Precise harvest is an important part of precision agriculture.The study first proved the importance of accurate crop mature date prediction by analyzing the influence of harvest date on crop yield and grain quality,then provided a review on crop mature date prediction through meteorological statistical method,crop growth model method and remote sensing method.After that,based on the development of remote sensing derivation of crop maturation indicators,this paper concluded that it is feasible to predict crop mature date and map fields harvest order with satellite-based remote sensing,especially when new sensors are launched continuously.The study also pointed out that the issues will be addressed in the field:(1)developing remote sensing oriented crop maturation indicators and studying their varying patterns;(2)improving remote sensing estimation of crop canopy chlorophyll and water content;(3)developing high-temporal & high-spatial resolution remote sensing data generation technology to implement field scale dynamic monitoring and;(4)improving mature date prediction by integrating different models.
This study presents an evaluation of two split-window algorithms applied to MODIS data for LST retrieval,including QIN and Wan-Dozier split-window algorithm.The absolute accuracy and the total errors of two algorithms were compared and analysed using data simulated by MODTRAN 4.0 code with TIGR data input.The results show that the bias of the absolute accuracy and the total errors between two algorithms were both small in all sample data,however,the absolute accuracy of Wan-Dozier algorithm was more effected by the change of input LST and water vapor than that of QIN algorithm.The main error sources for both two algorithms were the absolute accuracy and the uncertainty of the land surface emissivity.Example regional LST produced by two algorithms using MODIS data was compared.The results show that great consistency of derived LST exists between the two algorithms in the study area.The characteristics of LST spatial distribution pattern can be clearly identified.Larger LST bias of the two algorithms is found in regions covered by water and bare soil,whereas the mean bias is within 0.5 K in town and vegetation surface.
In the study,we have utilized daily ascending(1:30PM) and descending(1:30AM) brightness temperatures to monitor the near surface freeze/thaw states.The dual-indices algorithm has been used to identify the near surface freeze/thaw states twice a day,which can classify daily surface freeze/thaw cycle into daily frozen surface,daily thawed surface,daily surface freeze/thaw cycle and daily surface inverse freeze/thaw cycle.This study calibrates the thresholds of descending and ascending orbit 37GHz vertical polarization brightness temperature,respectively,by in situ daily maximum surface temperature and daily minimum surface temperature of 28 meteorological stations,and another 122 meteorological stations in China are used for algorithm validation.The classification accuracy of daily frozen surface is above 90%.The classification accuracies of daily thawed surface and daily surface freeze/thaw cycle is around 70%.On the basis of daily freeze/thaw state,annual surface freeze/thaw cycle is divided into stable frozen period,stable thawed period,spring freeze/thaw transition period and fall freeze/thaw transition period.According to results of 2004,in January,daily frozen surface and daily thawed surface is roughly bounded by Tsinling Mountains and Huai River.Boundaries of daily frozen surface south,daily thawed surface north and daily freeze/thaw cycle south/north move from southeast to northwest of China,with increasing surface temperatures,and then move back for decreasing surface temperatures.Stable frozen period mainly appears in January and February,and stable thawed period is in July and August.Spring freeze/thaw transition period occurs during March to May,and fall freeze/thaw transition period in September,October and November.
It is promising way to analyze large scale surface energy balance from two source model,which widely used to map local evapotranspiration pattern,detect regional drought and climate change study.Based on thermal infrared remote sensing image and field meteorological parameters,the Atmosphere-land Exchange Inversion Model (ALEXI) can examine land surface process in continental scale,further the hourly datasets from geostationary platform make possible integrate field measurements and model results.However,such great capacities of ALEXI model have weakness in its validation results that involved in the inappropriate spatial scale between satellite images and field measurements,even Large Aperture Scintillometer (LAS) measurements can’t meet the heterogeneous area comparison requirements in 6 km×6 km area.In this study,we assumed that the surface energy balance algorithm for land model (SEBAL) from high resolution image could catch more reasonable regional surface energy balance pattern,which was verified by many studies in local scale.The land surface balance results from SEBAL model based on Landsat images were considered as the data source for ALEXI model (MTSAT satellite platform) validation in the same period.The study showed that ALEXI model and SEBAL model can get a more consistent surface energy exchange patterns,and statistical analysis of the ALXEI model also provided well distribution among different land cover.Combined with geostationary satellite data,ALEXI model can be a promising method for monitoring land-atmospheric interaction.In addition,this study was executed in small watershed over northern China,and the quantitative validation of ALEXI model also need further work to improve the model accuracy in more heterogeneous study area.
This paper used crossover data of Jason-1 altimeter after applying the least square method to analyze the effects of two inverted barometer model correction results on sea state bias (SSB).The results demonstrated that the SSB deriving from the conventional inverted barometer model and MOG2D inverted barometer model had better consistency in most domain of significant wave height (SWH) and wind speed (U),but there existed distinct differences in such strong sea state as SWH>6 m,U>10 m/s,and the maximum was 2.6 cm.The RMS of crossover differences in the certain strong sea state domain was calculated,and the value of MOG2D correction result was smaller of 1.4 mm than the convention IB correction result.Besides,the explained variance of SSB derived from MOG2D Model was larger of 3.14 cm2 than the convention IB Model.Therefore,MOG2D correction result is more effective than the conventional IB model to establish SSB parametric model and its effects in the strong sea state cannot be neglected.
Based on Neural Network (NN) algorithm,this paper estimated the Leaf Area Index of plantation forests in HEIHE area northwest china using the ENVISAT/ASAR data.Firstly,the relationship between ENVISAT/ASAR microwave backscattering coefficients (σ°) with different polarization,incidence angle and Leaf Area Index (LAI) of White Poplar and Desert Date planted forests were analyzed.The study showed that the homogeneity of plantation forests was the primary factor influencing the relationship.And there were significant difference in the relationship in different incidence angle conditions.Based on the analysis above,the NN algorithm for LAI retrieval was designed using ENVISAT/ASAR with different polarization,incidence angle as input parameters.The study compared the ground measured and predicted LAI for validated data,training data and all data.The determination coefficient reached 0.61,0.91 and 0.82 respectively.And the results showed that there was great feasibility to estimate LAI of plantation forests by NN algorithm using ENVISAT/ASAR data.
Longpaozi of ZhaLong marsh is taken as the test area.High spectrum and water depth data were collected,based on which water depth inversion models are established.According to the correlation coefficients between the first derivative of high-resolution reflectance and water depth,the best wavelength for inversion factor are selected to build single band model and multiband model.The best model for water depth included as predictors Band 832.05 nm,Band 839.87 nm,Band 809.08 nm,Band 774.76 nm,which has RE of 5.90% and RMSE of 10.869 cm.However,many complex factors influence the model accuracy in wetlands,so the further research will focus on spectral features of the water impurities and sediments to improve the models.
It’s worth to investigate that the false alarm is reduced with polarimetic information while keeping the detection ratio when the fully polarimetic SAR images used for ship detection.According to the nonnegative and sparseness characteristics of eigenvalues,an improved S-NMF (Sparseness-Nonnegative Matrix Factorization) based on eigenvalues is proposed in this paper,which is used for ship detection for SAR images and reduce false alarms effectively.We use the first two eigenvalues to consist matrix for NMF decomposition since they take most of the energy of the targets and could maintain the shapes of the targets,while the sparseness is decided by the clutter distribution,the decomposition result is regarded as Result I which contains all the ships with some false alarms;while the third eigenvalue which could get rid of the false alarms,such as “ghosts” caused by the ship movement,noi,etc.is used to multiply the Result I to get the Result II which could achieve higher TCR(Target-Clutter Ratio).Then the OS-CFAR method is used for ship detection,and become the final detection results.In this paper,this detection algorithm is validated through Radarsat-2 fully polarimetic SLC images of South Sea of China and the corresponding AIS data,and compared with the SPAN method,HV channel image and PWF method.The detection and comparison results show that the proposed method could not only detection all the ships correctly,but also reduce the false alarms effectively,and has high robustness for non-affected by changes of the detection window.
This paper aims at the research of the Doppler centroid real-time estimation technology of space-borne SAR based on GPU.In this paper,the complete system structure is especially discussed,the Doppler centroid estimation algorithm based on the CUDA programming model is presented,and the MLCC algorithm is designed optimally.In the end,Tesla C1060 is selected to build the GPU dedicated scientific computing platform for Radarsat-1 data processing.Experiments show that the processing rate of the Doppler centroid estimation is about 12 times as high as the Radarsat -1 satellite downlink rate without considering the global fitting of data processing.This study provides the technical basis for Doppler centroid real-time estimated accurately and real-time SAR imaging processing system manufactureed.
NDVI time-series data contain disturbances that limit their use and even yield false results.Although the adaptive Savitzky-Golay method could effectively filter some sudden fall - noisy data which is assumed traditionally to be contaminated by clouds or poor atmosphere conditions,it cannot preserve some sudden fall data with good quality,and cannot suppress the sudden rise noisy data.Although maximum NDVI values greatly reduce clouds and aerosols,the highest NDVI value does not necessarily correspond to small sensor viewing angles or to the least-contaminated measurement.This paper presents a VI-quality-weighted Savitzky-Golay method which is based on the Savitzky-Golay filter and weighted by VI qualities derived from MODIS VI product.The results illustrate that the quality-weighted methods could filter more noises,especially sudden rise noisy data,effectively preserve high-quality data and meanwhile do not sensibly elevate the values of the whole time-series.It can appropriately fit high quality data among serious fluctuations and better reconstructs wave crests compared with the traditional Distance-weighted Savitzky-Golay method.Statistically,the proposed method here has the following characteristics:(1) it has lower mean variation (or less shift) effect on original NDVI data;(2) it stabilizes high quality NDVI data;and (3) the resulting high quality data are better correlated with original good data,meanwhile the original noise are greatly decorrelated.
Tower cranes,as landmarks of construction sites,are alerts of illegal buildings and powerful basis of engineering process.H〖JP2〗owever,it’s very difficult to extract tower cranes exactly,because the background objects in urban remote sensed images are complexed and the influence of noise is severe.Moreover,there have not been any algorithms available in detecting tower cranes.The different tower cranes may have the influences in multi-directional and multi-scale structuring elements on the image,a method to detect tower cranes based on mathematical morphology with self-adaptive weight is put forward in this paper.It is adopted to do segmentation with Unmanned Aviation Vehicle Remote Sensing (UAVRS) image in our study.The weight is determined automatically based on the filling times that the structuring element has processed the image.Binarization,edge detection,line extraction and finally recognizing tower cranes precisely based on their specific geometrical characteristics are conducted.It has been revealed that this method can be feasible to segment aviation images and extract relative features.Moreover,it has important applications in evaluating investment in urban construction.
Hyperspectral remote sensing images are high-dimensional and non-linear,so information loss and distortion are easily caused by linear feature extraction.Based on the minimum noise fraction transformation (MNF) which is a hyperspectral remote sensing image linear feature extraction algorithm,kernel minimum noise fraction transformation(KMNF) is proposed by introducing the kernel method,so KMNF is a non-linear feature extraction method.Samples are mapped into high dimensional feature space through a kernel function,MNF is conducted in feature space,thus non-linear KMNF algorithm in the original space is realized.And hyperspectral remote sensing image feature extraction based on KMNF was carried out,the effects of sample amount to KMNF were analyzed,it was found that the sample number influences KMNF slightly,a small number of samples can get almost the same result as a large number of samples.KMNF and MNF feature extraction results were compared,and their dimension reduction efficiency and remaining information were analized,and it was found that KMNF can get almost the same dimensional reduction efficiency as MNF,and KMNF can extract non-linear information from hyperspectral remote sensing images.Using SVM for hyperspectral image classification based on KMNF and MNF,it is showed that classification accuracy of KMNF and SVM is higher than MNF and SVM.
Eco-hydrological Wireless Sensor Network in the Heihe River Basin can be considered as a successful application of wireless sensor network technology in the field of near-surface observation.This paper introduces a quality control approach for a wide variety and long-time massive data,especially for Heihe WSN observation data.Owing to the potential sources of systematic errors of wireless sensor network,there are some quality problems of the WSN data.Therefore,an adequate data quality control technique is of great importance.Firstly,the spatial and physical characteristics of WSN data were analyzed,and the definitions and implementation algorithms of WSN data quality elements were put forward as well.The algorithms involved timeliness and consistency assessment,data fragmentation processing,outlier detection,data redundancy processing,etc.Furthermore,there were two quality assessment cases were provided,and each one consisted of appropriate amounts of quality elements algorithms.The cases demonstrated that the proposed approach has the advantage of high efficiency and reasonable result,and played a vital role in the assignment of quality control for Heihe WSN observation Data.
A speckle noise filtering algorithm based on iterative Gaussian model is proposed.Firstly,the Noise Marker Matrix (NMM) is calculated using the DEM data,where the pixels are not fit the Gaussian distribution were marked as noise.And then,the true height value can be estimated using the quadratic surface fitting based on the statistical properties of its neighborhood window.Finally,the proposed approach is carried out by InSAR DEM data.The results on actual DEM data show that the proposed algorithm can filter the noise data without disturbing the correct data.In comparison,the filtering results of the low-pass filter,the median filter and the sigma filter algorithm,the proposed algorithm can get the best filtering result.
Cloud cover is an important factor that degrades the quality of remote sensing images.Generally,traditional cloud detection algorithms can not work effectively in scenes such as mountains,snow and dark clouds,thus resulting in lower detection precision.In this paper,we analyzed the difference of characteristics between cloud area and earth object,then defined a new edge feature descriptor,and lastly proposed a new cloud detection algorithm based on image block classification.The edge features and gray features are extracted and classified by the AdaBoost classifier,and the result is corrected by using spatial neighbouring correlation.More than 100 thousand image blocks experiment shows that this algorithm has much better performance than traditional algorithms.To be specific,the accuracy is more than 96% and the operation is so fast that real-time requirement can be ensured.
Remote sensing plays an important role in monitoring greenhouse gases emissions and the source and sink of greenhouse gases at regional and global scales.GOSAT (Greenhouse gasesObserving SATellite) is the first satellite for space borne measurement of the main greenhouse gases CO2 and CH4.An Introduction is made about the emission background,satellite platform,instrument characteristics,and the ground systems of GOSAT,in order to thoroughly understand the advances of satellite greenhouse gases observation in the world.This paper also presents an overview of GOSAT data products,calibration and validation strategies.Thermal And Near infrared Sensor for carbon Observation Fourier\|Transform Spectrometer (TANSO\|FTS) is based on the Michelson interferometer,and combine with several observation modes which provides atmospheric greenhouse gases concentration and profile data with high precision.While TANSO-CAI (Cloud and Aerosol Imager) is a radiometer with Ultra Violet (UV),visible,and short wave infrared (SWIR) bands to reduce the interference of cloud aerosol on greenhouse gases measurements.The satellite design and data applications of GOSAT provide important references for developing greenhouse gas monitor satellites in China.
Sequential data assimilation methods have been widely applied in many data assimilation systems and each method has its own characteristics.In this paper,we introduce three typical assimilation methods,for example,Ensemble Kalman Filter,Ensemble Transform Kalman Filter and Deterministic Ensemble Kalman Filter.Based on the classical nonlinear model (eg,Lorenz-96 model),the numerical experiments were developed to test the sensitivity of all these methods.Different key parameters were investigated with respect to four aspects,which were the number of ensembles,the number of observations,the inflation factor and the localization radius.The results show:the number of ensembles and observations will directly influence the assimilation results; the optimal selection of the inflation factors and the localization radius will improve the accuracy of the assimilation.According to the final comparative studies,the deterministic EnKF is a method that has a better robust performance.It can achieve a better assimilation effect,especially in the occasion of the small ensemble numbers.
The Compact Airborne Spectrographic Imager(CASI)hyperspectral image was used to finely classify the plant structure of the vegetable base in Ganzhou District,Zhanye City.The image was strictly pre\|processed by using the field spectral measurement data and CE\|318 data.The raw spectral information and calculated NDVI,textural features information were used in the classification and four classification schemes,and the Support Vector Machine(SVM)was adopted as the classification method.The aim of this paper is to explore the application potentiality of CASI in crop fine classification of the plant structure.The results indicated that CASI data contains abundant information,it thus has a great potential in the vegetation fine classification.The overall classification results of four schemes were consistent with ground investigation measurement data.The classification accuracies were 96.78%,97.21%,88.00%,88.38% and Kappa coefficient were 0.9676,0.9691,0.8674,0.8716 respectively.The vegetation sophisticated category could be done well by combination of CASI ,spatial information and NDVI,which is an effective approach to improve classification accuracy.
Remote sensing data sets from GIMMS NDVI,VGT NDVI and MODIS NDVI/EVI have been widely used in the studies related to vegetation change.Many previous studies have evaluated the consistency among these different data sets.However,less attention has been paid to study of the influence of seasonal cycle of vegetation,which probably amplifies the relationships among different data sets.Interannual variability of vegetation and the influence of the seasonality of vegetation on the consistency of different remotely sensed datasets had been investigated using GIMMS NDVI,VGT NDVI and MODIS NDVI/EVI products in eastern China from the period 2000 to 2006.The results indicate that all of these datasets show similar trends of the interannual variability of vegetation,whereas,the seasonality of the vegetation data overshadows the difference of different remote sensing data sources.The vegetation indices derived from GIMMS,VGT and MODIS show significant correlation before removing the seasonality,while the relationships among them decrease obviously after removing the seasonality.In general,the consistency of GIMMS,VGT and MODIS still remains in the northern part of eastern China.The differences among them mainly occur in the southern part.In particular,the most rapid decrease of correlation occurs between GIMMS and MODIS.The correlation between them has not been observed in large parts of the region located south of the Huai River.
The estimation of soil moisture in Maqu in northeast Tibet Plateau on August 9 and October 6 2009 is carried out using the European environment satellite (ENVISAT) Advanced Synthesis Aperture Radar(ASAR) APP mode of HH and VV polarization active microwave data.Surface microwave backscatter geometrical optics model (GOM) is used in the bare soil,the effect of the vegetation layer can be eliminated from the backscatter coefficients by a semi-empirical “water-cloud” model,which have achieved good results.The RMS error R between estimated soil moisture using remote sensing and ground measurements is less than 0.05,while the coefficient of determination R2 is greater than 0.82,which indicates that this method is suitable for the inversion of the soil moisture in Maqu area.The inversion results of remote sensing show that shadow valley and the steep sides of mountains are relatively poor,and the inversion results in relatively flat area are good,and the value of the estimated soil moisture are mostly between 0.20 and 0.50 m3/m3.
Vegetation mapping based on EVI time series matching have been widely used due to the usefulness of EVI datasets in distinguishing phenological differences,and many scholars have made great results.However,most studies just made the ideal time series profile from the unclassified image,and research on constructing the ideal profile from datasets of previous years is even less.This study used the time series Moderate Resolution Imaging Spectroradiometer (MODIS) EVI datasets with intermediate spatial resolution(250 m),high temporal resolution ( 16 day composite period) ,and cost free status,made use of Savitzky-Golay method to reconstruct the EVI time series image,odopted the dataset between 2006 and 2010 to create an ideal EVI time series profile of cotton and collected the plant distribution of cotton in 2011 by comparing the Euclidean distance between the ideal profile and the time series EVI profile of unclassified pixels.This study also used the cotton plot surveyed in 2011 as region of interest,checked out the accuracy of the plant distribution of cotton in 2011 using confusion matrix.The result is that the overall accuracy is 82.31%,demonstrating that collecting the plant distribution of cotton using the ideal profile created by the datasets in pervious years is feasible.
For the demands of road ecological environmental monitoring,according to road construction and ecological characteristics,we studied on 8 road constructed recently considering on phase characteristics,spatial characteristics,data characteristics and price.Phase characteristics selection includes various phases of road construction,local vegetation conditions and coordination of various phases.Spatial characteristics selection means different scales by different resolution.And data characteristics are combined with visible light data,multi-spectral data,cloud cover and light reflected factors.Consideration of price can make remote sensing data used more effectively.Through these studies we could provide scientific judgments about appropriate remote sensing data selection for road ecological environmental monitoring research.
Water resource of Jilin Province were uneven distribution in time and space,which led drought had been one of the largest natural disasters in Jilin Province.Therefore,it was urgent to strengthen the monitoring,forecasting and damage assessment of drought,while Remote Sensing technology had more irreplaceable advantage than other technical means.LST-NDVI feature space was established with Land Surface Temperature and Normalized Difference Vegetation Index which were from MODIS products.According to the feature space,models were built to calculate the temperature vegetation drought index,which was a drought situation indicator.This paper studied the drought space-time distribution characteristics of Jilin Province in 2005 through temperature vegetation drought index,which was validated by soil moisture.The results showed there was significant correlation between temperature vegetation drought index and soil moisture,and this method could be used as a reference in the drought monitoring of Jilin Province.We found that drought distribution of Jilin trend from southeast to northwest,presenting from moist to normal,light drought,drought and severe drought; On August 19,August 25,and September 8 of 2005,ratios of the normal and light drought distribution area to the total area were 26.84% and 59.53%,41.31% and 41.73%,40.40% and 32.83%,respectively.The distribution of drought and light drought in mid-September was the most widely,which were 38.27% and 36.26%,respectively; Severe drought and drought were mainly distributed in Baicheng and Songyuan; Light drought mainly distributed in Changchun,Siping and Liaoyuan;The normal drought concentrated in Jilin,Tonghua and Baishan,and moist mainly distributed in Yanbian.
The overlying iron cap formed from Limonite with the thickness of 0.5~9 m of bauxite is an important characteristic of bauxite in study area,which makes the bauxite altered information can be represented by iron cap information in a certain degree.Based on the feature analyses of ore\|contained system,spectral and remote sensing images,utilizes the correlation between iron cap information and bauxite altered information to extract bauxite altered information indirectly by extracting iron cap information is proposed.This paper presents a method of “multivariate data analysis + ratio analysis + principal component analysis + classification” to extract bauxite altered information.Test results show that the coincidence rate with given ore bodies and bauxite altered information extracted by the presented method is 63%,while the extraction accuracy of principal component analysis and ratio analysis is 44.6% and 54.7% respectively.The extraction accuracy is much higher than that of principal component analysis and ratio analysis.The method proposed in this paper can extract bauxite altered information in Kindia,Guinea rapidly and effectively by inhibiting vegetation coverage and Quaternary effectually.With the help of the presented method,the distribution of bauxite in the study area is verified preliminarily,which provides effective instruction information for the further prospecting and exploration.It is of great realistic significance for overseas exploration area with high cost of ore exploration.
The hydrologic information is the foundation of the hydrologic researches.At first,this paper reviewed current hydrologic information systems and hydrologic data sharing systems all over the world.Based on the spatial and observation data collected in the Heihe River Basin,a typical inland river basin in the northwest of China,hydrologic data sharing and on-line services were implemented with the help of open source HIS(Hydrologic Information System)from CUAHSI (Consortium of Universities for the Advancement of Hydrologic Science,Inc.).The specialized HIS for the Heihe River Basin adopted the ODM (Observation Database Model),WebGIS and Web service and integrated watershed science data from the Heihe River Basin.This established hydrologic information sharing system functions with hydrologic observed data retrieval and browsing,as well as on-line data service and analysis demonstrated by an application case in the Heihe River Basin.
The combination of three Dimension (3D) visualization and spatial analysis has become the development tendency of virtual reality research.Therefore,we try to proceed an organic combination between GIS software and professional modeling software in this study.This study is of great significance for the realization of 3D scene visualization,scene dynamic roaming,information query and spatial analysis and so on.The southern campus of Northwest Agriculture and Forest University was used as an example in this study.Integrated with Google SketchUp 3D modeling technology,Geodatabase data models of ArcGIS9.3,3D application environment and the spatial analysis functions of ArcScene,using secondary development technology of ArcGIS Engine 9.3 and Visual Studio 2005(C#),the 3D campus geographic information system was well designed and accomplished,and further discussed the specific technology and realization procedure.The system has many functions,such as layers management,symbols design,edit management,spatial query,optimal path dynamic analysis,length and area measuring and thematic mapping of two-dimensional data,etc.In addition,3D models system could realize importing and exporting,layers management,3D scene browsing (including navigation,flying,zooming,roaming,etc.),spatial query,measurement and the interactivity with two-dimensional vector.The design and Google
