Innovative Approaches to Assessing Urban Space Quality: A Multi-Source Big Data Perspective on Knowledge Dynamics

In the context of urban development, this research paper delves into the intricate relationship between urban space quality perception and the psychological well-being of residents. Urban spaces are dynamic and possess characteristics that significantly influence individuals' psychological states. This study focuses on the specific niche of understanding the influence of spatial environment quality on residents' psychological perception from a spatial perspective, challenging conventional assumptions and aligning with evolving trends in urban studies. The study employs a unique approach, combining micro-psychological perception analysis, web-crawled Baidu Maps street data, semantic segmentation using the PSPNet model for street image elements, and a novel "man-machine confrontation-iterative feedback" evaluation methodology. Deep learning techniques are harnessed for processing street images, and human-computer interaction scores are incorporated to gauge urban block space quality perception. The findings shed light on factors influencing spatial quality perception, such as green spaces, urban infrastructure, safety, and aesthetics. Furthermore, the research highlights the practical implications for urban planning and policy development. It introduces a novel "human-machine interaction and feedback" methodology that empowers decision-makers to create more resident-centric, data-driven urban environments. The study underscores the importance of community engagement in the planning process and advocates for inclusive and sustainable urban environments. This research contributes to both theoretical and practical domains, bridging the gap between advanced technology and perceptual evaluation in the urban context. It provides a deeper understanding of human interactions with urban surroundings and offers valuable guidance for building resilient and livable cities that prioritize the well-being and happiness of their inhabitants.