Research On Urban Functional Land Attribute Recognition Method Based On Regional Flow Characteristics

As the urbanization process continues to accelerate,the city has gradually derived many different types of functional land such as working areas,residential areas,and commercial areas.Optimizing the urban functional structure and spatial layout is the core element of urban planning.Identification helps to grasp the urban spatial layout,improve the urban functional land classification system,promote the rational development and utilization of urban land,and help urban decision makers to formulate reasonable planning measures.Mobile phone signaling data contains the spatio-temporal information of individual travel.It has the characteristics of large data volume,low acquisition cost and high update frequency,which can provide data support for the acquisition of traffic information.This paper mainly uses mobile phone signaling data to analyze and analyze the flow characteristics of different types of functional land,and to mine the differences in flow characteristics between different functional areas,so as to realize the identification of the functional land attributes of urban blocks,and provide a reasonable Basis for decision.First of all,this article introduces the research background,purpose and significance.By searching for urban land use classification standards,urban functional area recognition,multi-dimensional time series data mining,and mobile phone signaling data-based traffic research,relevant domestic and foreign related research literatures,learning mobile communications Principles and relevant theoretical knowledge such as data mining,clarify the technical route,and lay a theoretical foundation for subsequent research.Subsequently,the mobile phone signaling data,Beijing POIs(Points of Interest)data,and Beijing road network data needed in this paper were preprocessed,and the corresponding flow characteristics indicators were selected to multi-dimensionally analyze the flow characteristics of some base stations in well-known urban areas.Through analysis and comparison,it is found that there are obvious differences in the characteristics of the flow of people between different types of areas and the reasons for the existence of the differences are explained,thus verifying the rationality of selecting feature variables.Based on the selection of the above multi-dimensional feature variables,the Beijing road network was used to divide the area within the fifth ring road of Beijing,and 1148 block research units were obtained.The spatial overlay analysis of the base station's Tyson polygonal map and the block is performed to obtain the flow feature vector of the block,and then the FCM(Fuzzy C mean)algorithm based on the feature weight is used to perform block clustering to obtain multiple clusters with distinct traffic characteristics.Different clusters represent with different types of functional land,combining the urban POIs data and cluster center flow characteristics,the functional land attribute of the cluster is calibrated.Finally,combined with Baidu map,Beijing city center land planning map and membership matrix,the recognition results are evaluated and analyzed,a certain number of samples are selected to construct a recognition confusion matrix,and the accuracy of the classification results is obtained more accurately.Overall,the proposed recognition method can It is better to identify and classify urban functional land.Based on the recognition and classification results,we analyzed the overall distribution pattern and mixed degree of functional land and briefly put forward corresponding development suggestions.This paper contains 41 pictures,19 tables,and 63 references.