Millimeter-wave is close to the size of cloud particles, so millimeter-wave radar has now been constructed and operated for cloud observations. In the CloudSat Mission there will be a 3-millimeter cloud profile radar (CPR). It is very interesting to investigate the combination of CloudSat/CPR with TRMM/PR which has been in space to get the information of the vertical structure of clouds from thin to thick clouds. Thereinto, the correction of attenuation is a big problem in millimeter-wave cloud radar detection.A study of the characteristics of atmospheric attenuation at 37 and 94 GHz is presented in this paper. The primary aim of this study is to provide some basic information to the atmospheric correction for 94 GHz cloud radar. The atmospheric attenuation at 37 and 94 GHz have been calculated and analyzed in different cloudy atmospheres. The feasibility of performing the correction is discussed on the basis of dual-or triple-wavelength transmission and backscattering measurements.
Image information with different scale displays distinct spatial structure. Scale is different for each remote sensing image analysis. Image analysis with single scale could not meet the heterogeneity and dynamic of pattern and process and objects extraction in the same resolution or scale is unreasonable. To acquire the high quality data of the surface, multi-scale image analysis is necessary. Object-oriented image analysis approach offers a solution to extract different objects from various scale images. Object-oriented image analysis creates meaningful objects and builds a hierarchy levels close to surface character using multi-scale segmentation. Objects could be analysed in their suitable scale levels. The patient advantage of this method is obvious that classification is based on the meaning objects, rather on the pixels. In object- oriented image analysis“scale”is not the pixel size, but the segmentation scale threshold. This paper presents a snapshot of work using multi-scale image analysis with several resolution images. The results show the characteristics of objects multi-scale analysis, the relationship of spatial scale and resolution.
Image object levels from multi-scale segmentation are different from other images with various spatial resolutions. Different objects in surface could be represented in corresponding image object levels. Context information and texture information are very important factors for image information extraction.
Remote sensing is one of the most rapidly developing scientific technical fields in the late 20th century, which is characteristic of macro-lawexploring and micro-truthseeking. However, how to transfer the technical strategies of eyes-based interpreting to the automatic classification of remote sensing images is the main research direction of remote sensing application, thus making land use survey more objective,RS technology more quantified, and the attainment of land use messages more quick. Owing to the varied landforms and diversified land use in Southwest China, this paper aims at the establishment of a complete set of technical strategies and methods by the application of RS image automatic classification in a wide range of land use survey. Organically combining unsupervised classification and supervised classification,outside survey and expert knowledge, it will greatly improve the operating easiness and classification accuracy.
This paper explores the temporal-spatial dynamics of four grasslands (temperate meadow-steppe, temperate steppe, desert-steppe, and sand steppe) in Xilingol, Inner Mongolia using 1999 monthly AVHRR-NDVI. The averaged monthly NDVIs show that there is only minor variation for desert-
steppe in whole year. Temperate meadow-steppe and temperate steppe start to grow in May, but NDVI of the other two steppes reach their nadir during the same month. June is an important period for NDVI gain of meadow-steppe and steppe. Taking April as NDVI baseline, the net gain of NDVI in whole year is 0.42for meadow, yet by the end of June, 81% of it has completed. Similarly, the NDVI net gain for temperate steppe is 0.30, and 73% of it completed before July. The spatial dynamics are characterized by an eastwest profile. The iso-line of NDVI=0.1 is between 116°-117°E in May, and pushes forward to 113°-114°E in August when all grasslands reach their potentials. Since NDVI is very sensitive to the change of vegetation density in low covered area, the shape and position of NDVI=0.1 iso-line maybe a good indicator for long term climate changes.
Taking full advantage of existing remote sensing images, and fusing remote images with different sensor and different time by selecting appropriate algorithm, a clear-cut and abundant-colored image is made. Then we eliminate adverse infection due to fuse different time images through GIS' s overlay function combining with image' s color and field investigation. So the monitoring and analysis on the dynamic change of wetland resource in Peking Widgeon-lake from 1998 to 2000 and to 2002 were made.The result showed that: in recently years, the biggish changing in the area and special position of Widgeon-lake can be due to the durative drought and the influence of human activity. The decrease in the area of water and paddy field are always accompanied by the increase in swamp meadow and dyke. The water area and the paddy field area have decreased 16.12% and 4.85% respectively from 1998 to 2000, 44.2% and 64.87% from 2000 to 2002. At the same time, swamp meadow and dyke have increased 52.56%
and 20.16% respectively from 1998 to 2000,34.09% and 551.34% from 2000 to 2002.
Combining the method of integration development of GIS,MIS and multi-technology, a GIS-based forest field management information system is developed aiming at solving the deficiency of traditional forest field management. The paper analyses and designs the system objectives, system
hardware and software environment, system construction and system function etc. At the same time,several key technology problems such as the organization of the system data, the system integration of MapInfo and Delphi and the security design of the system are pointed out. At last, an application example is given.
This paper introduces the meanning of eco-city and city ecology system simply. Then some indexes of the eco-city costruction are estimated for the Lanzhou city. There are great achievements for the Lnazhou ecologic construction. But Lanzhou nature environment system has its shortcoming, which results in bigger costruction cost of the artificial material system and worse basis. So to construct eco-city Lanzhou would face the seletion of holiotic direction. This paper sugguests that the costruction of city and town system is the important direction of Lanzhou eco-city construction. Based on GIS, We describe the conformation of the construction of city and town system.
This paper describes the monitoring of land covers using remote sensing technique. IKONOS magery which is the world' s first 1-meter mmercialized remote sensing satellite imagery is used to obtain he land cover and its dynamic information. The methods for geometrical correction method, image fusion ethod and supervised classification method are discussed. The information of the land cover types of anji Islands is derived from these images using supervised classification method, threshold method,egetation index method and man-machine exchange method. The results show that the natural land cover ypes such as meadows and shrubbery are the main types which occupy about 70% area of the Islands, that
the accuracy of the supervised classification method is about 76.31% and the Kappa coefficient about 0.71.Paying the especial attention to the main island, its area and shoreline derived from IKONOS images are7.52 sq. km and 34.64 km, respectively. Comparing the statistics values, their biases are 1.46% and4.56%.
As the first Chinese moderate resolution spectrometer in SZ-3 spacecraft, CMODIS contains bundant spectral information with 34 channels in the range from the visible to the infrared. But sensor to ensor variation within instruments often leads to striping in many channels of CMODIS. The striping oise can distractingly and obstructively affects the interpretation and application of CMODIS data. This aper discusses the methods reviously used in striping removal of TM, MSS, MOS-B and presnts a new ethods based on moment matching. The assumption in moment atching is that changes in spatial ariability are small over the certain length scale associated with a given sweep. This assumption can be nvalid if:①an object oundary runs nearly parallel to a scan line;②an object is too small to be imaged by ll detectors within a given sweep. The Application results of the improved moment matching to non-geometrically corrected CMODIS data and analysis of the results show that the method can get the better result han the previously used methods mentioned in this paper in removing striping of CMODIS data,especially removing the disadvantage of moment matching.This method is also applicable in striping removal of othermultisensor remote sensing data.
Directly intellective selection and classification of the vegetation feature in the TM have differentmethods of dealing with remote sensing. But the presicion of selection by traditional methods is not highbecause they are influenced by many factors. The wavelet is a new method of image manipulation which isdeveloping in two or three years. In this paper, the image is transformed by the two-dimensional Wavelet andtransformed by NDVI, finally classicified by unsupervised. The result is compared to ones which isnot transformed by wavelet. The four conclusions are given. Firstly, the image which is transformed bywavelet obviously strengths the edge discrepancy of different sorts of vegetation. Secondly, the presicion
of classification of the transformed image is higher than that of raw image. Thirdly, after imagetransformed by wavelet, the results of classification relatively accord with the facts. It can be spreadextently and have a value of application. Lastly, even though the method which the image is transformedby wavelet and classified offers a new approach of selecting image feature, but it is not maturely becausethe fuctions and coefficient of wavelet are not confirmed. So the method will be more improved.
Airborne SAR could gain high-quality image during aircraft flight. Due to its theoretical advantages, SAR imaging has been widely applied in both military and civilian areas. Range and azimuth prosing are the two crucial steps of SAR imaging. The major challenges of real-time implementation with digital signal processing techniques lie in the algorithm complexity and high data rate of radar system. This paper describes a high-speed DSP system based on C67 to realize real-time azimuth compression of Range-Doppler imaging algorithm. Through optimized board structure and software pogramming, processing system' s data throughput is greatly enhanced. Both experimental simulation and field flight have shown that this system is successful.
Synthetic Aperture Radar (SAR) system simulation is the most effective and commercial way toverify the system design. Using the computer, it can simulate the radar functions and results, and bychanging the coefficient, it can achieve the result of the different terrain, different radar coefficients and
different radar operation modes. Comparing with emulation, it spends little time and cost. In this paper, aspaceborne synthetic aperture radar simulator is presented. It is based on the radar real operation, andsimulates some parts of radar system. It is approved that this approach can simulate the SAR image morerealistically.
Some primary land surface parameters needed in hydra-meteorology models can be retrieved from multispectra remote sensing data, and then the distribution map of Evapotranspiration(ET),which is very iportant in water resources management, can be made. Remote sensing retrieval of ET relates to some physical processes, and the validation and improvement of the retrieval results need to know about the techiques and methods to be used to monitor ET in situ, which is introduced briefly in this paper. Then the methods and development of retrieving ET by means of remote sensing are reviewed, especially the SurfaceEnergy Balance Algorithm for Land(SEBAL in short) is demonstrated in detail.
This paper generalized all-roundly the current study status of artificial neural network in remote ensing image classification and introduced its major application to remote sensing images classification abroad and home. For example, vegetation information extraction in remote sensing images using artificial neural network method, snowfall and rainfall information extraction in SAR images, information content extraction in mixed pixel or its decomposition by the study of multi-information content and uncertainty of object property in sample data. Compared with other classification methods, this paper also pointed out the advantage of artificial neural network method in remotely sensed images classification and made
prospects for its future development trend, including the integration of ANN and expert system, the calibrationf ANN and traditional analysis method, the fusion of ANN and geographical knowledge, the itegration of ANN and image fusion technology, and so on. Finally it is summarized that the artificial eural networkethod is an useful and effective means in remote sensing images classification. Thoughthe study and application of artificial neural network method in remote sensing image classification is stillin the beginning stage, the compositive analysis ability of artificial neural network models provided crediblenew approach for the analysis and classification of remote sensing images and other multi-source data.Artificial neural network, Remote sensing images, Classification
The Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) is an advanced multispectral imager provided by the ministry of International Trade and Industry (MITI), Japan launched on NASA' s Earth Observing System morning (EOS-AM1) platform in Dec 18th,1999.It is one of the five state-of-the-art instrument sensor systems on-board Terra with a unique combination of wide spectral coverage and high spatial resolution in the visible near-infrared through shortwave infrared to the thermal infrared regions. With its capability to schedule on-demand data acquisition requests , ASTER is expected to contribute to a wide array of global change-related application areas since its data collection in September,2000 .In this paper, the characteristics of ASTER and its advantages compared with other sensors (e.g. ETM+) were introduced firstly:①Multispectral thermal infrared data of high spatial resolution;②Highest spatial resolution surface spectral reflectance, temperature, and emissivity data within the Terra instrument suite;③The Visible Near Infra-Red (VNIR) telescope' s backward viewing band for high-resolution along-track stereoscopic observation. Then the present situation in research and application fields of ASTER data, including glacier, vocanology, geology, hydrology, agriculture, urban change, nature hazard, climate change and DEM, were summarized respectively. Finally, the future improvement in ASTER was discussed.
