| China is a country where various natural disasters happen frequently, natural disasters such as earthquakes are sudden, massive, frequent, and dangerous, it decides that the government must establish an effective disaster monitoring and integrated data processing system based on remote sensing, telemetry values recorded automatic transmission and bring emergency management into the daily management and operation, making it an important part of daily management of the government. Remote sensing technology as the primary means of access to multi-temporal, multi-resolution, multi-types of remote sensing images and remote sensing image processing technology has been applied to the segments of various natural disasters prevention and control. Particularly in the last 10 years, with the remote-sensing technology continues to mature and remote sensing image processing improves constantly, remote sensing technology has become one of important and indispensable means of dynamic monitoring in early warning for hazards and disasters,real-time survey and assessment. Remote sensing image processing as the main body of remote sensing technology is playing an increasingly large and irreplaceable role in natural disasters, information extraction and post-disaster reconstruction.The timely and effective treatment to post-disaster information, particularly in obtaining multi-source image information timely and processing effectively based on remote sensing technology should be further in-depth, and comprehensive analysis of the use of the data to be further expanded and improved.5.12 Wenchuan earthquake happens unexpectedly; its destruction and intensity are beyond our imaging. The roads were cut, the bridges were destroyed, communications were interrupted, earthquake -stricken areas were in the status of information isolation. Telecommunications and transportations in the quake-hit areas have been severely damaged; satellite remote sensing and aerial remote sensing technology become the best ways to access the situation of a disaster quickly. In fact, in 5.12 earthquake relief, not only the multi-branch domestic scientific research strength, foreign countries including the United States, the European Union and Japan provided a large number of remote sensing data for the Wenchuan earthquake. Remote sensing images'sources were rich, types were various and each has a corresponding characteristics. In front of such a catastrophe risk, how to deal with image data and extract earthquake damage information effectively are the challenges that the sudden disaster give to application of remote sensing technology ,making it plays its proper role in post-disaster emergency and assessment and post-disaster reconstruction.It is also worthy of further study and reflection.In this study ,we analysis of the actual situation after the earthquake, disaster emergency relief, disaster assessment and the actual needs of reconstruction comprehensively , put forward a theory based on "disaster response" according to the principle of sub-level priorities to process post-multi-source remote sensing image, and establish a set of practical multi-source image processing and analysis system initially. After specific analysis of the obtained data and comprehensive results of previous studies, we propose remote sensing image processing and other key technologies and methods such as region rectification, uncontrolled rapid mosaic images ,using topographic maps and other vector pre-disaster information to deal with many types, different spatial resolution multi-temporal remote sensing image effectively based on the use of remote sensing image processing and photogrammetry technology.On this basis,by use of remote sensing image classification,object-oriented automatic identification technologies ,we extract Wenchuan earthquake disaster information effectively, determine the disaster information of landslide, debris flow, earthquake lake, collapsed town houses, cottages and schools, damaged roads, railways, farms and other disasters. Research results provide a scientific basis for the earthquake relief headquarters to accurately grasp the situation of a disaster, scientifically assess the situation and then take effective measures for disaster prevention and rescue work and reconstruction.This paper's main research and innovation are as follows:(1) Through the analysis of the source, characteristics and relevant parameter information of the multi-type, multi-temporal and the different spatial resolution remote sensing images (TM, IKONOS, etc.) .This paper use remote sensing image enhancement, geometric correction, mosaic, integration and other methods to deal with satellite image data effectively. Especially, some areas of the terrain and landscape in post-disaster is different from pre-disaster,the original topography such as topographic maps is not in accord with the objective situation, it is hard to select the necessary control points and using the conventional method of remote sensing image correction to correct local serious deformation,the study take IKONOS images as example, according to the actual situation after the earthquake, disaster emergency relief, disaster assessment and the actual needs of reconstruction, through a comprehensive analysis of previous studies results, based on the current status of remote sensing image data acquisition.This paper put forward a theory based on "disaster response" according to the principle of sub-level priorities to process post-multi-source remote sensing image,and establish a set of practical multi-source image processing and analysis system framework.according to the spatial distribution of pre-disaster vector information (such as water, etc.) and morphology, this paper use auto-matching method of vector and image to selection control point correction needed, and take local "window" region rectification methods to generate a post-digital correction image maps.(2) The image processing study of WorldView (WV) which a new high-resolution remote sensing data less eaborated. In this paper, proposed block correction method for the WorldView (WV) image processing. Because lacking of attitude parameter information of WV original image and having serious distortions problems such as distortion corrected image as a whole big. So the original image is divided into a number of small regional, after correction of sub-block image and other processing in order to improve the accuracy of image correction and provide a new way for the WorldView (WV) image correction.(3) According to the characteristics of UAV aerial images and the "disaster response" principle, This paper put forward emergency measures against the lack of image parameter, use corresponding points automatic match method to complete the uncontrolled UAV image mosaic, mosaic, produce uncontrolled mosaic responsed to the needs rapidly , and get the macro-mosaic image of earthquake-striken area, meet emergency relief needs.(4) Aerial images having orientation attitude elementary obtained by aerophotographical camera, aerial digital camera and IMU / DGPS system loaded in low-level platform. In the condition of lack ground calibration and the necessary calibration, this paper take of direct orientation method to produce digital elevation models and digital orthophoto map with three-dimensional model, regional or map sheet as the basic unit.(5) For the earthquake damage information extraction, First, Using automatic extraction methods to extract post-disaster information such as geological disasters based on processing images in image-based correction. According to the differences between body information on the disaster and its surrounding features in different types of tone and texture after the color correction processing, This paper use high-precision color aerial edge detection algorithms combined single-band edge enhancement with multi-band color composite weighted to extract disasters body information automatically. Second, extract information of earthquake-damaged buildings, using object-oriented, high spatial resolution remote sensing image.
|
PDF Full Text Request