The urban green space is an important place for daily activities and contributes to the physical and mental health of people. Accurately analysing the complex morphological characteristics of green spaces and exploring their influence on crowd activities can help designers and managers optimise and manage these spaces, as well as enhance attractiveness and service effectiveness. However, current research on people-space preferences often overlooks the continuity and integrity of landscape spaces. Guiding overall optimisation based on local elements and point analyses may lead to discrepancies between expected and actual outcomes. To address this issue, this paper proposes a refined method for analysing spatial morphology and people’s distribution, which includes data collection and processing, line-of-sight simulation and 3D morphology analysis, and mapping to compare people's distribution with spatial morphology. Using an urban green space as a case study, the research demonstrates the required data, numerical techniques, and workflow. It also illustrates calculation methods for enclosure and edge degree based on line-of-sight simulation. The findings reveal that crowd activity distribution exhibits spatial differentiation and preferences for specific spatial patterns. The method has potential applications in establishing people-space relationships, predicting visitor distribution, and guiding human-centred green space renewal and optimization.