The spatial qualities of public spaces shape the livability and experience of urban environments. A well-designed urban environment encourages people to interact, leading to a sense of comfort, interest and staying behaviors. To understand why certain urban spaces are positively or negatively perceived, a set of physical and social factors needs to be evaluated through qualitative and quantitative methods. However, since many urban quality measurement tools do not combine both subjective and objective aspects, it is often difficult to accurately evaluate the quality of streets, and the level of regeneration needed. Recent advancements in various data sources and Machine Learning (ML) have offered new opportunities to investigate the relationship between urban design features and human perception of the built environment. This paper aims to systematically quantify the effect of streetscape design on the visual preference of people in the urban streets of Alexandria. The methodology integrates spatial mapping, panoramic imageryPanoramic images collection, and AI-driven urban scene segmentation for a precise analysis of the streetscape elements. It further investigates the influence of the detected urban elements on the perceived indices of comfort, safety, pleasurability, and accessibility. Results revealed significant positive associations between safety and sidewalk, and between comfort and vegetation, underscoring the value of integrating green and pedestrian infrastructure. Conversely, the negative correlations of car and road presence with multiple perceptual indices emphasize the need to reduce car-centric planning in favor of more human-centered design. This paper contributes to the studies of quantitative methods for linking physical urban features with human perception. Streetscape perceptionIt supports the principles of urban sustainability by advancing strategies that prioritize walkability, green infrastructure, and user comfort, which are key components of healthier cities.