Volume XLII-3
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-3, 1319-1326, 2018
https://doi.org/10.5194/isprs-archives-XLII-3-1319-2018
© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-3, 1319-1326, 2018
https://doi.org/10.5194/isprs-archives-XLII-3-1319-2018
© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.

  30 Apr 2018

30 Apr 2018

ESTIMATING HOUSING VACANCY RATE IN QINGDAO CITY WITH NPP-VIIRS NIGHTTIME LIGHT AND GEOGRAPHICAL NATIONAL CONDITIONS MONITORING DATA

X. Niu X. Niu
  • Shandong Provincial Institute of Land Surveying and Mapping, China

Keywords: Housing vacancy rate (HVR), NPP-VIIRS, Nighttime light data, Geographic National Conditions Monitoring Data (GNCMD), resident population distribution data, SVM, Moran's I

Abstract. Accompanying China's rapid urbanization in recent decades, especially in the new millennium, the housing problem has become one of the most important issues. The estimation and analysis of housing vacancy rate (HVR) can assist decision-making in solving this puzzle. It is particularly significant to government departments. This paper proposed a practical model for estimating the HVR in Qingdao city using NPP-VIIRS nighttime light composed data, Geographic National Conditions Monitoring data (GNCMD) and resident population distribution data. The main steps are: Firstly, pre-process the data, and finally forming a series of data sets with 500*500 grid as the basic unit; Secondly, select 400 grids of different types within the city as sample grids for SVM training, and establish a reasonable HVR model; Thirdly, using the model to estimate HVR in Qingdao and employing spatial statistical analysis methods to reveal the spatial differentiation pattern of HVR in this city; Finally test the accuracy of the model with two different methods. The results conclude that HVR in the southeastern coastal area of Qingdao city is relatively low and the low-low clusters distributed in patches. Simultaneously, in other regions it shows the tendency of the low value accumulation in the downtown area and the increasing trend towards the outer suburbs. Meanwhile the suburban and scenery regions by the side of the sea and mountains are likely to be the most vacant part of the city.