Urban areas in Lima, Peru, grow fast and with almost no control. Furthermore, resources to monitor urban areas are limited in Peru. Thus, the evolution of exposure of urban areas to natural hazards is highly uncertain. In this paper, we report a practical use of deep learning-based land use and land cover mapping to quantify, with unprecedented accuracy, the increment in exposure against floods in the District of Ate, Lima, Peru. We use Google Satellite imagery recorded in 2002 and 2023, generate land use maps, and quantify the increment of urban areas in the floodplain of the Rimac River. Results show an increment of exposure to floods of about 10.3% after 21 years.

In this paper, the Huangtupo landslide in the Three Gorges Reservoir (TGR) was explored by the multi-source satellite SAR data from ALOS-2 and Sentinel-1A/B based on time-series InSAR analysis. One track of ALOS-2 and two tracks of Sentinel-1 SAR datasets acquired from August 2016 to October 2017 were used to investigate the surface deformations. The maximum line-of-sight (LOS) deformation rate measured over the slope surface was up to -50 mm/year from the ALOS-2 measurement, while it was as high as -40 and -30 mm/year for Sentinel-1 Track 11 and Track 84 measurements, respectively. Cross-validation between the multi-source InSAR results showed that the ALOS-2 and Sentinel-1A/B measurements are highly correlated. Time series displacement analysis of selected point clusters based on InSAR measurements from ALOS-2 and Sentinel-1 data stacks was also investigated along with the TGR water level and daily rainfall over this area. Seasonal fluctuations caused by changes in rainfall and TGR water level can be clearly observed from the time series evolutions of deformation. It is clearly found that the lower parts of the Huangtupo landslide are more likely to influenced by the TGR water level, while the upper parts mainly responses to the seasonal changes of rainfall.

The recent surge in floods requires the development of a digital terrain model (DTM) with a high spatial resolution, as this type of model is essential for risk assessment at the building level. To generate a DTM, it is necessary to remove nonground objects from the corresponding digital surface model (DSM) and interpolate the elevation of the removed area. However, automatically conducting this process requires input data other than the DSM and/or setting parameters that are suitable for the target region. Here, to directly generate a DTM from only a DSM, we used the pix2pix model that performs domain conversion using conditional generative adversarial networks (cGANs). Pix2pix and the generator-only model were tested in areas with different topography and building characteristics. For both models, the root mean square errors (RMSEs) of the generated DTMs were approximately 0.4 m for areas where small buildings were distributed on flat terrain. In areas with large buildings with flat roofs or rapid elevation changes, the RMSEs were 1 m or even larger. In areas with rapid elevation changes, pix2pix showed an RMSE that was 1 m smaller than that of the generator-only model, suggesting the effectiveness of cGANs.

After a major disaster, a rapid assessment of building damage is highly required for emergency response and prompt recovery. Remote sensing technologies have been widely applied for building damage mapping. Combining machine-learning algorithms (e.g., deep learning) and satellite images has recently demonstrated success in boosting damage recognition methods. Although previous techniques have shown great success, they primarily adopt supervised settings, often requiring a minimum number of training samples to achieve acceptable accuracy. Moreover, previous methods also are developed for specific target areas, which makes it challenging to apply them to other regions in case of future disasters. This paper presents a novel unsupervised approach for building damage mapping, focusing on collapsed structures, using modern convolutional neural network (CNN) models and high-resolution remote sensing imagery. We apply our mapping framework to revise the building damage following the 2007 Peru-Pisco Earthquake and the recent 2023 Turkey and Syria Earthquakes.

Recent works have explored the automatic extraction of road marking features from sparse point clouds using convolutional neural networks (CNN) in an attempt to support the use of low-cost light detection and ranging (LiDAR) sensors for reducing the cost of mobile mapping tasks. In this work we follow through with this ideology and propose D-Fast-SCNN, a modification of the Fast-SCNN model, to improve its classification accuracy while still leveraging its speed. We propose the addition of a dilation block, which utilizes the morphological process of dilation. This additional process takes full advantage of the observed property of sparse point cloud-derived imagery, wherein misclassifications can be omitted in the areas with no corresponding point cloud value. After multiple trials, we can see an increase of around 10% in the mean f1-score and a decrease of roughly 6% in uncertainty for the target road marking class. In future works an ablation study will be conducted to investigate its effects when employed elsewhere in the model.

Estimating surface normals from 3D point clouds is critical for various applications, including surface reconstruction and rendering. While existing methods for normal estimation perform well in regions where normals change slowly, they tend to fail where normals vary rapidly. To address this issue, we propose a novel approach called MSECNet, which improves estimation in normal varying regions by treating normal variation modeling as an edge detection problem. MSECNet consists of a backbone network and a multi-scale edge conditioning (MSEC) stream. The MSEC stream achieves robust edge detection through multi-scale feature fusion and adaptive edge detection. The detected edges are then combined with the output of the backbone network using the edge conditioning module to produce edge-aware representations. Extensive experiments show that MSECNet outperforms existing methods on both synthetic (PCPNet) and real-world (SceneNN) datasets while running significantly faster. We also conduct various analyses to investigate the contribution of each component in the MSEC stream. Finally, we demonstrate the effectiveness of our approach in surface reconstruction.

3D point clouds are discrete samples of continuous surfaces which can be used for various applications. However, the lack of true connectivity information, i.e., edge information, makes point cloud recognition challenging. Recent edge-aware methods incorporate edge modeling into network designs to better describe local structures. Although these methods show that incorporating edge information is beneficial, how edge information helps remains unclear, making it difficult for users to analyze its usefulness. To shed light on this issue, in this study, we propose a new algorithm called Diffusion Unit (DU) that handles edge information in a principled and interpretable manner while providing decent improvement. First, we theoretically show that DU learns to perform task-beneficial edge enhancement and suppression. Second, we experimentally observe and verify the edge enhancement and suppression behavior. Third, we empirically demonstrate that this behavior contributes to performance improvement. Extensive experiments and analyses performed on challenging benchmarks verify the effectiveness of DU. Specifically, our method achieves state-of-the-art performance in object part segmentation using ShapeNet part and scene segmentation using S3DIS. Our source code is available at https://github.com/martianxiu/DiffusionUnit.

Digital representation of road markings is an essential component in High-Definition (HD) maps to support the navigation of autonomous vehicles. As such, works on automatically classifying road markings from mobile sensor data have increased. One of which is the use of U-NET, a convolutional neural network, on dense mobile LIDAR point cloud-derived images. In this paper, we will explore road marking classification on sparse mobile LIDAR point cloud-derived images as well as using an FSCNN model trained with focal loss, as an alternative to U-NET to drastically improve processing speeds without compromising classification accuracy. Tests done for this work have shown a maximum improvement in classification of 77% in the F1-score. More so, processing speeds reached 6x and 15x faster for training and prediction speeds.

Three-dimensional (3D) point clouds are becoming an important part of the geospatial domain. During research on 3D point clouds, deep-learning models have been widely used for the classification and segmentation of 3D point clouds observed by airborne LiDAR. However, most previous studies used discriminative models, whereas few studies used generative models. Specifically, one unsolved problem is the synthesis of large-scale 3D point clouds, such as those observed in outdoor scenes, because of the 3D point clouds' complex geometric structure. In this paper, we propose a generative model for generating large-scale 3D point clouds observed from airborne LiDAR. Generally, because the training process of the famous generative model called generative adversarial network (GAN) is unstable, we combine a variational autoen-coder and GAN to generate a suitable 3D point cloud. We experimentally demonstrate that our framework can generate high-density 3D point clouds by using data from the 2018 IEEE GRSS Data Fusion Contest.

During the last few years, in the field of computer vision, sophisticated deep learning methods have been developed to accomplish semantic segmentation tasks of 3D point cloud data. Additionally, many researchers have extended the applicability of these methods, such as PointNet or PointNet++, beyond semantic segmentation tasks of indoor scene data to large-scale outdoor scene data observed using airborne laser scanning systems equipped with light detection and ranging (LiDAR) technology. Most extant studies have only investigated geometric information (x, y, and z or longitude, latitude, and height) and have omitted rich radiometric information. Therefore, we aim to extend the applicability of deep learning-based model from the geometric data into radiometric data acquired with airborne full-waveform LiDAR without converting the waveform into 2D images or 3D voxels. We simultaneously train two models: a local module for local feature extraction and a global module for acquiring wide receptive fields for the waveform. Furthermore, our proposed model is based on waveform-aware convolutional techniques. We evaluate the effectiveness of the proposed method using benchmark large-scale outdoor scene data. By integrating the two outputs from the local module and the global module, our proposed model had achieved higher mean recall value 0.92 than previous methods and higher F1 scores for all six classes than the other 3D Deep Learning method. Therefore, our proposed network consisting of the local and global module successfully resolves the semantic segmentation task of full-waveform LiDAR data without requiring expert knowledge.

Disasters such as earthquakes, hurricanes, and flooding are responsible for large-scale infrastructure damages and loss of human lives. Immediately after disaster strikes, one of the most critical and difficult tasks is accurately assessing the extent and severity of the disaster. This task is especially challenging in areas isolated by the disaster; in such cases, remote sensing information provides the best alternative to tackle this problem. This paper presents a damage mapping framework using remote sensing imagery acquired from previous disasters. The proposed deep learning-based framework is trained to learn features related to building damage using imagery from previous disasters that were collected from different regions around the world. Then, it is tested to recognize damage from a different urban environment.

With some complex surface structure, it is difficult to understand the surface displacements by using SAR data. In addition to the complex geometry, identifying and understanding the surface displacements becomes quite challenging when non-linear displacements exist. There are several types of the interferometric stacking approaches, and each approach has different characteristics to estimate the displacements. In this paper, the conventional PSI, NN-PSI, and SBAS techniques are applied to a PALSAR-1 dataset, and the resulting displacements are compared. By using the three approaches, it is possible to understand the surface displacements in the study area, and the strength and drawback in each techniques are recognized. It is also recognized that the spatial baseline distribution should influence the displacement estimation of PSI approaches according to the EV-spectrum.

This letter improves the computational efficiency and proposes a methodology to retrieve the missing feature pairs when utilizing binary-based features for image matching. The 64-byte feature descriptors of binary robust invariant scalable keypoints (BRISKs) are rearranged by combining human retina ganglion cells distribution and visual accommodation to speed up the image matching. In addition, an interactive two-sided matching is designed to determine the most probable keypoint pair when a feature point in the reference image is mapped to multiple candidates in the target image. Experimental results indicate that the proposed inverse sorting ring can reduce the processing time by more than 10% compared to the accelerated BRISK while maintaining the same reliability. Also, additional point-to-point feature pairs can be regained from the point-to-multicandidate cases by the proposed method in order to increase the number of matches.

The building change detection, which is important for monitoring human activity in urban and sub-urban area, can be done by the using of remote sensing data. To overcome the cloud cover problem that cause the limited utilizing in optical images which is repeatedly occurred in tropical areas, we decided to use synthetic aperture radar (SAR) data, a kind of remote sensing data that does not get affected by weather condition, to fulfill this change detection purpose. In this study, Bangkok, Thailand were selected as our study area in order to demonstrate that the SAR data can be used in the area that optical data cannot be used such as tropical areas like in Thailand. As our target is to detect changes of buildings, so we have to consider separating the change that cause by seasonal and other non-target objects. To complete this purpose, we decided to use one of the deep learning techniques called U-net which is created for the image segmentation task. We have created the ground truth and used one portion for training the network and another portion for validate the result. Finally, the model that trained by our training data was able to provide promising results.

The distribution of ground displacement due to the 2016 Kumamoto earthquake was obtained from the analysis of the synthetic aperture radar (SAR) satellite image and the relationship with damage to the low-pressure gas distribution pipes was investigated. First, we estimated the co-seismic ground displacement with an accuracy of approximately 0.04 m by differential interferometric SAR (D-InSAR) analysis of the PALSAR-2 imagery on the ALOS-2 satellite which observed the affected area before and after the earthquake. Then, we calculated the displacement of the quasi-east-west and the quasi-vertical directions from the combination of two line-of-sight (LOS) displacements estimated from two pair images with different observation conditions and examined the relation with damage ratios of gas distribution pipes. As the results, a correlation between the gradient of the ground displacement and the damage ratio was found and the damage ratio tended to increase as the gradient became larger.

The purpose of this study is to monitor the surface displacement on the large scale of the infrastructures such as highways and railways with the interferometric stacking technology (SBAS). Both ascending and descending acquisitions are used to decompose the multi-temporal displacement into east-west and vertical components. The RMSE value between the resulting vertical displacement and the vertical measurement of the permanent GPS station located in the study area was 5.1 mm. Also the obtained displacement shows the subsidence phenomena on National Route 464 due to the extended constructions along the route and railway.

Deep learning methods that directly handle three-dimensional point clouds, e.g., PointNet, have recently been proposed. Moreover, deep learning-based-techniques have demonstrated excellent performance for supervised learning tasks on point clouds such as classification and segmentation for certain open datasets. In this study, the possibility of using a deep learning method, which is an unsupervised representation learning method to extract features from airborne raw full-waveform data, is investigated. Thus, a novel end-to-end autoencoder network called FWNetAE is proposed to address representation learning challenges on raw full-waveform data. On the encoder side, a PointNet-based network extracts the latent vector of raw full-waveform data. Subsequently, a fully connected network-based decoder deforms the latent vector into the input full-waveform light detection and ranging (LiDAR) data, thus achieving fewer reconstruction errors.

Semantic segmentation of 3D point cloud data is a crucial step for various applications such as autonomous driving. Conventional methods for semantic segmentation of point cloud heavily rely on numerous hand-crafted features often derived from the 3D covariance matrix, which are typically time-consuming. In this work, a dynamic scale graph convolutional neural network is proposed to perform semantic segmentation on 3D point cloud without relying on extensive exploitation of hand-crafted features. Using only spatial coordinates, backscattered intensity and spectral information derived from aerial images, our proposed method aims at modeling multi-scale local structural information by combining dynamic scale sampling and multi-scale neighbor graphs. Sampling and neighbor graph construction are both implemented on the fly such that no preprocessing or data augmentation is needed before training. Evaluated using the ISPRS 3D Semantic Labeling Contest, our approach has achieved the performance on par with the current state-of-the-art in terms of average f1-score without hand-crafted features and less computational overhead.

On September 19, 2017, the Mw7.1 Puebla Earthquake caused significant destruction in several cities in central Mexico. In this paper, two pre- and one post-event ALOS2-PALSAR2 data were used to detect the damaged area around Izucar de Matamoros town in Mexico. First, we identify the built-up areas using pre-event data. Second, we evaluate the earthquake-induced damage areas using an RGB color-coded image constructed from the pre- and co-event coherence images. Our analysis showed that the green and red bands display a great potential to discriminate the damaged areas.

Ground analysis is important after an earthquake occurred in order to understand the surface variations. By optical imagery before and after a disaster, this paper tries to find the co-seismic ground displacement using region-based image patches. By comparing the similarity of pre- and post-event orthophotos, regions of ground changes can be detected. Conjugate objects were then recognized through feature extraction and image matching. An image with pose parameters can estimate the three-dimensional positions of those objects in the world space. Thus, three-dimensional geometric correlation between the pre- and post-event objects can measure the ground displacement. The results indicate that the proposed approach is able to estimate the co-seismic ground displacements based on the spatial information from optical imagery. In addition, surface shifts can be identified based on the spatial correlations.

In this paper, we propose a method for the translation from Synthetic Aperture Radar (SAR) to optical images using conditional Generative Adversarial Networks (cGANs). Satellite images have been widely utilized for various purposes, such as natural environment monitoring (pollution, forest or rivers), transportation improvement and prompt emergency response to disasters. However, the obscurity caused by clouds leads to unstable monitoring of the ground situation while using the optical camera. Images captured by a longer wavelength are introduced to reduce the effects of clouds. In particular, SAR images are known to be nearly unaffected by clouds and are often used for stably observing the ground situation. On the other hand, SAR images have lower spatial resolution and visibility than optical images. Therefore, we propose a deep neural network that generates optical images from SAR images. Finally, we confirm the feasibility of the proposed network on a dataset consisting of optical images and the corresponding SAR images.

On the 11thof August 2012 two earthquakes within 11 minutes struck NW Iran with considerable fatalities and damages. Four Synthetic Aperture Radar (SAR) images from Wide Multi-look Fine mode of RADARSAT-2 data have been considered for rapid damage assessment through SAR interferometry processing. All images are from ascending orbits with 8 m spatial resolution. The first pair (2012.04.13 and 2012.09.08) covers the whole study area while the second one (2012.03.13 and 2012.08.28) partially encompasses the study area. Since the study area comprises 38 rural sites, the block-based building damage assessment method has been pursued instead of individual parcel assessment. The aggregated pixel method for the coherence map shows that the accuracy of initial damage assessment (and Kappa coefficient) is 60% (0.51) which profitably can be used in proper disaster management and any “where to go first” operation.

In this study, a practical methodology was presented to map damaged buildings using high resolution synthetic aperture radar (SAR) images and post-event building damage data from the 2013 Super Typhoon Haiyan, in Tacloban city, the Philippines. To detect destroyed structures, we focused on the changes in the radar signal within footprints of buildings between pre- and post-event SAR images. The method was tested using a 1.0 m resolution COSMO-SkyMed SAR images taken over Tacloban city, the Philippines. The method proves, with 73% accuracy in this case, to be suitable for estimating destroyed buildings.

There are several new SAR satellites launched in the last years, and it becomes quite important to understand the measurement capability of the new sensors. The Japan Aerospace Exploration Agency (JAXA) launched ALOS-2 PALSAR-2 satellite in 2014, and the measurement of surface deformation with this new sensor is expected to be more accurate than the previous sensors such as JERS-1 and ALOS PALSAR-1. In this paper, we used PALSAR-2 data in order to measure displacement induced by the 2014 North Nagano earthquake (M6.7) occurred on November 22nd in Japan. A standard differential interferometry process (DInSAR) was carried out with PALSAR-2 data, and the results were validated with the permanent GPS data and the field survey data measured by disaster response activities for this earthquake. Although the validation is not quantitative analysis, the direction and amount of the deformation derived from DInSAR correspond to the ground truth data, and we conclude that PALSAR-2 should be essential data source to measure surface deformation with wide area.

In this letter, a new approach is proposed to classify tsunami-induced building damage into multiple classes using pre- and post-event high-resolution radar (TerraSAR-X) data. Buildings affected by the 2011 Tohoku earthquake and tsunami were the focus in developing this method. In synthetic aperture radar (SAR) data, buildings exhibit high backscattering caused by double-bounce reflection and layover. However, if the buildings are completely washed away or structurally destroyed by the tsunami, then this high backscattering might be reduced, and the post-event SAR data will show a lower sigma nought value than the pre-event SAR data. To exploit these relationships, a rapid method for classifying tsunami-induced building damage into multiple classes was developed by analyzing the statistical relationship between the change ratios in areas with high backscattering and in areas with building damage. The method was developed for the affected city of Sendai, Japan, based on the decision tree application of a machine learning algorithm. The results provided an overall accuracy of 67.4% and a kappa statistic of 0.47. To validate its transferability, the method was applied to the town of Watari, and an overall accuracy of 58.7% and a kappa statistic of 0.38 were obtained.

In this study, the extent of the flooded areas by the Super Typhoon Haiyan in the Philippines were extracted using ASTER VNIR images taken over Tacloban city in the Visayas. In order to constraint the affected area, we employed the normalize difference vegetation and water indices (NDVI and NDWI) from the pre- and post-event images. The extension of the flooded area was determined by comparing the index characteristics before and after the event. A phase-based change detection method indices was applied to classify the affected area into three classes according to the changes between the pre- and post-images. Through NDWI the flooded areas were detected despite the moderate resolution of ASTER images. In addition, the phase-based analysis successfully detected level of change within the affected area that may be correlated to the damage observed on field surveys. The results from the phase-based analysis were verified with damage levels obtained through visual damage inspection using high resolution satellite images.

In this study, a semi-automated method to estimate building damage in a tsunami affected area is developed using pre- and post-event high-resolution synthetic aperture radar (TerraSAR-X) data. For development, some coastal areas affected by the 2011 Tohoku earthquake tsunami were focused. The method for estimating building damage consists of three steps, 1) To detect flooded areas by the tsunami, 2) To detect built-up areas, 3) To estimate building damage inside the flooded built-up areas. The previously proposed methods using high-resolution SAR data needs building footprint data for estimating building damage[1]. However, this problem was improved by developing a new method which does not need building footprint data to estimate building damage caused by the tsunami. The developed method was validated on the other test sites and the estimated results showed good consistency with the ground truth data.

A strong typhoon “Haiyan” affected Southeast Asia on November 8, 2013, caused gigantic destruction in the Philippines. In this study, two pre- and one post-event COSMO-SkyMed SCSB data were used to detect the damaged area around Tacloban City, Leyte Island. First, the severe damaged areas were detected according to the difference between the pre- and post-event speckle divergence values. Then the pre- and co-event coherence (NDCI) and correlation coefficient (NDCOI) were calculated from the three temporal data. The relationships between the four building damage levels and NDCI or NDCOI value were obtained by introducing the visual interoperation result. Using this relationship, the possibility of each damage class was estimated in the whole urban area.

In order to grasp the quantitative and spatial information of the earth surface deformation due to earthquakes, well-georectified remotely sensed data with high pixel resolution taken before and after the event are useful. Among satellite optical sensors, Terra and PRISM loaded on Terra and ALOS, respectively, have the function of along-track stereoscopic shooting to generate a digital elevation model (DEM), then an orthorectified image based on derived DEM is also created by a single fight observation. This paper performs to apply the subpixel-based image processing using pre- and post-earthquake orthorectified images to map the displacement distribution around the fault rupture region of the 2010 El Mayor (Baja) earthquake.

The present paper reports a preliminary result of searching tsunami-affected area using recent advances of GIS analysis and remote sensing combined with a numerical modeling of tsunami propagation/inundation and world population database. Applying the method of searching tsunami affected area to the 2009 Samoa earthquake tsunami and the 2010 Chilean earthquake tsunami, the potential tsunami affected area have been detected at some coastal cities/communities. The results are utilized to detecting tsunami impacted area for conducting disaster relief activities.

In order to expand the existing C-band SAR based damage estimation model into L-band SAR, this paper introduces a likelihood function to estimate severe damage ratio by earthquakes on the basis of dataset from JERS-1/SAR (L-band SAR) images observed the 1995 Kobe earthquake and its detailed ground truth data. The model is applied to JERS-1/SAR images taken over the tsunami affected areas by the 1993 Hokkaido Nansei-oki, Japan earthquake.

GEO Grid has been proposed by AIST in order to contribute to earth science. GEO Grid mainly provides satellite and field observation data related to earth science through data search service, data processing service and data providing service. Recently, we have developed ALOS PALSAR data on-demand processing and providing system as one of GEO Grid data providing system. The system allows users to easily search and quickly receive PALSAR products without careful considerations and advanced skills. There are two important points in the system. One is seamless connection between AIST and an external archive system. The other is that the system can provided calibrated PALSAR products according to observations using Corner Reflectors. As a future plan, OGC-CSW is applied to this system for data search service.

An earthquake occurred in the coast of Algeria on May 21, 2003. The cities of Boumerdes and Zemmouri were the most extensively damaged areas. Canadian SAR satellite, RADARSAT, observed Boumerdes area by the fine-beam mode, on 4 days after the event. European SAR satellite, ERS, also observed the same area on June 7, 2003. On December 26, 2003, another strong earthquake occurred beneath the city of Bam, Iran. Severely damaged areas were found widely being distributed in the city from high-resolution optical satellite images obtained after the event. ENVISAT also captured the hard-hit areas on January 7, 2004. In this paper, we investigated the characteristics of damaged areas in these SAR images by visual interpretation and clarified the effect of spatial resolution for the detection of damaged buildings. Then, we applied our automated damages detection technique, which was developed based on the data set of the 1995 Kobe earthquake, to the SAR images of Algeria and Iran

In this paper a new transformation which enables us to detect boundaries of cast shadows in high resolution satellite images is introduced. The transformation is based on color invariant indices. Different radiometric restoration techniques such as gamma correction, linear-correlation correction and histogram matching are introduced in order to restore the brightness of detected shadow area

An automatic method for LIDAR-based (Light Detection And Ranging) change detection is proposed. Highly dense LIDAR point clouds are recommended as the most suitable gathered data for dense urban areas. The main goal is to develop an up-to-date building inventory database, which is in great demand for the earthquake-prone areas like Japan, using LIDAR as primary data. Two LIDAR surveying flights in 1999 and 2004 provide the test data over Roppongi, Tokyo, Japan. Detected results are visual evaluation using orthophoto produced by LIDAR surveying flights. The highly automated processing proved the efficiency of using LIDAR for a quick and reliable updating. Moreover, it also implies the feasibility for detection of damaged buildings due to earthquake.

QuickBird observed the city of Zemmouri, Algeria, before and after the May 21, 2003 Algeria earthquake. Using the pre-event and post-event pan-sharpened images, visual inspection of building damage was carried out by the five authors of this paper individually. A total 1,399 buildings were classified into five damage levels of European Micro-seismic Scale. The results from the different interpreters were reasonably close for collapsed buildings but the difference becomes larger for smaller damage levels. The locations of refugee tents in the two post-event images were also identified. These observations indicate that high-resolution satellite images can provide quite useful information to emergency management after natural disasters.

A degree of earthquake damage can be estimated using temporal decorrelation by employing a coherence ratio which is defined by dividing a post-event coherence image by a pre-event image. In the case of applying both C and L bands SAR data for evaluating the damage of the 1995 Hyogoken-Nanbu Earthquake in Japan, the probability of the degree of damage could be approximated by a linear function of the coherence ratio. In this paper, we examine whether the post-earthquake damage estimation model is applicable for the 1999 Kocaeli earthquake in Turkey as another case. As a result, significant correlation between the probability of the degree of damage and the grade of damage surveyed by disaster researchers is also clarified by employing the coherence ratio computed from three ERS- 1/2 SAR data set including the event.

One of the remarkable characteristics of synthetic aperture radar (SAR) is to record physical value called the backscattering coefficient of the Earth's surface not depending on weather conditions and Sun illumination. Therefore, SAR could be a powerful tool and be used to develop a universal method for grasping damaged areas by disasters such as earthquakes, forest fires and floods. Detailed ground truth data for building damage due to the 1995 Kobe earthquake provided us the opportunity to investigate the relationship between the backscattering property from SAR images and the degree of damage. From the above analysis we have already developed a method to detect areas of building damage. In this paper, we applied this method to the images taken over the area hit by the 1999 Kocaeli, Turkey and the 2001 Gujarat, India earthquakes, and then the accuracy of the proposed method was examined by comparing the results of the analyses with those from the damage surveys.

In multi-stage remote sensing performed by NASDA just after the 1995 Hyogoken-Nanbu (Kobe) earthquake, stricken areas were observed by the airborne MSS, which has twelve bands between visible and thermal infrared. In this study, after some training areas were selected in Nada Ward using GIS data based on a field damage survey, spectral characteristics of damaged and non-damaged buildings were investigated. Then, the distribution corresponding to the damage level of buildings in the hard-hit area of Kobe City was estimated by the maximum likelihood classifier. The estimated result of areas with burned and severely damaged buildings was relatively in good agreement with the field survey data. An application of this method, based solely on the post-event image, to early damage assessment systems can be expected.