2011 Vol. 36, No. 6
By analyzing the gain matrix,we find that the general index for verifying the accuracy of the attitude parameter is not reasonable on maneuver condition because the attitude estimators,and their update covariance matrix are influenced by inaccurate predictors in classical Kalman filtering.A new verifying idea according to some components of the velocity vector in body frame equal zero in the GPS/INS system is proposed.The new method need not add the additional sensors,so it is economical,convenient and simple.
The concept and partition of GPS service volume have been established and specified in GPS SPS PS(standard positioning service performance standard) published recently.And the terrestrial service volume(TSV) includs near-Earth space,in which spatial exploration is much active.The new changes give the point that the technology of satellite navigation and positioning is feasible and promising to support users in near-Earth space.Based on the partition of the GPS TSV and SSV(space service volume),we discussed the judgment of the visible satellites in the both space and the feasibility of the received signal power,and carried out the UERE budget of the GPS signal in near space under the standard positioning service condition.Then the navigation performance statistics at the different heights at the middle latitude single point and over the global volume were carried out respectively with the Monte-Carlo method.The results show that the navigation accuracy will improve gradually with the users' height increasing in TSV,and into SSV,the accuracy deteriorates rapidly,and the service interrupt will appear at some height.The conclusions also show that the GDOP at the certain height in TSV will change its precision dilution property,which means that the positioning accuracy is much higher than that of UERE,and that the vertical accuracy is becoming higher than the horizontal accuracy gradually with the users' height increasing in TSV.
GPS common-view approach is currently a widely used technique in time transfer.We analyze the principle and model of GPS common-view method and present a method to gain geopotential difference and orthometric height difference using GPS common-view data based on gravity frequency shift.We choose the time difference series among four stations from BIPM Circular T data to carry out the experiment.Results show that due to accuracy limit of GPS CV,the mean difference and standard deviation between calculated value and theory value of the height difference are at the order of tens of meters.
Failures often exist in dynamic precise point positioning due to the effects of the environment or GPS receivers,so the precision will degrade quickly.Interacting multiple models(IMM) based on the ratio of the probability of failure models is introduced,in order to improve the efficiency of diagnosis,control the outlier's influences on the dynamic precise point positioning.In the new procedure,IMM including all failure models is used for correct detection and diagnosis of failures at first.Failures are detected and diagnosed by comparison with the ratio of the probability of failure models to improve the validity and reduce the skip.Robust theory is used to control the influences of outliers and improve the precision of dynamic precise point positioning.An actual data is employed to test the modified IMM.The results show that the efficiency of detection and diagnosis is improved,outliers are controlled and the accuracy of dynamic precise point positioning is also improved.
The theory and mathematical model of uncombined precise point positioning were proposed.And several effective data processing strategies were proposed to improve PPP receiver clock solution by avoid absorbing some un-modeled system errors.Full-featured PPP software was developed base on the GPSTk core library.Experiments with data from IGS stations distributed all over the world were carried out to evaluate the performance of GPS timing with uncombined PPP algorithm.The results support that PPP algorithm can achieve sub-nanosecond(0.3 ns) level receiver clock solution,and the uncombined PPP algorithm has better accuracy than the classic PPP algorithm with all things being equal.
Combining neighboring swaths is an important step in swath sounding data processing.The error structure of ping in swath sounding is brought forward.Then the adjustment defect model is given for the swath combination net,and we poin out the lack of datum data is main factor to make the adjustment mode defect.To solve the large equation based pings,the formulae is derived based on the theory of adjustment division.A practical example for detecting systematic errors is given.Experimertal results show that the defect model can detect the systematic errors on a ping,or on a swath even on the whole surveying area.
In the strap-down gyro north seeker,gyro drift are primary factors to degrade orientation accuracy.On the basis of eliminating high frequency periodic noises,a new algorithm based on robust estimation is presented to degrade the influence of abnormal noises on the accuracy of FOG north seeker.The robust estimation formula of FOG and its influence function are deduced.Robust estimation serves the median of observations for the robust initial value and applies the median of the residuals of the first epoch robust estimation to obtain stable standard deviation factor.The mixed iterative algorithm based on high breakdown point initial value and assisted by IGGⅢ scheme is described.The calculations show that robust estimation based on reliable initial value and standard deviation factor can effectively degrade abnormal disturbances,and in turn,can improve the accuracy of FOG north seeker.
rt-term warning model is built on the analysis of the cumulative sum statistics.Then the relationship of the minimum warning deformation(MWD),the average run length(ARL) and the warning delay(WD) with respect to the warning threshold are presented.Based on these the definition of short-term warning model availability,false alarm rate and missed alarm rate are given.We build the integrity and the availability monitoring algorithm of the short-term warning system.According to the simulated deformation signal(μ0=0,σ=2),different early warning thresholds are computed to give the different false alarm rate.The results show that the model is able to give an effective short-term warning when the abrupt deformation or the slowly growing deformation are considered.At the same time,the short-term warning delay is related to the ARL.The model is successfully applied to the short-term warning of the field collected data,which provides an overall quality control for the establishment of short-term warning theory.
In the geometric processing of high-resolution satellite imagery,the general rational function model(RFM) consists of 78 rational polynomial coefficients(RPCs),and the correlation between these coefficients will result in difficulties and accuracy degradation in solving the RPC.The rigorous error equations for solving the RPCs are derived,and an optimized method of selecting the RPCs by analyzing the multicollinearity between column vectors of the design matrix is proposed.A SPOT-5 HRG image in level 1A is selected and used.The empirical results have shown that selecting 20 to 30 RPCs could effectively eliminate the correlation between the coefficients,remove the oscillation in approximating the terrain by using the RFM,and obviously raise the solution accuracy of the RPCs and geometric processing accuracy based on the RFM with less ground control points(GCPs) available.When the GCPs are sufficient,the selected RPCs could acquire a consistent geometric accuracy in approximating the terrain with the 78 RPCs solved by the traditional least-squares method.
The traditional approach for relative orientation is the least-square solution of coplanar equation with more than five correspondence rays when the approximations acquired by direct methods.We present a hybrid genetic algorithm which utilizes the global convergence of the traditional genetic algorithm and the local convergence of the gradient-based search algorithms.And unit quaternion is used to represent 3D rotation instead of traditional Euler angles to enlarge the convergent region.The algorithm can effectively eliminate the premature of the traditional genetic algorithm and rapidly converge to the optimal solutions without initial values.
Terrestrial laser scanner(TLS) is a kind of new sensors which has been developed in recent years.In order to construct the objects using the point clouds,one of the problems has to be tackled first is feature extraction.In order to extract linear and circular features of ancient architecture,we segment point clouds and identify these segments according to their geometric features and physical properties of targeting objects.The Hough transform and the least squares method are used to extract lines and circles from laser point clouds of ancient architecture.The effectiveness of these two kinds of algorithms is compared and tested with real data.Several experimental results are presented as well.
Extraction and matching of conjugate image features is prerequisite for registration of multi-sensors images.Feature of image includes points,lines and polygons.We focus on area feature-based image registration for the reason that area features improves registration accuracy.More importantly,area features are often the sole basis for image registration.We propose using 'dilation' operator in mathematical morphology as a pre-processing procedure to prevent boundaries extracted using conventional Laplacian of Gaussian(LoG) operator from becoming discontinuous.We use a boundary algebra algorithm to mark area features with closed boundaries rapidly.We explory singular value decomposition(SVD) to match images based on centroids of area features extracted beforehand.Experiments confirmed that the proposed methods are superior over conventional methods in terms of speed and accuracy.
The mixels in the hypersepectral images directly influence the accuracy of target recognition.A large number of spectral unmixing methods are all based on the convex simplex geometry and the hypothesis of the pure pixels' existence.Actually,this hypothesis is very hard to be met in practice.We present the minimum volume constrained linear spectral unmixing algorithm,which isn't based on the presence of the pure pixels,and calculates the endmember matrix with the quadratic programming method in the reduced dimensional space.Then we estimate the abundance and extract the endmembers with least square method.Experimental results demonstrate that the proposed scheme for decomposition of mixels overall outperforms the MVC-NMF algorithm in the mass.
We propose a spectral matching algorithm,spectral angle sensitive forest(SASF),which improves the spectral matching efficiency in high dimensional large-scale hyperspectral dataset.The locality sensitive hashing(LSH) is expanded to the metric space of spectral angle.Moreover,we introduce a new scheme to index the data bucket,which remove the flaw of the original LSH method that a part of the query points won't get any neighbors.We provide systematical analysis of the parameters,and theoretical and experimental evaluation of the algorithm.The computational efficiency of SASF is proved outperforming the former algorithms.And SASF also provides a tradeoff between efficiency and precision of spectral matching by which make the user has more choices in the applications.
To further improve the accuracy of image retrieval,an algorithm based on nonsubsampled contourlet transform(NSCT) and Krawtchouk moments is proposed.Firstly,the image is decomposed to the low resolution image with NSCT,extracting the contourlet coefficients of each decomposition level of different direction and using the statistical features of the coefficients as texture features of the image.Then,the Krawtchouk moments are employed to extract shape features of the image.Finally,the image retrieval is achieved based on weight distance measure.A comparison is made with the method of using NSCT and Zernike features.Experimental results show that the extracted features have the properties of translation,rotation and scaling invariance,and the proposed algorithm can acquire the higher retrieval accuracy.
A color edge detection approach based on proportion decomposition and nested windows is presented.At first,from information point of view,the proportion decomposition algorithm was presented.We used the principal component analysis algorithm to decompose the color picture,with reference to the human vision system.Then the nested windows were used to detect the multi-scale edge,and the vector templates were used to detect the hue edge.Finally,each component edge was fused and thinned to obtain the color edge.Experiments have been carried out and the results show that the proposed approach can fuse the brightness component and the hue component effectively.The method also can deal with the noise,and can be used to analyze the multi-scale edge.
We apply the corner reflector interferometric synthetic aperture radar(CR-InSAR) to the monitoring of highway.Because the traditional method has large uncertainty in estimating the row and column information of corner reflector(CR) points,we present a method for CR point identification,which is based on the comprehensive assessment index of the intensity and correlation coefficient.This method is used to find the CR points in the five SAR images of the study areas respectively,and the distances between the two consecutive CR points in different SAR images are used to verify the accuracy of the results.Finally,we monitor the deformation of the highway using these identified CR points and acquires the displacement in the period of 2008 ~2009,which are accordant very well with that of leveling.The results show that the CR point identification method presented is effective and reliable in highway deformation monitoring.
Based on the principle of photogrammetry,we analyze the features of the geometric conditions for photography of the power line.Then,a strategy to determine the locations of the power transmission line is proposed and realized,as well as one method that meets the requirements of real-time ranging based on object-space matching is extended.Also the technical problems of real-time measurement are researched.Finally,experiment results prove that this method is feasible and effective.
agriculture,hydrology and meteorology,monitoring the soil moisture in real time and in the long term by the remote sensing technology is very meaningful.However,how to quantitatively retrieve the properties of objectives is a unsolved problem in remote sensing.A new polarimetric method,which is different from the traditional remote sensing by radiometry,was employed for retrieving the soil moisture.A scattering polarization model was developed to analyze the polarimetric characteristics of soil moisture.And this model was examined by experiments.We found that the modeling data were very close to the experimental data.So the quantitative relationship between the parameters of this model and the soil moisture could be found.
We mainly discuss about some of the quality requirements in national standards.Some conflicts between these requirements and real data in practices are indicated.Based on the data quality theory of GIS,the error theory of DEM,ordered isomorphism of elevation and the generalization theory of DEM are analyzed in terms of resolution ratio,data logical consistency and integrality.DEM should not be the elevation of a specific sampling point but the typical elevation within the grain range(dx×dx),and DEM generalization plays a key role in DEM production.
The Gestalt cognition psychology plays an important role in the recognition of linear island alignment by several perceptual grouping principals,namely the law of proximity,compactness,good continuation.Based on the perceptual grouping principles mentioned above,a new method has been developed for linear island alignment recognition.The spatial neighborhood analysis is first conducted by Delaunay triangulation.And based on the neighborhood graph,the corresponding MST is generated to represent spatial structure of island cluster.Then the edge of MST tree is pruned successively according to three different Gestalt conditions.The experiment shows that result is consistent with the manual identification.
In the network geographic information system(network GIS),to improve its quality of service(Qos),we usually use cluster to manager servers,and use load balancing techniques to balance the web loads.We propose a GA-based dynamic load balancing algorithm combined with the existing load balancing methods and the data feather of network GIS.Both the request's content and the state of the server are considered.We also design the ingenious coding and genetic operators of GA accordingly.The experiment's results show that the GA-based load balancing method can reduce response time and bring better service to network GIS compared with other common algorithms.
Query optimizer is critical component in spatially enabled databases.Processing complex spatial and non-spatial queries in seamless fashion is responsiple.However,the lack of spatial statistics and spatial operators limits the accuracy of cost derivation and the generation of optimal query execution plan.We propose an integrated cost evaluation framework in optimizer.Based on the important keys of each node and the characteristic of pipeline,the cost of query execution plan is derived bottom-up.An improved cumulative density histogram is introduced to describe the distribution of spatial data,and is implemented as a module in database server through extensible server-side programming interface.Optimizer uses these statistics to improve the accuracy of cardinality estimation.The experiments illustrate that the cost evaluation framework can correctly and efficiently evaluate different execution plans.The results of different hybrid query statements show the positive correlation between estimated cost and actual cost.
We present a quantitative analysis model for indeterminate topological relations between uncertain spatial regions.Grey set is used to represent uncertain spatial regions,which have a pair of membership function instead a single one to represent the indeterminacy distribution.Using the cut-set operation on the Grey set,uncertain spatial regions can be turned into crisp region pairs.By applying relative qualitative topological analysis model such as the Egg/Yolk model on these crisp region pairs,the indeterminacy of topological relations can be quantified to a probability range.
A quality evaluation model is given out for automatic label placement concentrating on urban rail transit schematic(URTS) maps with analytic hierarchy process(AHP) method.Besides independent elements of legibility threshold,label-feature overlap,position optimal level and label-feature association,we propose two compensatory indicators that related to the number of labels per unit and distortion level for the comparison of label quality among different URTS maps.The rail transit schematic maps of Shanghai,London,Moscow,Milan,Sao Paulo are presented as example verification.The result shows that the two compensatory indicators are reasonable.
We focus on how to automatically place annotation for hash area features and proposes a new method with reference to the compilation specifications for "cartographic symbols for national fundamental scale maps" and Gestalt psychology theory.At first,the candidate positions are generated with convex-grid Method.Then,they are evacuated with the Gestalt factors on the annotation.Finally,the global optimum location of annotation is determined by "conflict" rule.Utilizing the peripheral zone of features is a notable merit of this method.The method has been successfully used in producing topographic maps at 1∶50 000 scale.
We discuss multinomial logistic regression for land use change.Binary logistic regression could be extended to multinomial logistic regression,which exists to handle the case of dependents with more classes than two.Coefficients for each land use types are calculated and also used for prediction of competition.The result shows that the simulation in case of Jiayu County with correct percentage of 73.5% is satisfying.
We propose a new price based method for land grading and benchmark land price evaluation.Vector-grid hybrid model is used to dicide the assessment cells.Firstly,we use trend surface analysis model to eliminate abnormal samples.Then shortest path distance is used instead of Euclidean distance to do the interpolation of sample points.After the overlay analysis,we do some statistical analysis to decide the land grade and get the benchmark land price for each grade.At last,we take the city of Shanghai as an example to show the practicality of our method.
According to characteristics of village and town land use regulation,we propose a novel regulation model considering the land use monitor and evaluation results.The model integrates spatial unit division,regulation units discrimination and spatial allocation.Using simulated annealing algorithm,the model realizes optimal spatial regulation which reconciles the conflicts among multi-objects.Experiments results show that the model is reasonable and applicable.