顾及空间相关特征的最大熵DEM插值权系数计算  

Calculation of DEM interpolation weight modulus based on maximum entropy with self-correlation characteristics

作　　者：陈天伟[1] 康传利[1] 陈明 段青达 韦海福 CHEN Tian-wei;KANG Chuan-li;CHEN Ming;DUAN Qing-da;WEI Hai-fu(Guangxi Key Laboratory of Spatial Information and Geomatics,Guilin University of Technology,Guilin 541006,China)

机构地区：[1]桂林理工大学广西空间信息与测绘重点实验室,广西桂林541006

出　　处：《桂林理工大学学报》2020年第3期546-550,共5页Journal of Guilin University of Technology

基　　金：广西自然科学基金项目(2017GXNSFAA198308);广西空间信息与测绘重点实验室主任基金项目(15-140-07-09)。

摘　　要：针对传统格网DEM插值建模时由于主观因素的干扰,在解算权系数时存在负权现象的问题,提出了一种解算DEM插值权系数的最大熵模型算法。该方法以熵函数作为目标函数,设置非负权值和参考点数据的0、1、2阶统计矩作为约束条件,为了揭示参考点的空间自相关特征,建立最小最大权系数约束条件,通过最大化熵值来求解格网DEM插值的非负权系数。最大熵法解得权系数大小比例与点位关系相适应,在MATLAB软件平台利用遗传算法求解,并与传统插值方法进行了比较,其估值精度优于传统插值方法。Due to the interference of subjective factors in traditional grid DEM interpolation modeling,there is negative weight when solving the weight coefficient.A maximum entropy model algorithm for solving DEM interpolation weight coefficients is proposed.The entropy function is taken as the objective function.Non-negative weights are set 0,1,and 2 statistical moments of reference point data as constraints.In order to reveal the spatial autocorrelation characteristics of reference points,minimum and maximum weight coefficient constraints are established,and the non-negative weight coefficients of grid DEM interpolation are solved by maximizing the entropy value.The proportion of the weight coefficient obtained by the maximum entropy method adapts to the points.The genetic algorithm is solved by MATLAB,the estimation accuracy is better than traditional interpolation method compared with traditional interpolation method.

关 键 词：最大熵 负权 约束条件 相关特征 遗传算法 

分 类 号：O29[理学—应用数学] P207[理学—数学]