The Research And Applying Of Principle Of Radarclinometer

Synthetic Aperture Radar (SAR) allows production of geography information, especial digital elevation model (DEM), and detection of small earth motions using SAR data obtained by remote sensing satellite. How to get precise DEM is the topic research today. The recent technology and radarclinometer have become the most potential means to get DEM because of needing single SAR images.Radarclinometer is evaluated from shape from shading and based on linear integration deduced by reflectance map of optical image. The algorithm built the relation of terrain and gradation of radar image based on radar equation or Cosine Backscatter Model, and get the slope and DEM by constraint condition. The algorithm need many constraints that didn't comform the real terrain, especially omit azimuth angle that induce large error. Based on this model, we introduce a new algorithm, the region growth algorithm based Cosine backscatter model for Radarclinometry.The region growth algorithm based Cosine backscatter model for Radarclinometry that doesn't omit azimuth angle and can get DEM in mountainous area. First, the Radar image can be divided into several regions by edge information using Laplace algorithm. In one region, the image gray level changes slowly. So the variables in large image become invariables in regions. Second, in the same region, we could assume that slope changes slowly, azimuth angle and range angle are affected by their neighbor pixels, the image gray level of pixel is changed by its neighbor pixels. In the same region, we can get azimuth angle and range angle from a seed. From known point, we can get azimuth angle and range angle respectively by derivative, balance the value through iterative computation by ratio data and Cosine Backscatter Model. In the neighbor region, we assumed the region to a pixel and ignored the azimuth angle. The gradation of the pixel is average gradations of the region and the seed pixel is the nearest distance and the most approximate grey pixel. The slope and DEM of the seed is deducted by linear integration algorithm. To enlarge the thresholds of zero cross point by Laplace algorithm, we can get slope and DEM of whole image.One of the key factors of radarclinometer is the precision of potion and vslope of seed. The effective observation system isn't built in our country, so we introduce the new algorithm based on EM algorithm to get precise position and slope. First, we find the position of notable geographic characteristic in the SAR image and the contour map that determined by visualizeand Laplace algorithm. Second, the postion of seed is adjusted the distance between the ridge and notable pixel. Third we correct the position and slope of seed through iterative computation by Cosine Backscatter Model based on EM algorithm.Eventually we introduced other algorithm for radarclinometer based studying the principle of image, the closure path based Radar Equation for radarclinometer. The algorithm fully applied the feature that geography information of the terrain mapped by a pixel is invariable, that is to say, the elevation and slope of the terrain mapped by a pixel is invariable by different path and different means. But in fact, the error exists in the processes of deducing. The best deduced way is to control the errors of results to the minimum extent. Based on that, we choose the closure path to deduce and verify. We can get two different DEM and slope value of arbitrary pixel from two different path in closure path so that we can match and iterative. In this algorithm we apply simple 2*2 closure path to deduce that make the process simple and fast. The algorithm through neighbor pixel made the process feasible and the algorithm apply invariable reduced constraint conditions that promote the DEM precision. Even we apply the parameter that the radar image by equal distance slant to correct that made the precision of DEM more accurate.By applying this model, we assessed the DEM of Zhangbei of Hebei Province. Zhangbei has a lot of mountainous areas, litter water. Through ch...