地理国情普查数据支持下的人口空间化方法

发布时间：2018-04-28 07:33

本文选题：地理国情 + 人口空间化　；参考：《测绘科学》2017年01期

【摘要】：针对传统的人口空间化方法存在数据源多、模型复杂、忽视居住建筑类型的影响等问题,提出了一种以居住建筑作为人口分布参考的人口空间化方法。在地理国情普查数据的支持下,文章选取株洲市境内的18个乡镇为研究区。首先,将研究区内的居住建筑重分类为多层房屋、低矮房屋和废弃房屋,采用最小二乘法求算各类居住建筑的居住密度。然后,根据各乡镇人口估计值的相对误差对各乡镇各类居住建筑的居住密度进行修正。最后,运用GIS的叠加分析技术得到250m格网人口密度图。检验结果表明,人口估计值与人口统计值的线性拟合斜率为0.965,另外基于居住建筑重分类的人口空间化方法使得人口估计精度较未重分类时有了明显提高。
[Abstract]:In view of the problems of many data sources, complicated models and ignoring the influence of residential buildings, the traditional methods of population spatialization are proposed, which take residential buildings as the reference of population distribution. In this paper, 18 villages and towns in Zhuzhou City are selected as the research areas supported by the survey data of geographical conditions. Firstly, the residential buildings in the study area are classified as multi-storey houses, low-rise houses and abandoned houses, and the residential density of various residential buildings is calculated by the least square method. Then, according to the relative error of the population estimate of each township, the residential density of various residential buildings in each township is revised. Finally, the population density map of 250m grid is obtained by using the superposition analysis technique of GIS. The test results show that the slope of linear fitting between population estimation and population statistics is 0.965. in addition, the population spatial method based on the re-classification of residential buildings makes the precision of population estimation obviously higher than that without the re-classification.
【作者单位】：河南工业和信息化职业学院;中国测绘科学研究院;
【基金】：国家科技支撑计划项目(2012BAH28B03)
【分类号】：C924.24;P208

【相似文献】