An improved automatic detection method for earthquake-collapsed buildings from ADS40 image

Abstract

Earthquake-collapsed building identification is important in earthquake damage assessment and is evidence for mapping seismic intensity. After the May 12th Wenchuan major earthquake occurred, experts from CEODE and IPSC collaborated to make a rapid earthquake damage assessment. A crucial task was to identify collapsed buildings from ADS40 images in the earthquake region. The difficulty was to differentiate collapsed buildings from concrete bridges, dry gravels, and landslide-induced rolling stones since they had a similar gray level range in the image. Based on the IPSC method, an improved automatic identification technique was developed and tested in the study area, a portion of Beichuan County. Final results showed that the technique's accuracy was over 95%. Procedures and results of this experiment are presented in this article. Theory of this technique indicates that it could be applied to collapsed building identification caused by other disasters. earthquake, airborne ADS40 data, collapsed building, reflectance similarity, automatic detection Change detection of pre-and post-earthquake sites using high-resolution remote sensing data plays an important role in assessing damages caused by earthquakes. In particular, earthquake-collapsed buildings can be extracted through change detection techniques, which use pre-and post-earthquake high-resolution satellite images such as SPOT5 and QuickBird, by radiometrically and geometrically correcting the image sets, then matching and subtracting them [13] . Incorporation of GIS data in this process can improve extraction accuracy [46] . Threedimensional information can also be used to identify collapsed buildings. Turker extracted DEM information using airborne stereo image pairs before and after the earthquake to see the landscape deformation, and by comparing the difference of three-dimensional images collapsed buildings were identified [7] . SAR data, the coherence and intensity information of radar images such as ALOS/PALSAR, ERS-1, ERS-2 and Envisat ASAR, have also been applied in pre-and post-disaster building detection [812] . In addition, optical and microwave remote sensing information on disaster can be synthesized [13] . However, the change detection method could not meet the requirement for regions without preearthquake high-resolution images to make rapid damage assessment.Airborne remote sensing techniques have a characteristic of rapid response, and also can support multi-optical sensors and SAR devices. Almost immediately after the Wenchuan major earthquake on May 12, 2008, two B4101 Citation II/Model 550 aircrafts belonging to CEODE were made ready with a high-resolution ADS40 optical camera and a high-resolution radar system respectively. They flew over the most damaged areas of

A 3D Shape Descriptor Based on Contour Clusters for Damaged Roof Detection Using Airborne LiDAR Point Clouds

“…Because the 3D shape descriptor is normalized, the number of bins is denoted as B, which is determined using Equation (12). The frequency of each bin of the intact roof histogram is FI i , as given in Equation (13), and the frequency of each bin of the damaged roof histogram is FD i , as given in Equation (14).…”

Automated processing of high resolution airborne images for earthquake damage assessment

“…LIDAR and SAR technology) sensors are effectively exploited. Optical images can be used to extract a number of building properties, such as grey scale, spectral, texture, shape and morphological features (Rathje et al, 2005;Guo et al 2009;Dong and Shan, 2013). Among optical systems, satellite platforms are usually preferred when a large area should be surveyed.…”

Infrastructure assessment for disaster management using multi-sensor and multi-temporal remote sensing imagery

“…In recent years, several systems have been developed estimating the extent and type of destruction on various buildings. The damage assessment was realized using different kinds of sensors such as light detection and ranging (LIDAR; Rehor et al 2008), SAR data (Gamba et al 2007) and high-resolution spaceborne (Chesnel et al 2007) and airborne images (Guo et al 2009). However, most methods focus on only one single sensor and, thus, the adaptability is limited depending on the availability of data sources after a natural hazard.…”