Cloud removal is usually one of the necessary preprocesses before the application of remote sensing images to the extraction of land surface information. Based on the formation mechanism of remote sensing images, this paper has put forward an improved homomorphic filtering method for reducing the influence of thin cloud on images and reflecting the land surface more truly. An experiment indicates that the method can achieve better results and perform more efficiently than traditional methods.
This paper has analyzed the basic theory and algorithm of the probabilistic neural network, and established the remote sensing image classification model based on the probabilistic neural network. Examples show that the probabilistic neural network model outperforms the improved back-propagation neural network model in classification precision and is close to the latter in time consumption. It proves to be an efficient image classification method.
The multiple model microwave remote sensor is an important part of the payload carried by Shenzhou 4 spacecraft launched on December 30, 2002 and is also China's first experimental microwave remote sensing system ever operating in space. The system includes three kinds of microwave remote sensors, namely microwave radiometer, radar altimeter and radar scatterometer. The radiometer instrument provides global passive microwave measurements of terrestrial, oceanic, and atmospheric variables for the investigation of water and energy cycles. In this paper the authors have described the data processing of the radiometer 1B level product and analyzed the data quality by comparing the data with those obtained by other radiometer sensors. The results show that the microwave radiometer observations of Shenzhou 4 can be used to study the variations of land surface parameters. Although it can be well compared with other satellite passive microwave sensors, the radiometer of Shenzhou 4 has its limitations.
This paper has designed a structural type identification method of the Suzhou river network picked up from the SPOT-5 image. According to the overall structural features of the Suzhou river network, some typical discrimination factors are selected, and then discrimination is made by using one or several discrimination factors. On the basis of the statistical features, it can be divided into some possible classifications and, by using other factors, differentiation can be further carried out in the resulting classification. Therefore, with the dichotomy, the types of river network can be classified. This method proves to be very effective in the structural type identification of the Suzhou river network and can attain the aim of differentiating structures of river networks.
Urban construction and development constitute the major symbols of social progress, and the spatial data of urban expansion make up an important index in evaluating urbanization. To monitor the change of the urban area, the authors employed remote sensing technology and Landsat TM/ETM+ images of 1989 and 2000 to extract the urban expansion information of Quanzhou City within the eleven years. The methods included the Normalized Difference Built-up Index (NDBI) and the image supervised classification. The spatial information was combined with the socio-economic data through GIS, which could be used in the analysis of urbanization core of Quanzhou City by means of the urbanization theory. The study reveals that the driving force of the urbanization process in Quanzhou City was the rapid development of village and township enterprises during the study period.
New methods have been proposed in this paper to improve the classification accuracy in addition to unsupervised and supervised classification procedures. In land cover classification, only the comparability of pixels is considered, while the figure and texture characteristics are not involved. In contrast, thematic information extraction can make good use of these characteristics and extract classification features more precisely. The authors therefore integrated the land cover classification with the thematic information extraction to improve the classification accuracy. During the study, three methods were selected, namely the expert system classification method for land cover, the decision tree extraction method for residential area and the iterative mixed analytical method for water extraction. These methods were tested in Shaoxing, Jiangning and Nanjing areas respectively. The gross classification accuracy of these new methods is 87.89%, which is about 7.86% higher than the common supervised classification method. Especially in fragmented plots as well as low mountain and hilly areas south of the Yangtze River, the problem of objects classification confusion has been solved effectively.
To investigate the eco-environmental destruction in the Dachang and Gaolong mines in Guangxi, the authors used Quickbird-2 & SPOT-1 remote sensing technology. It is known from the study that the fusion image formed by band 1,4,2 or 2,4,1 and band Pan of Quickbird-2 is effective in recognizing different ecological environments of the mines. Therefore, the technology can be applied to environmental investigation on the large scale and also at the 1-meter-level scale. The fusion image formed by band Pan of SPOT-1 and band 7,5,4 of Landsat-7 ETM+ can be applied to environmental investigation on the medium-large scale and also at the 10-meter-level scale. On such a basis, this paper further discusses the effects of the eco-environmental destruction in mines on the whole ecological system of the earth.
Coal fire is a dynamic process caused by spontaneous combustion of the coal layer in various directions, accompanied by the sensible thermal field variation both in time and in space. Using appropriate assemblies and sensors in different layers for effective identification, this paper has studied the technology for detection of underground coal fire. The test was carried out in Rujigou coal fire area of northern Ningxia on the basis of investigating coal fires and related physical features sensible to thermal detection. On such a basis, the effective and economical methods are proposed for the monitoring of underground coal fire.
In recent years, many researchers have been concerned with the evidence of surface processes in their study of plateau and mountain uplift as well as the evolution of orogens. At the same time, the traditional research is becoming quantified, and the key study puts its emphasis on geomorphology. Geological movements such as uplifting and denudation have led to the formation of present relief. Based on the techno-platform of Geographic Information System (GIS), in combination with the geomorphological study, this paper has constructed the Digital Elevation Model (DEM) of the Minshan Tectonic Belt (MTB) and adjacent regions by applying the data of terrain vectorization. Moreover, the authors have applied this technique to visualizing features of active structures of MTB and Minjiang Fault Zone (MFZ).
The data in this study were mainly from IKONOS images produced in USA. The methods adopted were geometric rectification and other image processing means through ENVI, manual and computer image interpretation through MapGIS and its automatic area measuring. In this way, the green area of Yueyang City was accurately figured out. This result shows that the greening ratio of Yueyang City has reached the national standard of gardenized city and that the city still has a great development potential in this aspect. The utilization of its water areas and peripheral forests seems to be a fast and economical way for further gardening this city. Yueyang City should set the strategy of becoming an international gardenized city.
Based on investigating river distribution, coastal type and shallow sea topography from multi-temporal high-resolution satellite remote sensing images, distinguishing tidal current field forms and diffusing characteristics of suspended sediments from flood-tide time and ebb-tide time temporal images, and utilizing sea charts of different periods, this paper has analyzed the bed load and its movement regularity as well as scouring-silting trend of Tong'an bay in Xiamen. The results show that the natural shore of Tong'an bay has been mostly replaced by artificial coast, that the bed load movement trend of the bay is in anticlockwise circumfluent direction, and that the beach and tidal channels are undergoing slow scouring-silting evolution with the driving factors being mainly tidal current and subordinately runoff. Since the formation of the artificial coast, the terrigenous bed load has not exerted remarkably influence on the sea area.
Starting with the accuracy of GIS spatial data, this paper discusses the factors responsible for the error of the spatial data, puts forward the method for evaluating spatial data accuracy, points out some problems to which we must pay due attention in working out the spatial data accuracy and, with some examples, explains the practical result of the study.