Research on urban landscape accessibility assessment model based on gis and spatial analysis

• Published in: GeoJournal Vol. 90; no. 2; p. 67
• Main Authors: Tian, Yunzhi, Jiang, Yi
• Format: Journal Article
• Language: English
• Published: Dordrecht Springer Netherlands 10.03.2025; Springer Nature B.V
• Subjects: Access; Accessibility; Algorithms; Analysis; Artificial neural networks; Cities; Data quality; Economic factors; Environmental Management; Evaluation; Geographic information systems; Geographical information systems; Geography; Human Geography; Indexes; Information systems; Land use; Landscape; landscapes; Layout; Layouts; Network analysis; Networks; Neural networks; Optimization; Quality of life; raster data; Remote sensing; Social Sciences; Socioeconomic aspects; Socioeconomic factors; Socioeconomics; Spatial analysis; Spatial data; standard deviation; traffic; Transportation networks; Urban areas; Urban environments; Urban planning; Path optimization; GIS Technologies; Spatial analysis techniques; Accessibility assessment model
• ISSN: 1572-9893; 0343-2521; 1572-9893
• DOI: 10.1007/s10708-025-11310-y

The accessibility of urban landscape is an important index in urban planning and design, which can effectively affect the city’s functional layout and the residents’ quality of life. With the rapid growth of urban data, traditional accessibility assessment methods face many challenges. This study proposes an urban landscape accessibility assessment model based on geographic information systems (GIS) and spatial analysis. The model combines the path optimization algorithm of network analysis with the spatial analysis method of raster data and realizes the comprehensive analysis of urban traffic network and geographical features. Urban landscape accessibility ensures equitable resource distribution, sustainable urban planning, and efficient land use. With the increasing complexity of modern cities, traditional methods of accessibility analysis often struggle to address intricate urban layouts and multifaceted socio-economic factors. This study leverages Convolutional Neural Networks (CNN) integrated with Geographic Information Systems (GIS) and spatial analysis to propose a novel model for urban landscape accessibility assessment. The model significantly enhances computational efficiency and accuracy by utilizing CNN’s capability to process high-dimensional spatial data. In the urban landscape accessibility assessment, after calculation, the average accessibility index of a certain area is 582.3, and the standard deviation is 3.15, reflecting the high consistency of the transportation network within the area. At the same time, the maximum accessibility index reached 482.51, indicating that some areas have significant advantages. The overall accessibility improvement was 7.22%, showing the effectiveness of the evaluation model in optimizing traffic paths.