The studies of the global change and the sustainable development are the two principal themes in the current geoscience researches.The terrestrial ecosystem has great significance in global change.Many of the terrestrial carbon reserves are stored in the freezing regions.However,the carbon in these regions is particularly sensitive to the global warming.Therefore it can play one of the important indicators of the climate change.As a result,the research of the soil freeze-thawing process in freezing regions and its relationship with the terrestrial ecosystem carbon cycle is significant to the global change research.In this paper,the mechanism of the interactions between freeze\|thawing process and ecosystem carbon cycle were summarized.The advances in the carbon cycle modeling were introduced and the mayor existing problems were analyzed.Meanwhile,the future research trends between freeze-thawing process and carbon cycle were put forward.In addition,a case study of the impact of the carbon cycle model considering of the freeze\|thawing process on the improvement of simulation accuracy was illustrated.
A dual-frequency rain profiling method without using a Surface Reference Technique (SRT) is described.Then,taking the valid precipitation data on 11 October 2010 in Airborne dual-frequency Precipitation Radar (APR) field campaign in Jangsu and using the dual-frequency method to retrieval Drop\|Size Distribution (DSD) parameters and rain rate.As a result,the retrieved DSD parameters are reasonable and the bright band peak can be detected in the rain profile.Moreover,the value of rain rate is similar to that detected by Spaceborne Precipitation Radar (PR).The possible errors happen in dual-frequency retrieval algorithm analysis indicated that the height of phase change and the volume ratio of liquid water may have an influence for the height lower than 4km and temperature may have an influence near the height of 3.5 km,what’s more,radar system error restrict rain rate accuracy.In summary,the dual-frequency retrieval results of APR data show the rain rate is almost consist with the actual air condition,which verifies the detection ability of APR.
The fully polarimetric microwave radiometer is a new passive remote sensing instrument,which can measure the whole of four Stokes parameters of sea surface radiation,that provides a new technique for sea surface wind vector,especially wind direction.First,the sea surface wind direction harmonics from WindSat,the first spaceborne fully polarimetric microwave radiometer,are introduced.In the basis of analysis of reason of wind direction inversion 180 degree ambiguity,the idea to solve this problem is proposed,which take advantage of two\|look capability of WindSat.Secondly,the sea surface wind field simulation scenarios are building,and the wind direction is retrieved by maximum likelihood estimation and median filter technique using the simulated WindSat brightness temperatures (Tbs) added gaussion white noise.Finally,the wind direction retrievals using between single\|look (only forward swath) and two\|look (forward and aft swath) Tbs data are compared.The analysis results show that retrievals using fore\|aft data are significantly higher accuracy,because it can effectively remove wind direction inversion 180 degree ambiguity.
Forest cover is an important forest condition indicator and a climate and hydrologic model parameter.In this study,a new statistic model was proposed,Classification and Regression Trees (CART) model is used to estimate tree cover by remote sensing.Chinese Environment and Disaster Monitoring and Forecasting Satellite(HJ) CCD data is used as main data source to estimate tree cover in Miyun reservoir upstream area by regression tree model .The estimation was validated by samples extracted from high spatial resolution aerial image,and the accuracy was compared with traditional regression models.The result shows that CART model used in HJ data has higher accuracy than regression models,with a higher R2(0.749)and lower RMSE in both modeling samples (0.068) and test samples (0.118).The study provided a new strategy for estimating forest cover by remote sensing in large area.
Sea surface salinity is one of the most important parameters in the ocean system,and it is also critical to the research of the global climate system and ocean circulation.The quality of the SMOS (Soil Moisture and Ocean Salinity) brightness temperature (MIRAS TB) is assessed and validated by comparing with simulated brightness temperature,which is computed using the default radiative transfer model implemented in the ESA (European Space Agency) L2OS processor,with auxiliary data from ECMWF (European Centre for Medium Range Weather Forecasts).Then the retrieval algorithm of sea surface salinity is built up based on the radiation transfer equation for spaceborne microwave remote sensing.The descending SMOS data of the Southeast Pacific Ocean region (45°~5°S,140°~90°W) during July 2012 is analyzed,and biases of several Kelvins are observed between average MIRAS TB and simulated TB,which are called OTT (Ocean Target Transformation) and heavily affect the accuracy of SSS retrieval.Two methods are proposed in the paper to deal with the OTT.The first is that the OTT is corrected,and the whole angular data is used for retrieval.The second is that the OTT is not corrected,and only the incidence angle range of 35 to 55 degrees are used.Then all measurements of a given Stokes parameter in a pixel are fitted to incidence angles in two order polynomial by using MIRAS multi\|angular property,in order to reduce the effect of random noise.At last,the simulated TBs computed by the guessed SSS and the forward model are compared to MIRAS TBs,and then the guessed SSS is modified in an least square iterative process until reaching the maximum similarity between both TB values.The retrieved SSS of the two methods is compared with Argo SSS and ESA L2 SSS to be validated,and SMOS SSS is collocated with Argo SSS using collocation radii of ±5 days and 50 km×50 km.The comparisons show that,the mean of the difference between the first method SSS,which is that the OTT is corrected before retrieval,and Argo SSS is 1.38 pss,and the standard deviation of the difference is 0.35 pss;The mean of the difference between the second method SSS,which is that the OTT is not corrected,and the incidence angle range of 35 to 55 degrees are used,and the Argo SSS is 0.03 pss,and the standard deviation is 0.33 pss;The mean of the difference between ESA L2 SSS and Argo SSS is 0.26 pss,and the standard deviation is 0.38 pss.So the second method can be used to retrieve SSS more accurately.
The influence and applicability of several aerosol models in the satellite image atmospheric correction are studied based on the radiance transfer simulation software 6S model usually used in atmospheric correction.The environment satellite HJ\|1 CCD sensor data in the Tianjin area under the cleaner air and severe pollution conditions is selected as a case to quantitatively estimate the influence of the selection in aerosol model for atmospheric correction based on the atmospheric correction of 6S model.By comparing the atmospherically corrected results of aerosol model with whole atmospheric aerosol parameter measured by sun\|sky photometer on ground at the same time,several conclusions show that:correction results of the continent and maritime aerosol model in 6S are closer while the result of urban aerosol model is very different from them because of its higher ratio of the soot component than the continent and maritime model;continent aerosol is more suitable for the atmospheric correction of Tianjin area under the pollution condition,while the maritime aerosol is more suitable for Tianjin area under the cleaner air.
Chlorophyll\|a concentration is an important indicator of water quality evaluation.Taking the Guishuihe River as a study area,chlorophyll\|a concentration was obtained on 25 September,2011 and 25 September,2012,and two quasi\|synchronous images of HJ\|1A satellite were acquired.This paper presents two linear empirical models and two Support Vector Machine(SVM) models to retrieve chlorophyll\|a concentration of Guishuihe River valley based on in situ collected Chlorophyll\|a concentration data and Multi\|spectral data of HJ\|1A.Determination coefficients and average relative errors are used to assess the accuracy.Models are used to retrieve chlorophyll\|a concentration,and the reasons for distribution characteristicsof chlorophyll\|a concentration in time and spatial are studied.The results show that SVM can obtain a better prediction results in the condition of small sample size for its strong nonlinear mapping ability.SVM model is more suitable for the inversion of chlorophyll a concentration.Compared to 2011,the Mean value of Chlorophyll\|a concentration of Guishuihe River increased 6.8603 μg/L in 2012,which indicated that Chlorophyll\|a concentration of Guishuihe River tends to increase.The spatial distribution of Chlorophyll\|a concentration shows that deep water areas present Chlorophyll\|a concentrations values lower than the shallow water areas,the upstream areas present higher Chlorophyll\|a concentrations values than the downstream areas.Domestic HJ\|1A CCD2 multispectral data demonstrates advantages on the dynamic changes of water quality monitoring for the 4d time resolution.
The soil line was usually used in vegetation detection,soil properties and drought monitoring.In the traditional method,the soil line is manually derived with low efficiency and large variability.Recently ,aimed to solve these problems,QIN Qiming,FOX et al.proposed an soil line automatic identification method,but it didnt take the water into consideration,especially the reservoir.Based on the analysis of their method,the method was developed by introducing the vegetation index to decrease the effect of water,and proposed the adjust soil line automatic identification method.In the comparison with real soil line,the differences is small,and the method was proved to a accurate and stable way.Used in the drought monitoring,PDI,the result showed that the method improved the effecient of the model.
Based on the water quality parameters and spectrum data measured in Taihu Lake from November to December 2007.Firstly,using the Gaussian equation to filter and disassemble remote sensing reflectivity to identify the fluorescence reflection peak after the chlorophyll a (chl\|a) 675 nm absorption peak.Then using the 662 nm reflectivity as a benchmark,adopting the normalized fluorescence height method to inverse Chlorophyll a concentration (C chl-a) and got the best inverse model.Based on Gaussian decomposition reflectivity R acquired chl\|a fluorescence reflection peak R(FP) and 662 nm at (662) ratio［R(FP)/R(662)］between C chl-a has a significant correlation .The model is the best inversion model of the fall Taihu Lake water.In high suspended sediment conditions,the model can be better expressed the relationship of chlorophyll fluorescence height and chlorophyll a concentration.This paper provides a new method and basis for C chl-a inversion,and provides a reference for the sensor sensitive band selection and settings.
Dry-hot wind is disastrous weather with high air temperature and low air humidity.Hebei province locates dry\|hot wind prone area.So it is significant to monitor the soil moisture of Hebei during the late wheat growth stage.The article uses 8 pairs of MOD09A1 and MOD11A2 products.Using enhanced vegetation index(EVI)derived from MOD09A1 and land surface temperature(LST)derived from MOD11A2,LST\|EVI two\|dimensional characteristic space was constructed,and then TVDI was extracted to indicate the top\|soil moisture of Hebei province.Furthermore,this paper analyzes the tendency and quantified relationship of TVDI and precipitation data,and find that the two groups of data showed a strong relationship.The results can provide soil moisture spatial distribution and trend of rainfall data.Combining with the meteorology data( wind speed,air temperature,et al.),we can evaluate the influence and harm of dry\|hot wind.
Synthetic aperture radar can observe the earth all day and all weather,which can guarantee multi\|temporal acquisition of satellite data.Taking Zhangpu Country of Fujian Province as our test site,this paper used ALOS PALSAR dual\|polarization data to identify land covers.To characterize the multi\|temporal behaviour of the local land covers and introduced two indexes:①multi\|temporal stability index from the multi\|temporal backscatter intensities and ② interferometric coherence from the phase information of the repeat pass data.With two temporal indexes as the primary source for classification,this study built a decision tree for land cover classification for our test site.We validated the produced land cover map of Zhangpu country with field inventory data,ALOS AVNIR\|2 image and Google Earth image.In addition,paper compared our map with that from the original backscatter intensity images with the maximum likelihood classifier.The comparison shows that our method produces higher overall accuracy (81.43%) than the maximum likehood classifier (63.06%).The result shows that the effective integration of the multi\|temporal parameters of ALOS PALSAR data can improve the accuracy of the identification of land covers.
Small mafic\|ultramafic intrusions were often omitted because of their small sizes and the large distance between two adjacent routes for geological mapping,which has seriously restricted the discovery of the relevant mineralization.This research takes Yinaoxia,east Beishan,Gansu,as the study area.First,band ratio and minimum noise fraction were adopted to enhance and identify the mafic\|ultramafic rocks using ETM+ data.Then,ETM+ data and CBERS 02B HR were fused to identify and interpret small intrusions.The results were compared with the geological map and conformed to be consistent with the geological map of the study area.Six new regions for mafic\|ultramafic intrusions were identified by synthetic analysis of remote sensing results and geological data.Through ground inspections,all the six regions were consistent with the remote sensing results.These methods are recommended for geological mapping and small intrusion\|type mineral resources targeting in the sparsely vegetated arid region of northwestern China.
As it is not well supported by existing 3D WebGIS system with respect to the effect of vector data rendering results,a geometry based approach to render vector data,which takes the advantage of the new WebGL technology,is proposed in the paper.And a optimization method for vector data rendering acceleration is also given.Then a prototype system of 3D\|rendering of vector data is developed to verify the rendering effect.At last,a test is carried out to prove the feasibility and effectiveness of the method proposed in the paper.
The three dimensional information of bridge is of great significance in building 3D digital city and virtual earth.Especially,the method of automatic and quickly extraction of bridge information from LiDAR data is a crucial problem.It is a concern of many scholars research focus.This paper presented a method of bridge extraction using airborne LiDAR data.The biggest advantage of the method is based on priori\|knowledge by analyzing the spatial structural characteristics and geometric characteristics of the bridge.Experiments show that this method has a good accuracy compared with the result of expert interpretation.
Pléiades is an optical observation system designed by National Centre for Space Studies of France.It consists of two identical satellites,that is,Pléiades-1A and Pléiades-1B,launched on Dec.17,2011 and on Dec.2,2012 respectively.Operating on a phased orbit,the Pléiades system affords a daily revisit capability on any point of the globe.And comparing with other systems,such as QuickBird and GeoEye,it has the advantage of 3 m position precision and of 20 km wider swath.Therefore,it will better address the civilian application in disaster response,precision mapping and so on.The paper explores 7 methods of data fusion,such as PCA,Pansharping,wavelet\|IHS,wavelet\|PCA,wavelet\|single band,HPF and Ehlers,to process Pléiades-1A multispectral and panchromatic images.In order to evaluate their performances,6 indexes are used,which are entropy,standard error,average gradient,deviation index,spectral correlation coefficient and UIQI.The results show that the method of wavelet-PCA and Pansharping algorithms are more suitable and can get the best results.
A novel unsupervised change detection algorithm of remote sensing images based on Nonsubsampled shearlet and Geometrical structure is proposed.Firstly,the difference image is composed of the absolute value of the difference of two remote sensing images.Then denoising algorithm based on Nonsubsampled shearlet adaptive Bayesian threshold is used to deal with the difference image to reduce the influence of noise.Finally,local geometric features and neighborhood information of the difference image are used to construct the cross\|feature vector,and then the cross\|feature vector is clustered by Fuzzy C\|means clustering algorithm.The results of clustering is change class and no change class,which are the final change detection results.Experiments show that Anti\|noise performance of the proposed algorithm is steady and effective and the proposed algorithm can get a better change detection results.
Image segmentation is the first step for object\|based image analysis.The size and quality of segmented objects directly affect the accuracy of the subsequent classification.Once the algorithm for image segmentation is determined,the choice of image segmentation parameter will directly determine the size and shape of image objects.How to choose optimal segmentation parameter is becoming the key important.The paper proposes a new goodness measure based on an inner\|segment homogeneity measurement with local standard deviation and an inter\|segment heterogeneity measure with local Moran index.The optimal segmentation parameter is chosen by discrepancy measures,including Potential Segmentation Error (PSE),Number\|of\|Segments Ratio (NSR),and Euclidean Distance (ED),were compared with this method to obtain an integrated optimal segmentation parameter.Four different categories of land cover,including cropland,grassland,ponds and buildings in a high\|resolution IKONOS image are experimented.The experiment demonstrates that different categories of land cover have different optimal interval for segmentation parameter,and the intervals derived from goodness measures and discrepancy measures are consistent on the whole.Then the optimal image segments are classified using nearest distance to means.The accuracy assessment of the results using optimal segmentation parameter are the best by comparing with classification results when using commonly selected three parameters.
The high\|resolution imaging techniques have been developing unprecedentedly in recent years.The land thematic information extraction puts forward higher request to the quality of high-resolution iamge processing.The research of registration errors influence on image processing and application is conductive to the quality control indicators determination of remote sensing data processing in the thematic information extraction.The IKONOS images wasnt acquired at the same time of north Tongzhou District,Beijing,were selected as the experimental data and images with different registration errors were produced for the simulation studies.The influence of different registration errors on the remote sensing application is analyzed from three perspectives:the image fusion,the land cover classification and the change detection,and different indices of processing quality are chosen for them.The results show that the distinguishablity of the fused image reduces with an increase in the registration error when the halved mean\|squared difference between the two image gray values rise up.If the latter increases by 3 pixels,the land cover classification accuracy will reduce by 2%~3%,and the virtual inspection rate goes up when the false land cover change information increases by 5%.Therefore,the quality of image fusion,land cover classification and change detection decrease with the increase of the registration error.
Because the UAV(Unmanned Aerial Vehicle) is light\|weighted,low\|loaded and the accuracy of its own build\|in POS data measuring instrument is not high,and the original image of the exterior orientation element is not only inaccurate,but also difficult to estimate,this paper based on the non-iterative solution of linear transformation, proposes a method of calculating UAV exterior azimuth elements that can calculate the exterior azimuth elements directly without the traditional Least\|squares Iterative Method.The accuracy of the exterior orientation elements obtained by this method is higher,and the obtained elements can revise the data of the POS at the same time.Compared with the coordinate precision of ground test point calculated by the original data,the POS data has been greatly improved.The result shows that the acquired exterior azimuth elements can be more definitely used as the initial value of the subsequent image matched,relative orientation to meet the application demands of UAV image analysis.
According to the demands of rapid maneuvering of military forces,the method of dynamic monitoring of military position based on remote sensing image is proposed.With the help of object\|oriented multi\|scale segmentation,the image of position is segmented into homogeneous objects for extracting their features,which avoids the phenomenon of “the different bodies with the same spectrum” and “content with the different spectrums”.For the trivialness of supervised classification and the blindness of unsupervised classification,through the apriori knowledge the method on making rules for the classification is proposed to identify the objects of the remote sensing images,and on this basis the results of change detection are output qualitatively and quantitatively.The experiments results show that the objected\|oriented method has high identification precision and can effectively monitor the military position.
According to the SAR image characters,this paper presents a SAR image speckle reduction method based on Bandelet transform domain of the total variation minimization.This method requires a Bandelet transform on SAR image to obtain the Badelet coefficient and geometric flow of the image;then the PCA will be used in extracting the image’s Bandelet coefficient and the principle component of the geometric flow;the threshold will be determined by the total variation minimization,and the reconstructed image will be processinged by the cycle recursion arithmetic.The experiments show that while denoising the image,this method has better preserved its geometrical structure information and has removed the pseudo gibbs phenomenon.
Modulation Transfer Function(MTF) is one of the most comprehensive and standard measure to characterize the performance of optical remote sensing payload.Estimation methods of MTF can generally be divided into three\|bar method,point source/array method,pulse method,knife\|edge method and radial\|target method according to the target.Among those methods,radial\|target method,due to its unique geometric structure,is relatively straight but more fully to characterize the spatial performance.Within this paper,the MTF estimation algorithm based on radial target is improved,and an assessment accuracy of the improved algorithm is analyzed:the main factors and their impact on MTF estimation accuracy in radial\|target method is stressed based on theoretical analysis and simulation experiment,including detection threshold,the number of detection bars,target width,target deployed angle,the random noise during image acquisition,and so on.The analysis results give some suggestions on the target deployment,and provide theoretical guidance for the on\|orbit MTF estimation based on radial\|target.
