Research On Recognition Methods Of Inefficient Construction Land Based On Remote Sensing And Open Data In Baiyun District Of Guangzhou City

With the rapid development of cities,the development of construction land has become the main form of urban land use.When the urban space is developed,the unreasonable use of construction land produces a large number of inefficient construction land.The inefficient construction land is the construction land with extensive utilization,imperfect supporting facilities and unreasonable use.The emergence of a large number of inefficient land use is a hindrance to the future development of the city.In this background,renewal of inefficient construction land has become the important way to enhance urban functions.Before the implementation of the reconstruction,the identification and analysis of the inefficient construction land is the necessary prerequisite for the successful promotion of the reconstruction,which plays an important role in improving the efficiency of the construction land after the reconstruction.The research on the extraction of inefficient construction land based on remote sensing images classifies and identifies the target mainly through spectral features,but there are still limitations in obtaining the social and economic attributes of the target through the information extraction of remote sensing data.Aiming at this deficiency,this study combined remote sensing data with Open data,including Landsat8 and all-color band data,DMSP(Defense Meteorological satellite Program)data and Google Maps data,POI(Point of Interest)data,OSM(Open Street Map)data,Baidu heat Map and Baidu road condition data.DMSP data and multispectral data were used as the basic data to extract urban built-up areas.Then fuzzy classification model based on rules is used to identify the impermeable water surface of the city.On the basis of impervious surface,distance analysis,superposition analysis and spatial assignment of GIS platform were used to extract developed construction land.The known regions were taken as learning samples and test samples,and the test samples were treated by BP neural network model for pattern recognition.The database data provided by the land management department was used as the reference for accuracy validation.The main results and conclusions of this study are as follows:(1)The identification accuracy of inefficient construction land in Baiyun district of Guangzhou is 87.7%.The number and scale of inefficient construction land are 122,with a total area of 40571331m~2 and an average land area of 332552m~2.The number of construction land in the research area is 821,with a total area of 165545747m~2.From the perspective of proportion,the quantity and area of inefficient construction land account for 14.9%and 24.5%of the total area respectively,which indicates that the identification results of inefficient construction land basically conform to the land use situation of the research area.(2)According to statistical analysis of various land use factors,factor 3(traffic conditions)of industrial land is higher than factor 1(development conditions)and factor 2(location conditions),which indicates that the development adequacy of industrial land is mainly affected by regional traffic conditions.Similarly,location factor is more important than the industrial type of land use in the business class land,focus on the development of residential land is better than other types of land use conditions.(3)Applying the identification method to Tianhe district,Guangzhou city,the accuracy of the identification of the inefficient construction land is 84.9%.The application results are basically consistent with the identification results of the study area,which indicates that the identification method is stable and suitable for urban built-up areas.(4)According to the analysis of the reconstruction functional area of the inefficient construction land in Tianhe district,the main reconstruction types of the central business district are reconstruction type and promotion type.The inefficient construction land in the cultural and educational functional area is mainly for reconstruction,and the reconstruction focus of the area is to improve the supporting facilities for scientific research.The decision makers in this region need to carry out industrial upgrading of factories and partial demolition and upgrading of rural residential areas.The reconstruction type of the inefficient construction land in the ecological control area in the north is mainly the clearing type,and the reconstruction focus in this area is to clean up the abandoned villages and factories,and restore the ecological services and supporting functions in this area.