Table of Content

25 August 2012, Volume 41 Issue 4

Previous Issue    Next Issue

学术论文

Change Detection in Multitemporal Remote Sensing Images based on Dynamic Fuzzy Fisher Classifier and Non Local Mean Weighted Method

2012, 41(4):  0-590. 

Asbtract ( )   PDF (3237KB) ( )  

Related Articles | Metrics

A novel change detection approach based on the dynamic fuzzy Fisher classifier for multitemporal remote sensing images is proposed in this paper, which detects the changes with the joint histogram. To increase the separability of the unlabeled pixels, a non-local mean weighted method is used to introduce the spatial information. The unlabeled pixels are labeled with a predefined probability based on their predictive values. The weights of the unlabeled pixels and the parameters of the dynamic classifier are adjusted according to the updated samples until all the pixels are classified. The proposed method is distribution free, context-sensitive and not affected by the comparison operators. Experimental results on real multitemporal remote sensing images confirm the effectiveness of the proposed technique.

Airy isostasy on the Moon：implications for crustal thickness

2012, 41(4):  0-590. 

Asbtract ( )   PDF (4155KB) ( )  

Related Articles | Metrics

Lunar geoid (or selenoid) anomaly and topogrphy variations are the expressions of its internal density anomaly and boundary undulations. So the geoid to topography ratio (GTR) can be used to study the isostatic state of the Moon and its crustal thickness. In this paper a new lunar crustal thickness model is derived by GTR technique from filtered gravity model SGM100h and topography model STM359_grid-02, with the mare fill removed. This model indicate that the lunar crust has an average thickness of 36.9 km, and the average farside thickness is 13.5km thicker than nearside. The crustal thicknesses at Apollo 12/14 sites of our crustal thickness model are 28.3 and 29.1 km, respectively. It also shows that the crust beneath the mare basins is generally thinner than the regions around.

Extracting Precise Atmospheric Propaganda Delays from Multiple Reference Station GPS Networks

2012, 41(4):  0-590. 

Asbtract ( )   PDF (17343KB) ( )  

Related Articles | Metrics

The atmospheric propaganda delays extracted from multiple reference station GPS networks can server for the fields of weather forecasting, precise positioning and so on, the related extracting method however consists of drawbacks that affect the reliability of the obtainable results, such as inadequate adoption of the known information, severe atmospheric modeling errors. Starting from the modified strategy for GPS network processing, a novel extracting method is derived, whose main points include: (1) directly processing of the un-differenced GPS observables, and elimination of the rank-deficiencies in the observation equations based upon re-parameterization; (2) exploiting the conditions of a-priori known coordinates of the reference stations and the integer-valued nature of the double-differenced ambiguities; (3) introducing the slant ionospheric delays as one of the unknowns, whose temporal behaviors are characterized as constant-velocity process. Experimental results demonstrate that, the short-term predicting precision of the atmospheric delays derived from the new method is at the level better than 1 cm. As the world-widely construction of the Continuous Operating Reference System (CORS), the perspectives of applying the proposed new method can be more and more expanding.

Noise Model Establishment of IGS Reference Station Time Series inside China and its Analysis

2012, 41(4):  0-590. 

Asbtract ( )   PDF (3108KB) ( )  

Related Articles | Metrics

Focusing on the 11 IGS reference stations inside China, this paper selects the coordinate time series of these stations during period of 1995~2010 under ITRF2005, implements noise analysis on them using different noise model combination, meanwhile calculates impacts of different surface mass loadings on station displacement, such as atmospheric pressure loading, nontidal ocean loading snow depth and soil moisture loading, thus yields optimal noise models of different station components before and after surface mass loading correction. We draw conclusion that noise model of IGS stations inside China represent diversity characteristics with different noise feature among different components, mainly behaving as flicker noise plus white noise model and band pass plus power law plus white noise model. Other complicated noise models also account for a certain proportion, which is different from the existing results. Based on these, we quantitatively analyze the relationship among noise model, surface mass loading together with station’s linear velocity and its uncertainty respectively, finding that different complicated noise model would influence station’s linear velocity and its uncertainty with maximum discrepancy larger than 1mm/year. Besides, the calculated four kinds of surface mass loading could introduce noise changes mainly representing as band pass and random walk characteristics. Results of this paper could provide more accurate data for crustal movement investigation, meanwhile providing reference for coordinate time series analysis of continuously operating reference stations in China.

Research on Quality Assessment of Atmospheric Correction Products Retrieved from TM/ETM+ Imagery

2012, 41(4):  0-590. 

Asbtract ( )   PDF (4976KB) ( )  

Related Articles | Metrics

Assessing the quality of atmospheric correction products retrieved from TM/ETM+ imagery is a necessary means to improve the product’s quality, and it is also important to the scientists who use the imagery. At the basis of the summarization of the present methods, a new approach using the available Global Land Survey surface reflectance product as reference imagery for validating the accuracy of atmospheric correction products retrieved from TM/ETM+ is proposed. The steps are as follow: Firstly, the spectrum sampling strategy is designed to get spectrum samples from the product imagery and the reference imagery, then using the coefficient of variation to determine homogeneous samples. After obtaining the homogeneous samples from the imagery, the PIFs (Pseudo Invariant Features, PIFs) samples are identified by using the spectrum correlation coefficient R and the Normalized Difference Vegetation Index (NDVI) differential between the product imagery and the reference imagery. Finally, the quality consistencies between the multi-temporal imagery are evaluated based on the PIFs samples. Experimental results have shown that the new approach can accurately identify the PIFs and obtain the quality information rapidly, and it is very suitable for large-scale applications and can provide the global/region change researchers with some essential quality information on base imagery.

Calibrating the device delay for The C Transfer measuring system

2012, 41(4):  0-590. 

Asbtract ( )   PDF (4353KB) ( )  

Related Articles | Metrics

The C transfer measuring system has superiority in the orbit determination of GEO. The precision of the device delay is the main factor to impact that system’s performance. Base on the orbit determination result which come from combining the carrier observation of multi-satellite, the method to calibrating the device delay with standard orbit is present, and the device delay and time bias can be estimated at same time by this method. The analysis result from regional station show that the orbit determination precision is about 5m after the device delay is calibrated, and the device delay and time bias have an imperceptibility variance.

An Automatic Registration Algorithm for SAR and Optical Images Based on Geometry Constraint and Improved SIFT

²

2012, 41(4):  0-590. 

Asbtract ( )   PDF (8616KB) ( )  

Related Articles | Metrics

An automatic registration algorithm for SAR and optical images based on geometry constraint and improved SIFT is proposed in this paper. Firstly a rough correction of the rotation and scale differences depends on the geometry constraint is applied. Then the SIFT features extracted by the improved SIFT from the two images are matched by SSIM as the similar measure according to the structure information of the SIFT feature. And then, scale and parallax restrictions are introduced to improve the matching performance by clustering analysis in the scale and parallax domains. Finally, the perspective transform parameters for the registration are obtained by RANSAC algorithm with removing the false matches simultaneously. The whole process is done automatically. Experiments show that the algorithm proposed in this paper is effective in the registration of SAR and optical images with large differences.

Interpretation of Oblique Buildings Based on Optimal Polarimetric Coherence Coefficient

2012, 41(4):  0-590. 

Asbtract ( )   PDF (11515KB) ( )  

Related Articles | Metrics

It could preserve polarimetric scattering characteristics for classification of polarimetric synthetic aperture radar image based on Freeman_Durden decomposition, but the classification accuracy is susceptible to target decomposition due to that the scattering mechanisms could not be changed in the process of classification. Buildings aligned not along with the flight direction of radar belong to volume scattering in Freeman_Durden decomposition. And it is difficult to distinguish oblique buildings from vegetation at present. According to the high coherence characteristic of building, optimal polarimetric coherence coefficient is introduced to the new algorithm by analyzing the backscattering characteristics of building, to improve the classification accuracy of tilted building and forest on the basis of target decomposition. The experiment results indicate the effectiveness of the algorithm by using the L-band PolInSAR images of Oberpfaffenhofen around DLR of E-SAR and X band PolInSAR images of Lingshui in Hainan Province of domestic X-SAR.

An Adaptive Method for Incremental Updating of Vector Data

2012, 41(4):  0-590. 

Asbtract ( )   PDF (8039KB) ( )  

Related Articles | Metrics

To maintain data consistency and eliminate the spatial conflicts brought by spatial database updating, this paper purposes an adaptive method for incremental vector data updating. Based on the matching of correspondent objects, a change-object detection and incremental updating method is discussed. Considering the constraint of spatial distance, semantic similarity and topology consistency, we propose a calculated method for edge matching evaluation. An adaptive edge matching strategy is also designed to maintain the consistency of spatial data. We also discuss the rule-based detection and manipulation of spatial conflicts. Topographical data are used to verify the practicality and efficiency of the method. The result shows that this method can achieve high computing speed and maintain the consistency and integrity of spatial data. This paper provides a new idea for incremental updating and consistency maintenance of GIS spatial database.

On the permanent tide in geodesy

2012, 41(4):  0-590. 

Asbtract ( )   PDF (2392KB) ( )  

Related Articles | Metrics

This paper is devoted to treating the permanent tide in geodesy. First the numerical expression of the permanent tide potential is derived, then formal expressions of three kinds of tidal values and transformations among them for six types of geodetic observables frequently encountered are given, and finally the characteristic and usage of the three tidal systems in geodesy are discussed.

Error probability controllable integer rounding method and its application to RTK ambiguity resolution

²,

2012, 41(4):  0-590. 

Asbtract ( )   PDF (6052KB) ( )  

Related Articles | Metrics

After RTK initialization, only new ambiguities introduced one after another with the observation continuity need to be fixed. In general, the ambiguities in the unknown ambiguity vector are few and significantly different in precision due to the different observation periods. Therefore, RTK ambiguity resolution is preferable to the bootstrapping method to partially fix the ambiguities that can be fixed with high reliability. Bootstrapping is a sequential rounding technique and its reliability relies, to a great extent, on the reliability of stepwise rounding. Hence, it is crucial to control the error probability of rounding. In this paper, we will address the issue of reliable rounding based on the theory of multiple hypothesis testing, where the rounding interval is adaptively determined in term of the user-defined critical level to tightly control the error probability of ambiguity fixing. Furthermore, regarding to the two special scenarios in real GNSS applications, two easy-to-use rounding strategies are developed. The dual-frequency GPS data collected from two CORS stations of 94.6 km baseline is carried out to evaluate the performance of the proposed new rounding strategies. The results show that, comparing with the traditional rounding strategies, the new rounding strategies can indeed control error probability of ambiguity fixing whilst keeping the moderate fix-rate.

Single-Frequency Precise Point Positioning With Simultaneous Ionospheric Delay Estimation

2012, 41(4):  0-590. 

Asbtract ( )   PDF (4450KB) ( )  

Related Articles | Metrics

This paper presents a new method for single-frequency precise point positioning using code and phase measurement. Ionospheric delay in each code and phase measurement is estimated as unknown parameters together with receiver clock offset, tropospheric zenith path delay, receiver position and ambiguity of carrierphase. Modified square root information Filtering & Smoothing algorithm with parameter elimination is employed for parameter estimation. The proposed method can be applied to real-time positioning and post processing, and it is independent of any ionospheric model. With the presented method, static positioning test was carried out with data collected at 32 stations of the International GNSS Service (IGS) through day 75 to 90 in 2011,which show RMS in ENU are 0.023m，0.018m and 0. 059m respectively. Kinematic experiment with airborne GPS data collected on September 5th 2008 show that RMS in ENU compared with kinematic relative differential solution are 0.168m,0.151m and 0.172m.

Regional Quasi-Geoid Refined Considering Terrain and Gradient of the Refined Spherical Bouguer Anomalies Correction

2012, 41(4):  0-590. 

Asbtract ( )   PDF (4525KB) ( )  

Related Articles | Metrics

Rigorous formulae of the boundary problem in physical geodesy had been proved considering terrain and gradient of the refined spherical Bouguer anomalies correction. Regional quasi-geoid had been calculated and tested using terrain data, gravity and GPS / leveling data with formulae derived in this paper. Results show that the effect extent by gradient of the refined spherical Bouguer anomalies correction could be up to centimeters. In order to improve the precision of the modeling, gradient of the refined spherical Bouguer anomalies correction should be considered besides terrain correction, especially in the region with great undulation.

The Direct Transformations between Normal Cylindrical and Normal Conic Projections with Different Distortion Properties

Shaofeng Bian²

2012, 41(4):  0-590. 

Asbtract ( )   PDF (727KB) ( )  

Related Articles | Metrics

The direct expansions of transformations between meridian arc, isometric latitude and authalic latitude function were given, and based on these expansions, models of the direct transformations between normal cylindrical and normal conic projections with different distortion properties were established systematically. These models are concise in form and strict in theory, and the fussy processes of traditional indirect transformation formulae are avoided. The formulae derived by this paper are symbolical expressions including the first eccentricity of the referenced ellipsoid, which could solve the transformation problems when different reference ellipsoids are used.

A Method Based on Integrating Real and Complex Correlation Function for InSAR Image Coregistration

2012, 41(4):  0-590. 

Asbtract ( )   PDF (4123KB) ( )  

Related Articles | Metrics

Abstract: SAR image coregistration is one of the key steps in InSAR data processing, while the coregistration accuracy can directly affect the quality of follow-up products. This paper presents a method for InSAR image coregistration based on integrating real and complex correlation function. Firstly, based on the analysis of characteristics of the existing coregistration measure function ,and with international SAR data processing experience, it’s concluded that a robust, universal InSAR image coregistration method should take correlation function as its coregistration measure function. Secondly, for the correlation function have two of real and complex correlation calculations, the author analyzed the characteristics and the adaptation of real and complex correlation function, and put forward the concept of coregistration sensitive factor, which can be effectively and adaptively select the appropriate correlation function. Finally, the implementation steps of the coregistration algorithm are given in detail. And the real data processing results show the effectiveness of the proposed coregistration method.

An Improved Spectral Similarity Measure Based on Kernel Mapping for Classification of Remotely Sensed Image with Kernel Mapping

2012, 41(4):  0-590. 

Asbtract ( )   PDF (2574KB) ( )  

Related Articles | Metrics

Spectral matching measure is the key technique in the classification of remotely sensed images. Most of the measures in traditional classification methods are generated based on Euclidean distance or cosine similarity which was difficult to adapt the requirements of accurately land types distinguish in multispectral data. Based on the characteristic of multispectral data, a new function called KSSV was designed in modifying the Gaussian kernel mapping by SSV matching technology. With this function the feature space of multispectral images could be mapped to high dimension space. Then in the high dimension space the old similarity measure based on Euclidean distance was replaced by SAM method. In this way the characteristic information in multispectral images can be exploited adequately and used in many remote sensing applications effectively. At last, the method was applied to unsupervised (k-means clustering) and supervised (minimum distance, SVM) classification experiments. The results show that the classification method with KSSV measure can significantly increase the accuracy of distinguishing between different land types and reduce inconsistency in one category. So the improved method can be more effective in the classification of multi-spectral remote sensing images and achieve better accuracy, and provide a new way to integrate field knowledge of remote sensing with pattern recognition methods.

Automatic construction of 3D valid solids for 3D cadastral objects based on facet sets

2012, 41(4):  0-590. 

Asbtract ( )   PDF (2909KB) ( )  

Related Articles | Metrics

EMD Threshold De-noising and Its Applications in Processing INS Errors

2012, 41(4):  0-590. 

Asbtract ( )   PDF (7192KB) ( )  

Related Articles | Metrics

Stochastic errors of inertial sensors are key factors in influencing the performance of Inertial Navigation System. Based on white noise model, wavelet threshold de-noising method is incapable of eliminating colored noise in sensor errors. Since colored noise is predominant, fractional Gaussian noise model for sensor errors is established and the model parameter estimation method by power spectral density is given. Noise variance in Intrinsic Mode Functions (IMFs) from Empirical Mode Decomposition (EMD) is derived. Noise thresholds of IMFs are estimated through variance and EMD threshold de-noising method is established. The method is applied in INS and compared with wavelet de-noising method. It is shown that wavelet threshold de-noising is poor in suppressing colored noise while EMD threshold de-noising is effective in reducing sensor errors for its connection with proper noise model. INS precision is improved through EMD threshold de-noising.

An Efficient Large-scale Point Cloud Data Management Method for Vehicle-borne Laser Scanning Applications

2012, 41(4):  0-590. 

Asbtract ( )   PDF (8499KB) ( )  

Related Articles | Metrics

Vehicle-borne laser point cloud data has become key 3d spatial information source in fields such as digital city and crisis management. Aiming at technical bottleneck of large-scale point cloud data management, this paper presents a new spatial index method–3DOR-Tree, which integrates Octree and 3D R-tree. This method utilizes Octree to forbid point-by-point insertion and generate leaf nodes of R-tree efficiently. This paper also extends R-tree structure to present levels of detail (LoD) generation algorithm of point cloud models. Finally, a data organization approach is put forward for large-scale point cloud, which easily use file mapping technique to accelerate data access. Experiments prove that this approach has fine space utilization and spatial query efficience with Lod representation capability and data cache mechanism, which lays a solid foundation for post-processing and comprehensive practices of large-scale point cloud data.

Combined 2D & 3D Indexing of Very large Point-cloud data sets

2012, 41(4):  0-590. 

Asbtract ( )   PDF (5824KB) ( )  

Related Articles | Metrics

In order to facilitate unified management of point cloud model, extract the point cloud model according to the need and improve the efficiency of visualization, a database management algorithm based on 2D&3D combined indexing of very large point-cloud is proposed in this paper. The single-view point-cloud set is managed by 2D quad tree and 3D MBB, and the multi-station point-cloud sets are indexed by 3D-R tree, and finally the organized hierarchical model and other attribute data were stored in large commercial database. The database storage, scheduling and visualization of very large point-cloud sets are achieved on the PC with representative ancient buildings’ massive point cloud from Forbidden City. Experimental results show that the algorithm is able to manage more than 10G-level data and one billion effective points with satisfactory drawing efficiency, the security and concurrency of point cloud data.