Zhiyi Dou

Landslide hazard assessment increasingly demands the joint analysis of heterogeneous remote sensing data; however, automating this process remains difficult due to the pronounced resolution and texture discrepancies existing between satellite and aerial sensors. To address these limitations, this study proposes a robust segmentation framework capable of extracting sensor-robust representations. The framework leverages a DINOv3 transformer encoder and exploits representations from multiple transformer layers to capture complementary visual information, ranging from fine-grained surface textures to global semantic contexts, overcoming the receptive field constraints of conventional CNNs. Experiments on the Longxi satellite dataset achieve a Dice coefficient of 0.96 and an IoU of 0.938, and experiments on the Longxi UAV dataset achieve a Dice coefficient of 0.965 and an IoU of 0.941. These results show consistent segmentation performance on both the Longxi satellite and UAV datasets, despite differences in spatial resolution and surface appearance between acquisition platforms.

The spatial distribution of urban commercial spaces significantly impacts the overall efficiency and vibrancy of adjacent neighbourhoods. As such, it is an important factor to consider during urban development. This study aims to examine the patterns of impact between commercial distributions and neighbourhood dynamics in a residential neighbourhood, based on the case study of a highly populated, thriving commercial, and culturally rich area situated in Mong Kok, Hong Kong. In this research, a series of numeric evaluations and statistical analyses of liveability and vibrancy metrics are presented, uncovering the tension created by existing commercial forms and local living patterns. This research started with multidimensional data mining, such as accessing planning data using Geographic Information Systems (GIS), perception data using Street View Images (SVI), and business performance data from Google; secondly, analysing the data via machine learning (ML) algorithm and statistical correlation to identify correlations overlaid with a mapping of spaces of measurable characteristics. The goal is to establish a measurable evaluation of the relationship between commercial vibrance and urban features that can further inform the impact of urban design strategies on fostering the vitality of community commercial centres.