Extraction Of Usage Patterns For Land-Use Types By Pedestrian Trajectory Analysis

Research on moving objects and analysis of movement patterns in urban networks can help us evaluate urban land-use types. With the help of technologies such as global positioning systems, spatial information systems and spatial data the study of movement patterns is possible. By understanding and quantifying the patterns of pedestrian trajectories, we can find effects of and relations between urban land-use types and movements of pedestrians. Understanding urban land-use and their relationships with human activities has great implications for smart and sustainable urban development. In this study, we use the data of various urban land-use types and the trajectory of pedestrians in an urban environment. This paper presents a new approach for identifying busy urban land-use by semantic spatial trajectory in which urban land-uses are assessed according to the pedestrian trajectories. Undoubtedly, the extraction of popular urban land-uses and analysis of the association between popular places and the spatial and semantic movement allow us to improve the urban structure and city marketing system. In this regard, for semantic analysis of urban land-use, all stop points are extracted by a time threshold and they are enriched according to semantic information such as age, occupation, and gender. We examine if and how habits of using land-use types depend on qualities such as age, gender and occupation. For analysis of effects of various urban land-use types, all stop points near each urban land-use are detected. Determining what type of urban land-use cause pedestrian traffic and high absorption coefficient and what relation such high traffic has with semantic information such as age, occupation and gender. By clustering the stop points, the results indicate that stop at urban networks for each gender have a spatial correlation. Also, the results show that some urban land-use types have high traffic and we have a correlation with some semantic information such as age, gender and occupation.