2006 Vol. 31, No. 1
This paper summarizes the main content of technique and theory of spatial data processing respectively based on the authors own understanding on spatial data processing,discusses the current research status of spatial data mining and its emphasis in the future research,and compares the similarities and differences between spatial data processing and spatial data mining.The development direction of each one is forecasted.
This paper analyzes the principle,error sources and precision of triangulated height surveying,points out the key problems about first order leveling replaced by triangulated height surveying;and for the first time puts forward that in some given conditions,it is not only feasible but also valuable to replace first order leveling by precise triangulated height surveying,and proves it by experimentation as well.
This paper uses the Galileo orbit parameter for simulating the Galileo satellite position and DOP,compares the visible satellite number and DOP between GPS and Galileo system,at same time and same place.The result indicates that Galileo constellation is better than GPS.And simulates GPS and Galileo combined constellation,more visible satellite brings diminished DOP, in contrast with single system,so that improves the position precision,and also broaden the scope of position due to limited elevation.
The realization of square-root information filtering and smoothing in orbit determination software is discussed,a new real-time quality control method based on SRIF/SRIS is presented.The on-board GPS data of CHAMP satellite is processed by SRIF/SRIS module of PANDA software.The results show that even small slips can be surely detected and the reduced dynamic orbit of CHAMP can reach 5-8 cm,when 24 IGS station were solved simultaneity.
In this paper,the model of generalized net adjustment based on coordinate pattern is researched.With this kind of model,the coordinates are used as observations.And the methods of holistic disposal different sub-net with point pattern are proposed.
Applying precise point positioning(PPP) technology to airborne survey without ground base station was implemented in this paper.The results show that cm to dm accuracy could be achieved.Epoch-by-epoch 3D RMS calculated from postfit residuals of observations in both kinematic and static case is smaller that 3 cm.The external RMS calculated from epoch-by-epoch differences between the PPP solution and double-differenced solution with kinematic data shows that horizontal RMS is smaller than 5 cm,vertical RMS is better than 10 cm.Making differences epoch by epoch between PPP simulated kinematic solution and known coordinate of static base station derives RMS shows that horizontal RMS is smaller than 3 cm,vertical RMS is better than 5 cm.
This paper analyzes the cluster index selected through the theory of carrier phase combination,uses the maximal tree method to classify triple-frequency carrier phase observations of GPS and discusses the results,proves the typical observation selected according to the matrix transformation and a living example.The results shows that the method is feasible and reliable.
Based on sumed up the current algorithm between long./lat.coordination and QTM code,a new algorithm called three orientation translating algorithm based on assistant geocentric rectangular coordinate that is used to translating algorithm between long./ lat.coordination and QTM code is presented.At the same time,in order to compare with the differences among the algorithms,a new criterion judge of precision-invert error rate.The result indicated that three orientation translating algorithm not only has good rate and absolute accurary,but also the orthembadism invert error rate is 0.
The TROS-I is a new generation mobile SLR system in China.TROS-I has successfully obtained SLR data in Urumqi and Lhasa in middle western China where no SLR station existed before.In this paper,construction and observation result(to Lageos etc.) of TROS-I is introduced.The result is shown that the normal point precision of TROS-I is better than 10 mm,and its effective range is 20 000 km.So TROS-I can help upgrade tracking ability of Chinese SLR network,and it also has great potential in precise orbit determination of home made satellites,research of China crustal movement nowadays etc.
A quantitative method of RS for monitoring soil losses based on GIS and the formula and algorithm of each model factor about the revised universal soil loss equation(RUSLE) are described in this paper,the spatial distribution law of soil erosion is discussed and the amount of soil loss of the small watershed is predicted running the model of RUSLE by the support of GIS.
Based on the combinational representation of topological spatial relations,the method of combinational reasoning with the basic topological spatial relationships is further perfected in this paper.A sound reasoning table about topological spatial relations is given,and a complete series of diagrams on the topological spatial relations between two lines is drawn.
In this paper,a binocular vision system and a method for determining 3-D motion parameters of an object from binocular sequence images are introduced.The main steps include camera calibration,the matching of motion and stereo images,3-D feature point correspondences and resolving the motion parameters.Finally,the experimental results,acquiring the motion parameters of the objects with uniform velocity and acceleration in the straight line based on the real binocular sequence images by the mentioned method,are presented.
In this paper,some initial investigations are conducted to employ resource limited artificial immune system(RLAIS) for classification of multi-spectral remote sensing image.The proposed method trains the samples of regions of interest using the RLAIS and obtains the optimal the centers of classification.Image classification task by RLAIS is attempted and the preliminary results are provided.By the experiment,it is demonstrated that the method of this paper is superior to the three traditional algorithms,and its overall accuracy and Kappa coefficient reach 91% and 0.88 respectively.
This paper proposes to estimate abundances from hyperspectral image using probability outputs of support vector machines(SVMs),trains a SVM with a gauss kernel function,and trains the parameters of an additional sigmoid function to map the SVM outputs into probabilities.An experiment of real hyperspectral image is conducted to validate the procedure.The abundances estimated by SVM comparison with those of the linear spectral unmixing are also given.The experiment results show that the method can provide effective result of abundance estimation for hyperspectral image.
This paper introduces the methods of hyperspectral images band selection based on the property of Hyperspectral remote sensing images,utilizes the Projection Pursuit approach to find optimal solutions for a selected projection index based on dynamical evolutionary algorithm and then project a high dimensional data set into a low dimensional data space to produce a sequence of projection images,explores zero-detection method to threshold projection images to detect anomaly target.The experimental results show that the Projection Pursuit based on dynamical evolutionary approach is an effective means to detect anomaly target in hyperspectral images.
Based on the Bayesian approach,this paper proposes a method to use marked point process to model the prior knowledge of the expected extraction object such as road network.The marked point process model cooperate the prior model and the image data to extraction the linear object in high-resolution remote sensing images.The whole process embedded into the simulated annealing algorithm.
A method to extract the object in nature-light image is introduced in the paper.In order to extract the object,a background with stripes texture has been set up in the measure field.Differential algorithm to find the object and its position in the image can be used according to analyze he characteristic of the noise image.The experiments show that the method is effective to extract the object in nature-light image.
This paper discusses the shortage of conventional algorithms of texture classification based on wavelet transform,presents two improved approaches of point feature weighting and smart windows.The point feature weighting algorithm constructs the deviation vector using means and variances of samples,and uses the deviation vector to restrain the feature units which do harm to the classification accuracy.The subsequent experiment results show that the performance of the presented algorithm is much better than the conventional one.
The optimal path algorithms on wide range road network are reviewed and existed problems are discussed in this paper.In order to make up insufficiency of existed optimal path algorithm,the thesis proposes optimal path algorithm based on the pyramid-like multi-level road network.The design of algorithm,mathematics expression,as well as algorithm implementation are narrated.The application example is implemented to prove the efficiency and the rationality of this algorithm
This paper puts forward a rough set using GIS product quality evaluation based on merits and defects that methods of GIS product quality possess at present.The example shows that the rough set is used to evaluate GIS product quality,which can achieve a better outcome.
An error model named the stochastic error process of curves is proposed to describe the errors of curves in GIS.In terms of the stochastic process,four characteristics concerning the local error of curves,namely,mean error function,standard error function,absolute mean error function and the correlation function of errors are put forward.The total error of a curve is expressed by a mean square integral of the stochastic error process.The probabilistic meanings and geometric meanings of the characteristics mentioned above are also discussed,and a scan digitizing test is designed.The result of the test shows that the system error in digitized map data is not negligible,and the errors of points on curves are chiefly dependent on the curvature and the concavity of the curves.
Based on the information theory and fuzzy set theory, in the light of the characteristics of randomicity and fuzziness of part spatial data in GIS,this paper proposes a hybrid entropy model of spatial data uncertainty,and hybrid entropy can be used to measure the total uncertainty of spatial data caused by stochastic uncertainty and fuzziness uncertainty.Taking the uncertainty of line segment as an example,this paper discusses the evaluating methods of statistic entropy,fuzziness entropy,and hybrid entropy of line segment uncertainty.And the entropy-band distribution of line segment uncertainty is shown.
This paper discusses the architecture of both mobile client side and server side in the embedded database based GIS application,and analyzes that what the features of the embedded database based as GIS are and how to deal with those key problems under the mobile computing environment.
This paper introduces some concrete methods and ideas on the construction of hydrologic integrated database,including the database classification,database store design and logic design and some delicate sides of them.These schemes provide effective technologic support for the faster development of the whole project.Meanwhile it intends to promote the widespread application of GIS spatial information technology in the hydrologic field.