Collaborative Innovation Center for Geospatial Technology

Wuhan, China

Collaborative Innovation Center for Geospatial Technology

Wuhan, China
Time filter
Source Type

Balz T.,Wuhan University | Balz T.,Collaborative Innovation Center for Geospatial Technology | Caspari G.,University of Hamburg | Fu B.,CAS Institute of Remote Sensing
IOP Conference Series: Earth and Environmental Science | Year: 2017

Central Asia is rich in cultural heritage generated by thousands of years of human occupation. Aiming for a better understanding of Central Asia's archaeology and how this unique heritage can be protected, the region should be studied as a whole with regard to its cultural ties with China and combined efforts should be undertaken in shielding the archaeological monuments from destruction. So far, international research campaigns have focused predominantly on single-sites or small-scale surveys, mainly due to the bureaucratic and security related issues involved in cross-border research. This is why we created the Dzungaria Landscape Project. Since 2013, we have worked on collecting remote sensing data of Xinjiang including IKONOS, WorldView-2, and TerraSAR-X data. We have developed a method for the automatic detection of larger grave mound structures in optical and SAR data. Gravemounds are typically spatially clustered and the detection of larger mound structures is a sufficient hint towards areas of high archaeological interest in a region. A meticulous remote sensing survey is the best planning tool for subsequent ground surveys and excavation. In summer 2015, we undertook a survey in the Chinese Altai in order to establish ground-truth in the Hailiutan valley. We categorized over 1000 monuments in just three weeks thanks to the previous detection and classification work using remote sensing data. Creating accurate maps of the cemeteries in northern Xinjiang is a crucial step to preserving the cultural heritage of the region since graves in remote areas are especially prone to looting. We will continue our efforts with the ultimate aim to map and monitor all large gravemounds in Dzungaria and potentially neighbouring eastern Kazakhstan. © Published under licence by IOP Publishing Ltd.

Balz T.,Wuhan University | Balz T.,Collaborative Innovation Center for Geospatial Technology | During R.,Airbus
IOP Conference Series: Earth and Environmental Science | Year: 2017

The construction of new infrastructure in largely unknown and difficult environments, as it is necessary for the construction of the New Silk Road, can lead to a decreased stability along the construction site, leading to an increase in landslide risk and deformation caused by surface motion. This generally requires a thorough pre-analysis and consecutive surveillance of the deformation patterns to ensure the stability and safety of the infrastructure projects. Interferometric SAR (InSAR) and the derived techniques of multi-baseline InSAR are very powerful tools for a large area observation of surface deformation patterns. With InSAR and deriver techniques, the topographic height and the surface motion can be estimated for large areas, making it an ideal tool for supporting the planning, construction, and safety surveillance of new infrastructure elements in remote areas. © Published under licence by IOP Publishing Ltd.

Wang H.,Wuhan University | Wang H.,Collaborative Innovation Center for Geospatial Technology | Xu Z.,Wuhan University | Xu Z.,Collaborative Innovation Center for Geospatial Technology
Knowledge-Based Systems | Year: 2017

Analyzing and mining time-series data by taking advantage of the correlation between the data values can provide outstanding beneficial. But data owners may be unwilling to publish the data's true values due to privacy considerations. Recently, researchers have begun to leverage differential privacy to address this challenge. However, the Laplace noise series used in the current state-of-the-art approaches has a drawback in that it is independent and identically distributed. An adversary can remove the independent noise from the correlated time-series by utilizing a refinement method (e.g., filtering), resulting in a lesser than expected effective degree of privacy. To remedy this problem, we propose an effective correlated time-series data publication solution based on differential privacy by enforcing Series-Indistinguishability and designing a correlated Laplace mechanism. Based on the concept of indistinguishability from the unconditional security definition, Series-Indistinguishability guarantees that the correlation between the noise and original series is indistinguishable to an adversary. Furthermore, instead of using an independent Laplace mechanism, a correlated Laplace noise series is produced using four Gauss white noise series passed through a specific linear system, to satisfy Series-Indistinguishability. Experimental results demonstrate that our solution outperforms the state-of-the-art differential privacy mechanisms in terms of security and mean absolute error for large quantities of queries. © 2017 Elsevier B.V.

Yuan X.,Wuhan University | Yuan X.,Collaborative Innovation Center for Geospatial Technology | Chen S.,Wuhan University | Yuan W.,Wuhan University | And 2 more authors.
ISPRS Journal of Photogrammetry and Remote Sensing | Year: 2017

Feature matching aims to find corresponding points to serve as tie points between images. Robust matching is still a challenging task when input images are characterized by low contrast or contain repetitive patterns, occlusions, or homogeneous textures. In this paper, a novel feature matching algorithm based on graph theory is proposed. This algorithm integrates both geometric and radiometric constraints into an edge-weighted (EW) affinity tensor. Tie points are then obtained by high-order graph matching. Four pairs of poor textural images covering forests, deserts, bare lands, and urban areas are tested. For comparison, three state-of-the-art matching techniques, namely, scale-invariant feature transform (SIFT), speeded up robust features (SURF), and features from accelerated segment test (FAST), are also used. The experimental results show that the matching recall obtained by SIFT, SURF, and FAST varies from 0 to 35% in different types of poor textures. However, through the integration of both geometry and radiometry and the EW strategy, the recall obtained by the proposed algorithm is better than 50% in all four image pairs. The better matching recall improves the number of correct matches, dispersion, and positional accuracy. © 2017 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS)

Zhu Q.,Southwest Jiaotong University | Zhu Q.,Collaborative Innovation Center for Geospatial Technology | Li Y.,Hong Kong Polytechnic University | Hu H.,Southwest Jiaotong University | And 2 more authors.
ISPRS Journal of Photogrammetry and Remote Sensing | Year: 2017

The semantic classification of point clouds is a fundamental part of three-dimensional urban reconstruction. For datasets with high spatial resolution but significantly more noises, a general trend is to exploit more contexture information to surmount the decrease of discrimination of features for classification. However, previous works on adoption of contexture information are either too restrictive or only in a small region and in this paper, we propose a point cloud classification method based on multi-level semantic relationships, including point–homogeneity, supervoxel–adjacency and class–knowledge constraints, which is more versatile and incrementally propagate the classification cues from individual points to the object level and formulate them as a graphical model. The point–homogeneity constraint clusters points with similar geometric and radiometric properties into regular-shaped supervoxels that correspond to the vertices in the graphical model. The supervoxel–adjacency constraint contributes to the pairwise interactions by providing explicit adjacent relationships between supervoxels. The class–knowledge constraint operates at the object level based on semantic rules, guaranteeing the classification correctness of supervoxel clusters at that level. International Society of Photogrammetry and Remote Sensing (ISPRS) benchmark tests have shown that the proposed method achieves state-of-the-art performance with an average per-area completeness and correctness of 93.88% and 95.78%, respectively. The evaluation of classification of photogrammetric point clouds and DSM generated from aerial imagery confirms the method's reliability in several challenging urban scenes. © 2017 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS)

Mao F.,Wuhan University | Wang W.,Wuhan University | Min Q.,Wuhan University | Min Q.,Albany Research Center | And 2 more authors.
Optics Express | Year: 2015

Fernald method is regarded as the standard method for retrieving lidar data, but the retrieval can be performed only when a boundary value is given. Generally, we can select clear atmosphere above the tropopause as a reference to determine the boundary value, but we need to use the slope method to fit the boundary value when the detecting range is lower than the tropopause. The slope method involves significant uncertainty because this algorithm is based on two hypotheses: one is that aerosol vertical distribution is homogeneous, and the other is that either molecule or aerosol components exist in the atmosphere. To reduce the uncertainty, we proposed a new approach, which segments a signal into "uniform" subsignals to avoid the first hypothesis, and then uses nonlinear twocomponent fitting to avoid the second one. Compared with the approach based on the slope method, the new approach obtained more accurate boundary values and retrieving results for both of the simulated and real signals. Thus the automatic segmentation algorithm and the two-component fitting method are useful for determining the reference bin and boundary values when the effective detecting range of lidar is lower than the tropopause. © 2015 Optical Society of America.

Chen J.,Wuhan University | Sang J.,Wuhan University | Sang J.,Collaborative Innovation Center for Geospatial Technology
Geodesy and Geodynamics | Year: 2016

Atmospheric drag, which can be inferred from orbit information of low-Earth orbiting (LEO) satellites, provides a direct means of measuring mass density. The temporal resolution of derived mass density could be in the range from minutes to days, depending on the precision of the satellite orbit data. This paper presents two methods potentially being able to estimate thermosphere mass density from precise obit ephemeris with high temporal resolution. One method is based on the drag perturbation equation of the semi-major axis and the temporal resolution of retrieved density could be 150 s for CHAMP satellite. Another method generates corrections to densities computed from a baseline density model through a Kalman filter orbit drag coefficient determination (KFOD) process and the temporal resolution of derived density could be as high as 30 s for CHAMP satellite. The densities estimated from these two methods are compared with densities obtained from accelerometer data of CHAMP satellite. When the accelerometer data based densities are used as reference values, the mean relative accuracy of the densities derived from precision orbit data using the two methods is within approximately 10%. An application of the derived densities shows that the derived densities can reduce orbit predication errors. © 2016 Institute of Seismology, China Earthquake Administration

Yao Y.,Wuhan University | Yao Y.,Collaborative Innovation Center for Geospatial Technology | Zhao Q.,Wuhan University
Meteorology and Atmospheric Physics | Year: 2016

Water vapor information with highly spatial and temporal resolution can be acquired using Global Navigation Satellite System (GNSS) water vapor tomography technique. Usually, the targeted tomographic area is discretized into a number of voxels and the water vapor distribution can be reconstructed using a large number of GNSS signals which penetrate the entire tomographic area. Due to the influence of geographic distribution of receivers and geometric location of satellite constellation, many voxels located at the bottom and the side of research area are not crossed by signals, which would undermine the quality of tomographic result. To alleviate this problem, a novel, optimized approach of voxel division is here proposed which increases the number of voxels crossed by signals. On the vertical axis, a 3D water vapor profile is utilized, which is derived from radiosonde data for many years, to identify the maximum height of tomography space. On the horizontal axis, the total number of voxel crossed by signal is enhanced, based on the concept of non-uniform symmetrical division of horizontal voxels. In this study, tomographic experiments are implemented using GPS data from Hong Kong Satellite Positioning Reference Station Network, and tomographic result is compared with water vapor derived from radiosonde and European Center for Medium-Range Weather Forecasting (ECMWF). The result shows that the Integrated Water Vapour (IWV), RMS, and error distribution of the proposed approach are better than that of traditional method. © 2016 Springer-Verlag Wien

Wang L.,Hubei Engineering University | Wang L.,Hubei University | Gong W.,Hubei Engineering University | Gong W.,Collaborative Innovation Center for Geospatial Technology | And 4 more authors.
Atmospheric Environment | Year: 2015

Aerosol optical properties including aerosol optical depth (AOD), Ångström exponent (α), single scattering albedo (SSA), aerosol size distribution and refractive index at urban Wuhan in Central China are investigated based on the measurements from a CIMEL sun-photometer during 2007-2013. AOD500nm is found to be relatively high all year round and the highest value 1.52 occurs in June 2012 and the lowest (0.57) in November 2012. α shows a significant monthly variation, with the highest value in June 2010 (1.71) and the lowest value (0.78) in April 2012. Analysis of AOD and α frequencies indicate that this region is populated with fine-mode particles. Monthly variations of SSA for total, fine and coarse-mode particles are closely related to the aerosol hygroscopic growth, fossil fuel and biomass burning. The aerosol volume size distributions (bi-modal pattern) show distinct differences in particle radius for different seasons, the radius for fine-mode particles generally increase from spring to summer month, for example, the highest peak is around radius 0.15μm in March, while the peak radius is around 0.25μm in June. Finally, monthly statistics of real and imaginary parts of the complex refractive index are analyzed, the highest averages of real (1.50) and imaginary parts (0.0395) are found in spring and autumn, respectively at wavelength 440-1020nm. © 2014 Elsevier Ltd.

Guo F.,Wuhan University | Guo F.,Key Laboratory of Geophysical Geodesy | Guo F.,Collaborative Innovation Center for Geospatial Technology | Zhang X.,Wuhan University | And 4 more authors.
Journal of Geodesy | Year: 2016

The latest generation of GNSS satellites such as GPS BLOCK-IIF, Galileo and BDS are transmitting signals on three or more frequencies, thus having more choices in practice. At the same time, new challenges arise for integrating the new signals. This paper contributes to the modeling and assessment of triple-frequency PPP with BDS data. First, three triple-frequency PPP models are developed. The observation model and stochastic model are designed and extended to accommodate the third frequency. In particular, new biases such as differential code biases and inter-frequency biases as well as the parameterizations are addressed. Then, the relationships between different PPP models are discussed. To verify the triple-frequency PPP models, PPP tests with real triple-frequency data were performed in both static and kinematic scenarios. Results show that the three triple-frequency PPP models agree well with each other. Additional frequency has a marginal effect on the positioning accuracy in static PPP tests. However, the benefits of third frequency are significant in situations of where there is poor tracking and contaminated observations on frequencies B1 and B2 in kinematic PPP tests. © 2016 Springer-Verlag Berlin Heidelberg

Loading Collaborative Innovation Center for Geospatial Technology collaborators
Loading Collaborative Innovation Center for Geospatial Technology collaborators