Research On Extraction Method Of Large Urban Blocks Based On OpenStreetMap

Neighborhood is the most basic unit of urban spatial structure,and it is the basic metadata for research fields such as urban planning,urban modeling,and digital urban twinning.The block geometry can reflect the block shape in the real world,and accurate block geometry data can ensure the scientificity of the follow-up research results.Therefore,the extraction of blocks is a particularly important part of urban research.The point of extracting blocks is to describe the outline shape of blocks using two-dimensional geometric shapes.Three key issues need to be addressed.First,how to extract the contour shape of real blocks.Second,how to ensure the high accuracy of the extracted blocks.Third,how to reduce the amount of computation during the extraction process and improve the extraction efficiency.The research work carried out in this paper on these three key issues is as follows:(1)Propose a block boundary screening standard in line with my country’s urban construction.Firstly,the data types in the Open Street Map dataset are analyzed,combined with the specific urban planning documents in my country,a hypothesis for the block boundary is made,which sets the data type in the dataset that can represent the block boundary.In order to verify the possibility of the hypothesis,experiments are carried out on some areas of a typical city and different data categories.After experimental verification,the block boundaries determined by the hypothesis are representative.(2)A block boundary extraction method is proposed.First,according to the existing block boundary data,in order to simplify the block boundary structure and extract the blocks more accurately,this study uses the full path algorithm to simplify the block boundary.In order to verify the effectiveness of this method,this paper conducts an experiment of extracting block boundaries in some typical areas.The experimental results show that the block boundary extracted by this method can highly restore the real block boundary structure.Compared with similar extraction methods,the method in this paper saves the extraction steps and improves the computational efficiency.(3)A block polygon clipping method is proposed.According to the simplified block boundary,this study abstracts the extraction of block blocks into a polygon and polygon overlay analysis problem.A polygon clipping method more suitable for urban blocks is proposed by improving the Weiler-Atherton algorithm.And use this method and Weiler-Atherton algorithm to do polygon extraction experiments on large cities.The experimental results show that the method in this paper can save computing time.(4)Based on the above-mentioned block extraction methods,a large-scale urban block extraction experiment was carried out in my country.Firstly,using the method in this paper,the block extraction of three cities with typical distribution characteristics in my country is carried out.Then combined with satellite images,three urban blocks were manually extracted,and the manual extraction was used as the evaluation basis.Finally,the extraction results are quantitatively analyzed by combining with the evaluation method.The evaluation results show that the blocks extracted by this method have high accuracy and reducibility and are suitable for most urban distribution in my country.