基于图像熵的冰川遥感数据程序化建模

Programmed Modeling of Glaciers Using Remote Sensing Data Based on Image Entropy

【目的】提高使用冰川遥感数据进行程序化建模的模型精度,降低因遥感图像RGB色域较窄导致空间矩阵边缘干扰。【方法】提出基于图像熵算法构建90 m的DEM高精度三维模型法。首先,对遥感图像进行灰阶处理,将RGB信息转化为满足图像熵计算的灰度信息;其次,设定特定规格的判定网格,应用HALCON平台来计算各网格的熵值,匹配高程噪波;最后,采用视差融合进行程序化建模。在建模过程中应用高程噪波补偿点云来构建模型细节。【结果】由仿真结果可知,在10 m×10 m高程判断网格下,图像熵能正确反映出高程噪波。【结论】该方法能有效提升冰川遥感图像程序化建模的精度,但要注意的是网格边缘熵值容易突变,从而影响高程噪波的正确性。

[Purposes]This paper is to improve the accuracy of programmed modeling using glacier remote sensing data and reduce the edge interference of spatial matrix caused by narrow RGB color gamut of remote sensing images.[Methods]A high-precision three-dimensional model method of 90 m DEM based on image entropy algorithm was proposed.Firstly,the remote sensing image is gray-scale processed,and the RGB information is converted into gray-scale information that satisfies the image entropy calculation.Secondly,the judgment grid of specific specifications is set,and the entropy value of each grid is calculated by using the HALCON platform to match the elevation noise wave.Finally,disparity fusion is used for procedural modeling.In the modeling process,the elevation noise compensation point cloud is used to construct the model details.[Findings]It can be seen from the simulation results that the image entropy can correctly reflect the elevation noise under the 10 m×10 m elevation judgment grid.[Conclusions]Therefore,this method can effectively improve the accuracy of programmed modeling of glacier remote sensing images.However,it should be noted that the entropy of grid edge is easy to mutate,which affects the correctness of elevation noise.