2020 Vol. 45, No. 2
The Sheshan 13 m radio telescope is designed as a new generation of astrometric and geodetic very long baseline interferometry (VLBI) observation system, equipped with 2-14 GHz broadband receiver and X/Ka dual-band receiver. Pointing precision is one of the most important technical specifications of radio telescopes, typically about one tenth of the beam width at the highest observation frequency. For the Sheshan 13 m radio telescope, the required pointing precision is about 18 as at 32 GHz of Ka band. Based on the pointing scan data of the telescope, the method of establishing pointing correction model is discussed, including the curve fitting of the power measurements of extragalactic sources, the tests on the effect of the integration time, and the parameter setting of the pointing correction model. The pointing precision of the telescope is measured and estimated. The resulted pointing correction model can be used as the basis for the system debugging and improvement. It is also the guarantee for the measurement of the systematic technical specifications and precisely tracking of observation target. The data analysis model and analysis process can be used for the daily pointing precision check of the observation system, and can also be used for similar engineering measurement as reference.
Due to the interference of various uncertain factors, the abnormal disturbances often occur in the satellite clock offset data, which reduces the reliability of the performance analysis of the satellite clock, destroys the validity of the modeling and prediction of clock offset, and affects the accuracy of the navigation positioning results. As to this problem, on the basis of the autoregressive integrated move average (ARIMA) model, this paper establishes an outlier detection model of clock offset time series. Based on the principle of Bayes statistics, the problems of outliers detection and the outliers magnitudes estimation are transformed into a model selection problem. Through the approximate calculation of the posterior probability of the model, the measurement standard of the model selection is derived so the complex iterative computation is avoided. Simulation Test examples of GPS and BeiDou illustrate that the proposed method can detect the outliers effectively and estimate the magnitudes of outliers accurately in the clock offset sequence; furthermore, it can obtain higher prediction precision when the method is applied in the medium and long term prediction of the satellite clock offset.
The West Liaohe River Basin (WLRB) is one of the most sensitive areas to climate change in China. Here we use the terrestrial water fluxes derived from the Gravity Recovery and Climate Experiment (GRACE) satellites, precipitation dataset, and in situ runoff records, to calculate the actual evapotranspiration (ET) in the WLRB based on the water balance equation. We find that during the drought period of 2005-2011, the mean annual ET is 350.5 mm, which is 9.8 mm more than the mean annual precipitation during the same time period. This difference can be explained by the groundwater depletion in the WLRB (about 6.8 mm/a). We also find that the ET products from remote sensing both underestimate the actual ET in the WLRB, compared with the GRACE-based ET results. However, the ET simulated by the global land data assimilation system version 2.1 (GLDAS-2.1) Noah is overestimated. This paper highlights the capability of GRACE to monitor actual ET in the WLRB and to validate different ET products from remote sensing and models.
Geomagnetic matching technology can provide passive external resource of correction information for underwater vehicle, which improves the long voyage navigation accuracy of inertial navigation system. The filtering matching algorithm is the core technology in geomagnetic aided inertial navigation, which can effectively mitigate the influence from the uncertainty of geomagnetic observation noise. Based on track simulation data, the earth magnetic anomaly grid 2 (EMAG2) is selected as reference map in this paper. To properly model the observation noise of geomagnetic anomaly in unpredictable geomagnetic environment and the error from measuring instruments, this paper proposes the matching algorithm of geomagnetic anomaly filter based on residual errors. The observation noise variance is estimated adaptively through the residual-based adaptive estimation (RAE) filter. Meanwhile, the availability and robustness of the algorithm are improved by combining the optimal filter selection criteria. The validation experiments of RAE filter are conducted in different maritime space of the South China Sea. It is shown that the drifting errors of inertial navigation system in longitude and latitude can be reduced based on the RAE filter. Moreover, the positioning accuracy and reliability of the aided navigation system can be improved significantly. In obvious geomagnetic fluctuating region under the simulation condition, position accuracy is raised to 0.751 53 n mile and 0.778 45 n mile respectively in latitude and longitude directions through RAE filter.
The successful launch of Chang'E-4 relay satellite in June 2018 has laid a good foundation for the mission of landing on the back of moon at the end of the year. Very long baseline interferometry (VLBI) system plays an important role in satellite positioning and orbital transfer. In this paper, the data accuracy of delay, delay rate and positioning of VLBI observations during the real-time phase of the Chang'E-4 relay satellite mission are calculated, and the application of positioning reduction in the real-time track monitoring of the orbit control arcs of the Chang'E-4 relay satellite is presented. In the process of the relay satellite entering Halo orbit, we show the real-time change of the orbit elements converted by our positioning method and compare the real-time positioning results with the post-time orbit determination results to further verify the validity and accuracy of our real-time positioning results. We provide a new Halo real-time track monitoring method.
On November 12, 2017, a strong Mw7.3 earthquake occurred in the Sarpol-e Zahab region on the Iranian-Iraqi border, killing more than 500 people. Earthquakes cause serious damage, but do not cause rupture of seismogenic faults on the earth's surface. In order to study the focal mechanism of earthquakes and the crustal deformation caused by earthquakes, this paper uses ALOS-2 and Sentinel-1 satellite data, obtains the seismic co-seismic deformation field by over-differential interferometry, and then uses the resolution constrained quadtree sampling method to observe it. The data are sampled for down sampling. On this basis, a two-step inversion algorithm is used to subdivide the fault plane, and the precise geometric parameters and the optimal fault slip distribution are obtained by inversion. The inversion results show that the seismogenic fault is a fault with thrust and dextral strike slip, and the fault strike is determined 347°. The inclination angle is 15°. The distribution of seismic slip is mainly distributed in16-19 km. The depth range and the maximum tilting slip are 4.5 m. Based on the inversion of the fault model, the 3D seismic deformation field is obtained. The results of the 3D seismic deformation field show that the deformation caused by the earthquake is mainly in the vertical direction, and the vertical uplift deformation reaches 78.8 cm. The inversion results of the focal mechanism and the three-dimensional deformation field results are in good agreement with the seismological research results.
In the autoregressive (AR) model, random errors in the observation vector are homologous to those in the coefficient matrix. In view of the unreasonable distribution of the observation weight matrix and the inaccuracy of the random model, the random quantities in the augmented matrix consisting of the coefficient matrix and the observation vector are extracted by the variable projection method. Then, we transform the errors-in-variables (EIV) model into the nonlinear Gauss-Helmert (GH) model and propose a structural total least squares (STLS) algorithm by the nonlinear least squares adjustment theory. Combined with the least squares variance component estimation (LS-VCE) method, the variance component estimation method of STLS problem is derived. Furthermore, it is applied to the variance component estimation of the AR model. Through the real example, the effectiveness of proposed algorithm is verified. Meanwhile, the results are consistent with those of modified existing variance component estimation methods, but the construction of observation weight matrix is simple, it can also applied to the estimation of covariance factors.
This paper derives the ellipsoid harmonic series of earth's gravitational field about the second kind of associated Legendre functions and its first and second derivatives recursive calculation method, and it compares their results with those of traditional Jekeli recursive method. The result shows that the numbers with maxima of about 30 is required terms of the revised Jekeli's recurrences is about half numbers compared with the traditional recurrences, when the recursive computation has the same convergence accuracy. As degree n, order m and convergent k increase, the revised Jekeli's recurrences is all fulfilled to 1×10-6. Therefore, we also can see that the relationship of the height h and degree n in the ellipsoid approximation, as similar with the approximation of spherical harmonics's (R/r)n+1 in different heights.
Aiming at the problem that Gauss projection in the line engineering project needs to be frequently strapped and has low precision when designing the plane construction drawing of the project, this paper proposes a large elliptical Gaussian projection of an elliptical line projecting from the ellipsoid elliptical line of the route for the new central meridian, and analyzes the parametric theoretical model of the elliptic ellipsoid. Firstly, it derives the meridian arc length formula by using the naturalized latitude instead of the geodetic latitude as a parameter. Secondly, according to the quadratic curve invariant theory, linear algebra, calculus and other knowledge, it deduces the model of solving the basic geometric parameters of the large ellipse ellipsoid with the plane equation coefficient as the parameter. Furthermore, a direct conversion model of the geodetic coordinates between the base ellipsoid and the large ellipse ellipsoid is deduced. Finally, based on some actual engineering data, the correctness and superiority of the theoretical model deduced in this paper are verified to be popularized in long-term projects.
To solve the problems of low accuracy and poor stability in the traditional estimation of zenith delay, a method of adding annual and semi-annual periodic terms on the basis of Hopfield model is proposed. Based on the atmospheric data of the global geodetic observation system at 45 stations from 2012 to 2014 in China, the time series and spectrum distribution of zenith tropospheric delay (ZTD) and residual Hopfield model are analyzed. With the introduction of annual and semi-annual cycle term, the GHop model suitable for China region is established, and the accuracy and adaptation of the two models are evaluated. The results show that the deviation and middle error of Hopfield model represent obvious seasonal variation in time, while GHop model is small and stable. In terms of spatial distribution, the Hopfield model varies greatly with the increase of elevation and latitude, GHop model can adapt to different latitudes and elevation ranges. Annual deviation of coincidence accuracy in GHop is 28% higher than that of Hopfield, and 76 radiosensory data in China are used for external coincidence test. The statistical results are better than those of Hopfield model. The ZTD results calculated by the proposed method are more reliable and have high practical value.
Urban land use/land cover classification and change detection based on remote sensing imagery are of great significance in land use surveying and updating. Based on Wuhan high-resolution aerial and satellite remote sensing images and corresponding GIS vector data, we propose a novel convolutional neural network to apply in the urban land cover classification and change detection. Firstly, a fully atrous convolutional neural network (FACNN) is proposed, which could take into account the different scale and LOD (level of detail) of polygons in the GIS vector data. Then, both pixel-based change detection and object-based change detection are analyzed according to the classification maps from FACNN and a previous GIS map. Finally, the effectiveness and advantage of our method are verified by the classification and change detection experiments in very high resolution remote sensing images of Wuhan city covering more than 8 000 km2. The proposed FACNN proved outperforming mainstream CNN based methods as FCN-16, U-Net, and Dense-Net, and the precision of the object-based change detection achieved 74.1% and the recall was 96.4%, indicating application prospects for unban GIS map updating.
Synthetic aperture radar (SAR) has been proved as an effective tool for agricultural monitoring and its effectiveness depends on the accurate and appropriate interpretation of SAR Information. Stokes parameters, which is proposed on the dichotomy principle of electromagnetic wave, describe the changes of the incident electromagnetic wave affected by objects which are radiated by electromagnetic wave, and then obtain the information from the objects. However, to our best knowledge, few reports focus on crop phenology monitoring using Stokes parameters. This study aims to explore the feasibility of Stokes related parameters for crop growth monitoring. In this paper, Stokes parameters and their subparameters are calculated based on assumption of wave transmitted in horizontal and received in both horizontal and vertical polarization. Then these Stokes parameters are derived from 5 multi-temporal Radarsat-2 images and averaged relying on each oilseed rape field area. The correlation between all of the Stokes parameters and oilseed rape growth parameters including above ground biomass (AGB), height and leaf area index (LAI) are computed by Pearson product moment correlation coefficient. The significance of these Stokes parameters for oilseed rape AGB, height and LAI inversion are derived from random forest (RF) model. The results indicate the potential of Stokes parameter for crop growth monitoring and growth parameters inversion. It demonstrates that scattering power related Stokes parameters reveal better performance for crop AGB inversion, but scattering mechanism related Stokes parameters such as degree of polarization (m), the degree of linear polarization (ml) and the ratio of linear polarization (μl) are more sensitive to crop height and LAI. Moreover, the results suggest that it is necessary to analyze the influence of crop structure on scattering power when crop growth parameters inversion is performed with Stokes parameters.
Aiming at the incomplete retention of features during the point cloud data procession by point cloud simplification algorithm, and data holes caused by small-curvature point cloud simplification algorithm, this paper proposes a new point cloud simplification algorithm integrated k-means clustering and Hausdorff distance. The topological adjacency is established in the new simplification algorithm based on the OcTree algorithm.Then the principal curvatures of all point cloud is calculated and the Hausdorff distance of the principal curvatures is calculated, and the Hausdorff distance threshold set by the requirements of the reduced target is used to extracted the point cloud feature. Finally, k-means clustering is performed on non-feature regions to extract feature points, and the extracted feature points are merged to obtain reduced results. Results show that the proposed algorithm can retain the feature information of the model more completely and avoid the void phenomena.
Visibility analysis is one of the most important parts of the spatial analysis in geographic information system, which is calculated based on the measurable models. However, it is hard to construct accurate models for the huge number of objects automatically because there is a mess of various objects in the urban area. Mobile laser scanning can acquire accurate three-dimensional information together with other physical properties (such as reflected intensity, echo waveform, etc.) on road and along roadside flexibly and efficiently, which provides an alternative data source for visibility analysis of the large-scale road scenes. This paper proposes a fast and robust depth-buffering method to analyze visibility based on point cloud data in road scenes efficiently and robustly. To achieve the goal, an adaptive spatial index construction strategy is firstly introduced based on the viewpoint and the corresponding field of view. Then the viewshed between the viewpoint and the field of view is analyzed efficiently using the depth-buffering method. The visualizing degree from the road to the traffic sign and the illuminated region on road of the street lamps are estimated respectively to verify the feasibility as well as the flexibility of the proposed method. The performance of the experiments shows that the proposed method can assist in monitoring the infrastructure health and decision support for the municipal planning department.
Using unmanned aerial vehicle (UAV) remote sensing method, this paper proposes a new approach to monitor the situation for oilseed rape based on multi-temporal spectral analysis. According to the rape canopy structure, this experiment determines the end-member combination of different growth stages and obtains the abundance data with the linear decomposition model. Aimed at the accumulation process of rape yield, we analyze the relation between the abundance data in three growth stages and the final yield, and propose three independent variable schemes to establish the multi-temporal rape yield estimation model. Experimental analysis shows that the combination of abundance and normalized difference vegetation index (NDVI) is more effective than the widely used NDVI. And the optimal abundance combinations are the same under different planting methods. This model is also suitable for different rape planting patterns. Experiment results verify the strong stability of the proposed model, and the multi-temporal combination of abundance data and NDVI can improve the effect of yield estimation model.
Maps play an important role in people's production and life. There is a lot of information in annotations. Identifying map name annotation categories is of great significance for computer reading maps and further drawing maps in the future. Recently, popular deep learning technologies, especially convolutional neural networks, have a good effect on solving image classification problems. Training sets are used to train deep neural networks, and deep neural networks can extract the features of the data set pictures themselves and continue to adjust model parameters until the training is completed. This paper uses Google's open source framework TensorFlow as the experimental deep learning platform to conduct intelligent classification research on multiple annotation datasets of multiple Atlases. Manually obtain annotation images from the Atlas as sample datasets to construct a convolutional neural network model and try to use two methods of mixed training and separate training to train the models. Experiments show that the model obtained by the mixed training method performs better.
The continuous advancement of economic globalization and informatization, along with the rapid construction of transportation infrastructure, have made cities more closely connected and made urban networking a prominent trend. Network news is easily accessible and contains bundant geographical information. Studying urban networks from the view of toponym co-occurrences in the news is a brand-new perspective, and conclusions may help clarify the status of cities, deepen the understanding of the structure of urban networks. This paper conducts research from contact, node and network successively. Firstly it proposes a new method aiming at measuring relatedness of city pair based on toponym co-occurrences; then it uses the centrality of social network analysis to characterize urban influence and explores its spatial distribution; finally it studies the characteristics and structure of urban networks. Results show that the proposed measurement approach highlights the strength of the relatedness, and makes up for the shortcomings of Ochiia coefficient method such as ignoring multiple diverse toponyms in the news; coastal cities possess higher urban influence than inland ones; the backbone urban network presents an approximately diamond-shaped spatial structure, metropolises such as Beijing, Shanghai, Guangzhou, and Chongqing are the core nodes.
Road network kernel density estimation is a cluster analysis method for event points under road network constraints. It is often used to study spatial distribution patterns of traffic accidents, urban crimes, vehicle trajectories and other events. The traditional serial algorithm of road network kernel density estimation has higher efficiency under the condition of small data volume, but with the increase of data volume, the performance of the algorithm is significantly reduced, which can not meet the actual application requirements. In this paper, an efficient parallel algorithm based on Spark computing framework is designed and implemented for road network segmentation and kernel density calculation in road network kernel density estimation method. Taking traffic accidents as an example, four groups of experiments are used for comparative analysis. The results show that the parallel algorithm of road network kernel density estimation based on Spark computing framework has high computational efficiency and good extensibility.
In received signal strength indication (RSSI)-based positioning, the deployment structure of beacon nodes greatly affects the accuracy of positioning. In theory, the intensive and well-proportioned distribution of these deployed can insure the high precise positioning results. However, the RSSI is seriously disturbed by noise in the actual environment, which makes it difficult to ensure stable signal propagation even if the layout is reasonable. In addition, dense deployment can lead to crosstalk between signals, and complex physical environments make it difficult to achieve ideal deployment of beacon nodes, which greatly affects the positioning result. In this paper, we propose an algorithm for optimal selection of beacon nodes in trilateration positioning. After beacon deployment, the optimal reference beacon combination for each local region in the positioning scene is constructed according to the geometric relationship of the beacon nodes. Then, the beacon sequence scanned by the device in real time is matched with the regional best beacon combination. The preferred beacon nodes are used to solve the position, and the positioning accuracy is improved by optimally selecting the reference beacon nodes. Experiments show that the proposed algorithm can improve the positioning accuracy efficiently, which is close to the result of selecting the best reference beacons.
Coordinate transformation between quarternary triangular mesh (QTM) code and longitude/latitude is one of the main factors in affecting the application of QTM. However, there are significant flaws in the existing algorithms. To overcome this deficiency, an improved transformation algorithm is proposed in this paper. The main principle is as follows:According to the row and column of QTM, the address codes are recursively approached in a certain direction. Then, the transformation results are obtained precisely by introducing the operation of the relation between the judgment point and the line segment. In this algorithm, not only the transformation result is accurate but also time consumption is only about 10.1%-10.4% of equal-triangles projection (ETP) method. In addition, the QTM code obtained in the improved algorithm still has the directionality. The improved algorithm works well for both traditional QTM and the QTM which using latitude-line instead of circle-line.
To study the problem of integrity authentication for vector map data, a fragile watermarking algorithm for locating tampered entity groups is proposed. Firstly, each geographic entity is represented by the midpoint of its minimum bounding rectangles. On this basis, geographic entities are divided into groups using optimized k-means clustering algorithm. Then fragile watermarking is generated through building integrity authentication parameter and combining it with chaotic mapping. Finally, authentication information is embedded in the sorted coordinates. Watermarking detection corresponds to the embedding procedure. Whether the map data have been tampered could be determined by comparing the consistency of the extracted watermarking with the generated watermarking. Experimental results show that the proposed algorithm is able to preserve the accuracy of vector map data effectively and authenticate the integrity of vector map accurately at the same time. Furthermore, the algorithm shows favorable tamper localization ability.