摘要:Inversion algorithms are very important in quantitative remote sensing. Currently, the classic least square method is still used widely. We suggest that remote sensing inversions are often typical constrained optimization problems. Many good constrained optimization methods may be used in remote sensing. After a brief review of the constrained optimization methods, we discuss the widely used augmented Lagrange multiplier method in detail. Only one penalty factor is used in this method, even if this factor is not required to be infinitive in theory, it may still increase larger and larger to meet several constraints with very different magnitudes. As a result, similar to the penalty function method, the ill-posed problem and low efficiency still bother the augmented Lagrange multiplier method. As a solution, we extend the penalty factor to be a diagonal penalty matrix, and present an extended augmented Lagrange multiplier method. Because different constraints are given different penalty factors in this new method, a priori knowledge can be used to help decrease the ill-posed problem and increase the iteration speed. After proving this new method in theory, we do detailed simulation and inversion as further validation. It is clear from the statistical analysis that the rate-of-convergence of our method has been improved of about 30 percent compared with the original penalty factor based method but with similar accuracies. Furthermore, it is also found that our extended method is resistant to ill-posed problems.
摘要:Speckle noise in high-resolution synthetic aperture radar (SAR) images is generated by the coherent processing of radar signals and exists in all types of coherent images systems. In this paper we study the characteristics and statistical properties of speckle noise in SAR image, redefine the speckle definition, analyse the reasons about the speckle noise generation and, give the statistics formula. The experimental results indicate the statistics result is more precise than the traditional statistics result. The theory of interferometry has been used to increase resolution of the radar, but the aberrance target will appear on the SAR image thus result in the reduction of SNR reduce, while electromagnetic waves irradiate target. The interaction of the transmitted wave and echo wave as well as the transmitted wave of other objects results in speckle, which is one of the most important characters of the SAR image. SAR image doesn’t directly display the spectrum characteristic of the point target, but displays the scattering characteristic. During the movement of the radar, the interaction of the transmitted wave, echo wave and other noise wave results in distortion of the echo wave. The speckle is considered traditionally as pixels, whose breadth is more than the mean breadth variance. However the traditional definition doesn’t completely reflect the characteristic of the aberrance target in SAR image. This paper based on the characteristic of the echo wave analyses the cause of the speckle during the period of the emission, take-over, data transmission, echo demodulation and imaging process of the radar signal. According to the characteristic of the echo wave area in SAR image, this paper brings forward the definition of the speckle, and the method of analysis and statistic for the characteristic. This paper creates the mathematics model and the methods for computing the speckle by means of averaging energy of each echo wave of the target. Thus the natural characteristic of the speckle in SAR image can be accurately reflected. Our result will contribute to the research of the interferometric SAR.
摘要:The Airborne Scanning Laser Ranging-Imager Sensor (ASLRIS) is a new generation of remote sensing and mapping system for providing 3D geosciences information. The system integrates the laser altimeter, IR imager, Differential Global Positioning System (DGPS) and Inertial Navigation System (INS), by which Digital Elevation Model (DEM) and Georeferenced Image (GI) can be generated quickly and effectively without ground control points. The electric synchronization technology makes sampling-time in consistency with laser ranger and scanning imager to match height data and image data.
摘要:A method for fast resampling of remote sensing imagery has been developed based on the nature of the geometric distortion of raw imagery. The method was applied to resample SeaStar SeaWiFS and NOAA AVHRR images. Examples of the resampling of SeaStar SeaWiFS images are presented by using the results have shown that the resampling using the method suggested in this paper is much faster than other resampling methods. A weighted neighbour intensity interpolation has also been proposed. The quality of the resampled images is improved. The weighted neighbour intensity interpolation. Advantages of the weighted neighboun intensity interpolation over the nearest-neighbour and bilinear intensity interpolations have been discussed.
摘要:Lots of knowledge can be discovered from GIS database. Knowledge discovery from database of GIS has been drawing more and more attentions in different fields such as geo-science field, and computer field. ESDA technique especially emphasizes the study on the spatial dependency and spatial heterogeneity. And it is used to preliminary discover some law hidden in the data sub-aggregate of interest in the field of knowledge discovery. It is one of the most important contents to some normal whole and local spatial statistics including Moran I, Geary C, G statistics, LISA and so on. And it is the key of making ESDA technique and GIS completely and closely integrated to fully make use of GIS visualization and spatial analysis technique under the condition of imperfect Expert System. The example, which is China floating population’s spatial distribution law discovered from GIS database, shows the method and steps to discover knowledge from database of GIS with the integrating ESDA and other knowledge. The results show that it is practical and effective to discover knowledge from from GIS database with ESDA. But it is impossible to completely and automatically discover knowledge from GIS database and other synthetic knowledge must be integrated into the course of knowledge discovery. The method of knowledge discovery in the paper is used in the mode of man-machine conversation and it needs urgently to be supported by Expert System.
摘要:Image resizing plays an important role in image processing and the corresponding methods are widely used in various kinds of image processing software. Recently, oblique projection based B-Spline image resizing has been developed. Compared to conventional image resizing methods such as the nearest neighbor interpolation, bilinear interpolation, and bicubic spline interpolation, this method performs better. The resized image using oblique projection method has more continuity than using other methods.For the same reason, the resized image is smoothed at the same time, especially at the places where exist sharp edges. Thus, some important information lying in the area is lost. This result is unexpected in image processing. In this paper, a sharpness preserving method in the procedure of image resizing is proposed in order to overcome the shortcoming of oblique projection B-Spline image resizing mentioned above.In the new method, the range of integral of an output pixel is changed when there are sharp edges including in the corresponding source pixels,thus the sharpness of the output pixel can be preserved. Results show that the method performs better in both continuity and sharpness preserving.
关键词:B-spline;oblique projection;image resizing;spline space
摘要:Synthetic aperture radar(SAR)is an instantaneous imaging system which gives a snapshot of the sea surface. There exists a 180° propagation direction ambiguity problem when ocean waves are imaged by a SAR.Complex SAR ima-gery must be used to resolve the problem in addition to in situ measurements at present.A new method is developed based on the propagation theory of gravity waves in this paper.By using this method,only a single frame of real SAR image is needed.The processing procedure is simple and especially suitable for operational use.The results from a case study show good agreement with in situ measurements.
摘要:Sandy Land of West Northeast China Plain lies in the eastern fringe of huge sandy belt of middle latitude in Eurasia Continent and is also the eastern fringe of sandy desertificartion land. Now sandy desertification landscape becomes very obvious, and we are very anxious for this situation. According to TM data, the landscape patterns and desertification features are analyzed and the mechanisms of sandy desertification process are analyzed. According to the VI index of desertification degree, the dynamic change of desertification is detected based on NOAA/AVHRR data and its driving forces is analyzed. It can be concluded that desertification monitoring should be conducted based on commonly used multi-resolution RS data, and there should be a relatively uniform criterion to data, types, grades and methods for monitoring desertification so that the results can be compared all over the world. The regional differences are very obvious in landscape patterns,desertification degree, development modes in this region. In 1990s sandy desertification developed rapidly and the regional differences which are consistent with the population growth, irreasonable cultivation and overgrazing are great.
关键词:sandy land of west northeast plain;sandy desertification;landscape pattern;remote sensing monitoring;dynamic change
摘要:As the largest source of momentum for the ocean surface, wind affects the full range of oceanic motion and plays a very important part in air-sea interaction. It is recognized that remote sensing (RS) satellite has the capability of acquiring high quality marine winds with fine resolution in both space and time. Meanwhile, geographical information system(GIS) is becoming an unprecedented powerful tool for storing, processing, analyzing, and visualizing geo-spatial data. Jointly using RS and GIS technologies, a marine GIS(MGIS) system based on satellite data is developed for the analysis of global sea surface wind speed. TOPEX altimeter data between January 1993 and December 1998 which cover the oceanic areas within ±65° are used as input. An analytical model of wind speed is proposed, on the basis of which seasonal,annual and interannual variabilities (both intensity and timing) are derived. The MGIS consists of five functional modules: A database construction module, a human-computer interface module, a wind speed data preparation and analysis module, a graphic and image production module and a typical illustration management module. The system also includes three data files of model coefficient, coastal line, and typical graphics and images, respectively. A relational database structure is adopted by the system. As examples of the application of this MGIS, results of three thematic cases are presented, namely, the global marine wind climatology, the Asian Monsoon, and the El Ni*ko related interannual wind variability. These applications demonstrate the advantages of jointly using RS and GIS in oceanographic studies. We will continue to improve this MGIS by expanding its functionality and incorporating other key geophysical parameters into the system.
关键词:ocean remote sensing;marine geographical information system;Sea surface wind speed;satellite altimeter
摘要:By means of Remote Sensing, the estimation of biomass and NPP mainly is on the basis of the models of solar energy efficiency. The biomass and NPP can be calculated through the relation among NPP, APAR and the rate (ε) of which transform APAR to the organic matter. The model is:NPP=(FPAR×PAR)×[ε *×σ T×σ E×σ S×(1-Y m)×(1-Y g)] On the basis of RS and GIS, the net primary production of terrestrial vegetation of China in every ten days using the NOAA AVHRR data with five channels and 8km×8km resolution cell was calculated. And then the annual NPP was accumulated. The result shows the total NPP of terrestrial vegetation of China is 6.13×10 9t/a in 1990 and the maximum NPP is 1812.9 gC/m 2. According to this result, the spatial distribution of NPP shows a decreasing trend from south-east China to north-west China. The distribution may have close relationship with the distribution of vegetation. The NPP varies with the different vegetation. It is indicated that the NPP of conifer, broadleaf, shrubbery, grassland, meadow/morass, desert, and corn are 1031.92 gC/m 2, 1023.60 gC/m 2, 822.99 gC/m 2, 502.10 gC/m 2, 690.86 gC/m 2, 132.21 gC/m 2 and 891.18 gC/m 2 respectively. According to the yearly NPP, the NPP of shrubbery is 1621.07×10 6 tCa -1 accounting for the largest part of China (26.4%). The sequences of the other vegetation types are forest and corn, grassland, meadow and desert. The yearly NPP of them are 1490.28×10 6tCa -1(24.3%),1485.54×10 6tCa -1 (24.2%),722.83×10 6tCa -1 (11.8%),566.13×10 6tCa -1 (9.2%) and 216.13×10 6tCa -1 (3.52%) respectively. The NPP of different land use types shows that the NPP of forest, agriculture land, grassland, water area, inhabitant area, wasteland and hard-using land are 966.08 gC/m 2,897.15 gC/m 2,545.92 gC/m 2,656.28 gC/m 2,550.46 gC/m 2,299.74 gC/m 2 and 106.56 gC/m 2. According to the yearly NPP, the NPP of forest is 2096.87×10 6tCa -1 accounting for the largest part of China (34.2%). The sequences of the other vegetation types are grassland, agriculture land, wasteland, hard-using land, water area and inhabitant area. The yearly NPP of them are 1911.41×10 6tCa -1 (31.2%), 1760.57×10 6tCa -1 (28.7%), 193.96×10 6tCa -1, 125.92×10 6tCa -1, 51.62×10 6tCa -1 and 3.84×10 6tCa -1 respectively.
关键词:remote sensing;NPP;APAR;solar energy efficiency;biomass
摘要:ETM data (Aug., 1999), statistics data of 25 sampling sites with size of 4m by 4m and 1:250,000 scale DEM data. and digitized 1 to 100,000 scale distribution map of endangered rare plants are used to study the vertical and horizontal distribution patterns. Endangered rare plants group information was extracted by using Masking Principal component transformation and Hue adjust (MPH) technique. And then the feature imagery was merged with DEM data to produce a virtual reality 3-D imagery to see the differences of rare plant in relation to topographical and soil enviroment. The soil type is affected mainly by topography in this desert area. An analysis model was established to organize different data set. The endangered rare plants in West Ordos Plateau, mainly Tetraena mongolica, Helianthemum soongolicum, Reaumuria trigyna, Reaumuria soongorica, Potaninia mongolica, Ammopiptanthus mongolicus and they are evolved from Tertiary, are for a long time regarded as environment change diagnostic indicators. They also have very precious value to study environmental evolution, plant diversity and geography in global scale. These plants have also played a very important role in maintaining ecosystem in terms of preventing desertification. The vegetation spectral feature ranging from 0.63 to 1.1 μm is called reflectance “shoulder”, which is the spectral character for NDVI, BIOMASS and NPP calculations. In the study we used TM and ETM blue bands( 0.63—0.69μm) and near infrared band4 (0.775—0.90μm). To differentiate spectral features of rare plant groups the MPH technique was used. The technique was implemented in two steps. Masking, and Principal component transformation. Firstly, water bodies and some alluvium were masked out from ETM 1,2,3,4,5,7 bands, which could reduce the variance features for next step processing. Principal component transformation algorithm had the function that can compress the common features into first component. In order to get rid of terrain illumination (mainly BRDF/ALBEDO)from all imported bands, we used the PCA compress function to reduce terrain illumination. After Principal Component Analysis the first component contained most of common BRDF/ALBEDO information. Three MPH resultants, MPH 2, MPH 3and MPH 4 were selected and combined to make a color composite. The method employed in the study can be widely used as a monitorying tool for the future decision making in protection of the endangered rare plants.
关键词:endangered rare plants;sample sites;3-D imagery;analysis model
摘要:The ecosystem of semi-arid region, the neighboring zone of agriculture and grazing area in Inner Mongolia is very sensitive and fragile. It is urgently demanded to assess the status of regional ecological environment quality degradation and o find out potential environmental degradation area.In this paper the vegetation, soil brightness, moisture and heat indices are extracted and evaluated from multi-band and multi-time TM and MSS remote sensing data to describe ecosystem status. An ecological environment quality evaluation model is built from these indices, climate data (air temperature, precipitation, evaporation and relative moisture) and other geographical auxiliary information (geomorphologic types, terrain elevation and land use information, etc.), on the support of GIS. The ecological environment change during 2 decades (from 1976,1987 to 1996)of middle and western Inner-Mongolia region is evaluated using this model. The ecological environment quality degradation of this region is analyzed from the aspects of average regional ecological environment quality index change, the area change of each ecological environment quality index, and from sample band analysis. Finally, this paper also analyzes the change of climate factors and the effect of these changes on ecological environment quality change quantitatively.Through correlation analysis of ecological environment quality change and climate factor change, this paper concludes that the climate changes are the major reason that leads to ecological environment degradation. By calculation of the correlation between ecological environment quality change and change of precipitation, evaporation and moisture, which is the combination of the former two factors, this paper puts forward that the major climate factor that affects ecological environment changes in semi-arid region is moisture. During the 20 years, the effect of human behavior increases a lot. Meanwhile this research also proves that the methodology used in this paper is effective.
摘要:The area of the plane distribution of the temperature anomaly field of urban heat island of Guilin city has been determined through the comprehensive processing and analysis of the image information of the heat infrared band, the near infrared band and the visible light band of TM. On this basis, it is discovered that a special effect, karst heat corrosion subsidence on the whole ground of the city caused ty the urban heat island effect on Guilin located in the special karst environment through combining the data of the limestone corrosion under the soil layer in the test karst hydrogeologic field in Yaji village and the data of the urban air temperature in Guilin observed in the past ten years from 1987 to 1996. Study shows that the karst heat corrosion subsidence of the whole urban area is a very slow and concealed interaction process between human and the earth, it is difficult to detect only in the dependence on the methods of traditional and local observation on the earth’s surface. But at least, three types of effects on the urban environments can be qualitatively prognosticated from a long-term point of view. The first, the whole foundation of city has to slightly subside with the result of the subsidence of the whole buildings. The subsiding range is less than or equal to the vertical range of the heat corrosion. Because of the small subsiding range, it usually can’t destroy the buildings. But it may bring about a certain effect on these buildings located in some special positions of corrosion and having precise mosaic texture in the horizontal direction. The second, the thickness of the Quaternary System within the urban region has to increase. The third, it will be possible to cause the karst collapses or the faults on the ground, if a certain original and stable natural support texture of the bedrock is destroyed by the heat corrosion subsidence. For evaluating these effects, a dynamic model of karst heat corrosion subsidence and a development and evolution conceptual model of the whole ground of Guilin have been put forward. $h$, the dynamic depth of the overburden of the underlying Quaternary System under the urban mat in the process of the heat corrosion subsidence, $v$, the velocity of the heat corrosion subsidence and $t$, the necessary time for subsiding to a predetermined depth can be quantitatively calculated by using the dynamic model. And the three developing stages of the karst heat corrosion subsidence of the whole urban area of Guilin can be audio—visually shown by using the development and evolution conceptual model. Thus, it provides a theoretical and technological basis for setting up the system of the quantitative analysis and prognostication of the heat corrosion subsidence of the whole urban area of Guilin and revealing the temporal and spatial locus of the heat corrosion subsidence of the whole urban area of Guilin.
摘要:The three-dimensional GIS (3DGIS) has been developed for a long time, but with the research work goes further deep, it is found that it’s almost impossible to establish a common 3DGIS platform, because objects data modeling and its spatial relationship expression is so difficult. As city is the central place that people living and information spreading, doing research of 3-dimensional urban GIS not only has great significance, but it is also one urgent problem that people want to resolve. Thus three Dimensional City Model (3DCM) research becomes a hot area in GIS field in recent years. The 3DCM research also has great significance in traffic, terra, mine, survey, and other fields, especially in city planning, construction, and environmentology. The 3DCM research has made a great progress in Japan, Germany, Austria, Canada and China. The 3DCM researches have been done abroad, and many works are largely related to specific applications.In addition, people still have no common understanding to 3DCM concept and its research meaning. In this paper, the authors consider that the `3-Dimensional City Model’ is a suitable concept, and 3DCM content meaning should include: various 3D entities data modeling, huge data management, 3D spatial relationship expression and operation, 3D spatial data storage, manage and query, and large city area real-time visualization etc. To 3DCM theory research works, methods used in data modeling could be classified into three types: method based on DEM and images, method based on primitives and 2DGIS, and method based on 3D data structure. For data management, the majority of systems are still using 2DGIS, while 3D expression only used in 3D visualization, and few systems manage and express 3D objects using 3D expression in its fundamental. At present 3DCM mainly uses CAD to establish objects model, and 2DGIS has managed a large number of data, so the main research works focus on 3D data and texture capture, as well as 3D objects reconstruction based on images. Although 3DCM research works have got a rapid progress, many problems still exists, e.g., data modeling needs further research works. Present model could only express regular 3D entities, and it’s difficult to express complex 3D entities. At the same time, present 3DCM system could only provide animation and fly functions, and many research works still need to be done in side of spatial query and spatial operation.
关键词:three-dimensional city model (3DCM);3DGIS;landscape model