis can detect the change of building height very well, but based on the acquisition method and development status of LIDAR data, it is usually difficult to have multi-temporal LIDAR data in the disaster area, and the stereo image pair also has similar problems, even if the single-temporal stereo image It can detect the collapse of buildings, but there are problems such as the small size of the stereo image pair, and the need for professional photogrammetry processing software, and the complicated work required to obtain DSM and three-dimensional buildings.

Images