基于数学形态学的统计数值离散化图谱生成

Transformation from Statistical Data into Discrete Grid-map Based on Morphology

为了提供满足应用要求的统计数值空间分布图,提出结合数学形态学方法和地理信息系统技术生成数值的空间分布图谱。对数值以行政区域为单位进行网格均匀离散化处理,以网格所处的区域性质、网格内包含的建筑物区块面积所占比例,建立腐蚀或膨胀系数,并通过形态学腐蚀或膨胀操作,生成数值专题图谱。实例验证了算法的合理性、可行性以及正确性。

This paper presents the morphological method supported by the geographic information system （GIS） technology to disperse statistical socioeconomic-data of district region, and builds up the thematic map, just for the needs of drawing statistical thematic map. The procedure divides the disposed scope into the row-column grid-cells identically by the size, and each cell is simultaneously evaluated by the average value of the statistical data grouped by its containing region. The value in the matrix of coefficient for convolution is determined by the percentage of the building area in each corresponding cell, and the vicinity of a belted region containing the cell and rounding the core area, weighted by its neighbor-distance to core area. Applying the action of erosion or dilation on the prepared grid-map, it can easily build up the thematic map based on these grid-cell. Experimental result demonstrates the reasonableness and feasibility of the method.