Daily Snow Coverage Changes Of Great Wall Station Based On CCD Images

On the background of global climate change,it will be of practical significance to monitor the climatic parameters in polar regions.At present,there is no efficient method to present the snow coverage of stations in high temporal resolution.In order to solve the problem,this paper introduces a method to calculate the snow coverage using CCD images.This research works contain the selection of the threshold value for classification,the delineation of the region of interest(ROI)and transformation of the coordinates in pixels.Finally,this method was used to calculate the snow cover ratio in the past three years based on the in-situ CCD images.For more details,as flollows:1.To analyze and eliminate the photos of all the time since the CCD camera has been put into work,and to determine the suitable CCD photos date for the long time series analysis.To select a detail photo for the specific experiment,and as for the OTSU method,the maximum entropy method,and the iterative method,describes three threshold segmentation method's principle and process of the algorithm,and use three kinds of algorithm for binarization of experimental photographs.After comparing the segmentation results of each experiment,we select the best algorithm of the OTSU method.And then in order to solve the complex features of the CCD photos,we remove the part of the ground obstacles,finally select ROI.2.Based on the theory of photogrammetry,we constructs the model of the spatial coordinate system of the CCD camera.Then,according to the coordinate transformation,we infer the expression from the pixel point of the image to the actual point coordinates.Then,according to the rules of photography,we analyse the actual ground surface,and calculate the actual ground surface area.Finally,we calculate the snow cover rate.In addition to,we analysed the angle which the camera deviating the coordinate system and obtain the error formulation,the conclusion is that the deflection angle less than 5 degrees,the error coefficient will be controlled within 0.1.Then we analyze the terrain slope error coefficient,calculating the DEM station using CryoSat-2 satellite altimeter data.The conclusion is that for the maximum slope of 3 degrees,the error factor can be ignored.3.We do some pretreatment of WorldView-2 high resolution image,including remote sensing image geometric correction and image fusion.And then we find the key points of the CCD camera in the fused image,and determine the CCD camera monitoring area by the key points.Then,we cut the image of the selected region,supervised and classified,and extracted the snow cover information,which is used to verify the snow cover rate calculated by CCD photos.The results show that they are approaching.4.We calculated the snow cover rate by CCD photos which are from March 1,2014 to December 31,2016,and analyzed the long time series of the results.And then we fitted the daily average temperature with the snow cover rate in the past three years,and analysed the trend.Finally we combined with the two results,describing the climate change of Great Wall Railway Station in Antarctic in the past three years.