Xiaoxiang Yuan

An improved SIFT algorithm in the application of close-range Stereo image matching

As unmanned aerial vehicle (UAV) remote sensing is applied in small area aerial photogrammetry surveying, disaster monitoring and emergency command, 3D urban construction and other fields, the image processing of UAV has become a hot topic in current research. The precise matching of UAV image is a key problem, which affects the subsequent processing precision directly, such as 3D reconstruction and automatic aerial triangulation, etc. At present, SIFT (Scale Invariant Feature Transform) algorithm proposed by DAVID G. LOWE as the main method is, is widely used in image matching, since its strong stability to image rotation, shift, scaling, and the change of illumination conditions. It has been successfully applied in target recognition, SFM (Structure from Motion), and many other fields. SIFT algorithm needs the colour images to be converted into grayscale images, detects extremum points under different scales and uses neighbourhood pixels to generate descriptor. As we all know that UAV images with rich colour information, the SIFT algorithm improved through combining with the image colour information in this paper, the experiments are conducted from matching efficiency and accuracy compared with the original SIFT algorithm. The results show that the method which proposed in this paper decreases on the efficiency, but is improved on the precision and provides a basis choice for matching method.

The development of Tianjin Remote Sensing Processing System Earthquake Damage Analysis

This paper introduces damage extraction and loss assessment methods and key technologies of earthquake emergency four work modes based on remote sensing, which are quickly display of post-earthquake images, rapid and rough damage assessment mode, standard damage assessment mode and detailed damage assessment mode. According to their workflows, Tianjin Remote Sensing Earthquake Damage Analysis and Processing System (TJ-RSEDAPS) was designed and developed to achieve image management, damage degree identification of buildings, key-objects and major facilities, losses assessment of life and economic, damage distribution mapping, and information service. The main functions are performed in detail.

Damage assessment of Haiti earthquake emergency using high resolution remote sensing imagery

This paper introduces the procedure of emergency remote sensing assessment for Haiti earthquake happened on Jan 12 2010. The procedure is divided into 4 steps: data preparation, data processing, information extraction and damage assessment, and contains three key targets which are damage information extraction, quantitative assessment and estimation of casualties and economic losses. In the first stage, the damage information of the buildings is the basis, and the other information, including building type, damage grade, built-over area, would be extracted by visual interpretation and automatically statistic with human-computer interaction from the high resolution disaster imageries. Then the remote sensing damage index and equivalent ground damage index of building could be counted in the second stage. According to this result, the specialists sketch a more exact intensity distribution in different regions of the metropolis. At last, the number of casualties is estimated by an empirical model adapting to worldwide earthquake as the detailed construction damage has been known. To assess the economic losses, we use a macro economic-based model which only needs population, per capita GDP and statistical macro-economic fragility related to seismic intensity. In this case, it is the first time to implement the methods of remote sensing assessment in foreign serious earthquake emergency, which is proven of being applicable outside China.

Study on a modified assessment of seismic intensity based on both multi-source RS images and spatial grid analysis

The seismic damage index (DI) has been developed to be calculated based on the damage interpretation of buildings et al from remote sensing, and further converted to the seismic intensity, but the statistical unit (usually a village or street block) for DI is pre-required. This will affect the timeliness in the emergency stage. The other case is that different capability exist in building damage extraction from multi-source RS images. So it must first convert the DIs evaluated from various images into comparable DIs. The authors in this paper propose an improved method of DI assessment of seismic damage and intensity in gridding unit with multi-source images, and then build new conversion equations according to building damage by both the extraction from aerial RS images and field survey in the disaster region of 2003 Ms7.0 Lushan earthquake. The validation check of the equivalent DIs show high consistency with the ground truth, which implicate that the method and empirical DI conversion models proposed in the paper is applicable in earthquake emergency stage.

The airborne hyperspectral image classification based on the random forest algorithm

The hyperspectral image has the advantages of wide spectral range and the high spectral resolution, and is widely applied in the terrain classification. In this paper, we study the airborne hyperspectral image classification methods using the airborne hyperspectral image. Considering the hyperspectral image has amounts of bands and there is redundancy among the bands, the principle component analysis is adopted to delete the redundant bands and the noise bands. And then the random forest decision tree algorithm is employed to classify the image. Through calculating the importance coefficient, the contribution of each band to the image classification is acquired. Finally, we design and develop the classification program using the C++ language with the OpenCV library to classify the hyperspectral image of Zhangye city collected by airborne AISA EAGLET sensor. The result of experiment indicates the validity and the robustness of random forest algorithm.

The loss assessment method of building earthquake damage using The Remote Sensing and building grid data

The earthquake disaster loss assessment is of great significance for the disaster recovery and reconstruction, the resettlement and the economic sustainable development. The Remote Sensing (RS) technique can help rapidly obtain the earthquake damage information. The paper proposes a methodology and procedure of building loss assessment based on the remote sensing and km grid-based building data, and presents an application to Lushan earthquake 7.0. The comparison to the results obtained by the average method of the administrative units implicates that the proposed method would achieve the more detailed and accurate loss assessment.

Design and realization of RS application system for earthquake emergency based on digital earth

The current RS-based earthquake emergency system is mainly based on stand-alone software which cannot meet the requirements of massive remote sensing data and parallel seismic damage information extraction after a devastating earthquake. Taking Shaanxi Province as an example, this paper explored firstly the network-based working mode of seismic damage information extraction and data management strategy for multi-user cooperative operation based on analysing work flow of the RS application to earthquake emergency. Then, using WorldWind java SDK, the RS application system for earthquake emergency based on digital earth platform was brought out in CS architecture. Finally, spatial data tables of classification and grade of seismic damage were designed and the system was developed. This system realized functions including 3D display, management of seismic RS image and GIS data obtained before and after earthquake for different user levels and cooperative extraction and publish of such seismic information as building damage, traffic damage and seismo-geological disasters caused by earthquake in real time. Some application to earthquake cases such as 2014 M s6.5 Ludian earthquake show that this system can improve the efficiency of seismic damage information interpretation and data sharing, and provide import disaster information for decision making of earthquake emergency rescue and disaster relief.

Study on RS and GIS based comprehensive analysis of multi-factors affecting the earthquake risk: An application of Monte Carlo simulation

The seismic risk, as the product of seismic hazard and vulnerability, is widely assessed based on the traditional inventory method with a detailed classified facilities and structures. The losses caused by historical earthquakes, show great variation in the same seismic intensity. Thus the study of uncertainty of building vulnerability is important to improve the accuracy of seismic risk assessment. Because of the complexity of the combined factors affecting the building vulnerability, it is hard to estimate the effects by using the simple models. The paper introduces Monte Carlo random testing method to simulate the seismic vulnerability and estimate its uncertainty in the case that complex affecting factors exist. By using the building damage data of 2008 Ms=8.0 Wenchuan earthquake in Sichuan, China, acquired through interpretation of post-earthquake RS images and field survey, the exemplary quantitative structural vulnerabilities are estimated. The results show that the vulnerability uncertainty with affecting factors can be accurately simulated.

The development of Great Earthquake Risk Assessment System based on high resolution grid data

The Great Earthquake Risk Assessment System (GERAS for Windows) is developed specially for the great earthquake risk assessment applied in the North-South Seismic Zone (NSSZ) of China. The system is based on GIS and database with gridding data in NSSZ. It has main functions of database management, the tools for data gridding, life loss estimation based on the gridding population, the economic losses estimation based on the gridding building distribution, the synthetic risk assessment considering the possibility of risks caused by both the ground seismic motion and the fault rupture, the disaster mapping and geological information processing. The system can be applied to assess risk of the potential great earthquake and fast estimate the loss immediately after the great earthquake occurrence etc. The design idea and main functions of the system is introduced in the paper.