.png){kind=small}
Monitoring urban impervious surface using multi-source remote sensing in cloud-prone areas
Ms LING Jing PhD Student, Department of Geography, HKU
Abstract:
Urban impervious surfaces (UIS) has occupied an essential role and aroused widespread interest in urban studies. However, the problem of cloud occlusion in optical images hinders continuous and timely UIS monitoring, especially for tropical and subtropical regions which suffer from a long period of rainy and cloudy weather throughout the year. Synthetic aperture radar (SAR) has all-weather imaging capability to overcome the challenge, and thus may provide complementary information about UIS to cloud-contaminated optical images, while the land surface information provided by SAR is limited as SAR suffers from various imaging mechanism problems, such as speckles, foreshortening, layover, and shadows. The effective integration of optical remote sensing and SAR remote sensing to make the land surface information obtained by the two complements each other is important to improve the accuracy of UIS monitoring. However, most existing fusion studies directly utilize cloud-free optical images in practical applications, which, actually, is not applicable in tropical and subtropical areas where clouds are often unavoidable. Previously, in the case of clouds occurrence, SAR data were often adopted alone, showing no particularly satisfactory accuracy. There are few studies on SAR and optical data fusion for UIS recognition in cloudy areas. The accurate and timely UIS monitoring for cloud-prone areas remains challenging. To fill the above-mentioned research gap, this study aims to propose a methodological framework 1) to quantify the cloud impacts on UIS estimation using optical images, 2) to investigate the mechanism by which SAR compensates optical images for land cover discrimination under the coverage of clouds, and 3) to develop a set of UIS estimation methods by fusing cloud-contaminated optical and SAR data considering the impact of cloud cover, to achieve accurate and timely UIS estimation in cloud-prone areas.