Fine-Grained Classification of Urban Functional Zones and Landscape Pattern Analysis Using Hyperspectral Satellite Imagery: A Case Study of Wuhan

Landscape mapping and pattern analysis of fine-grained urban functional zones (UFZ) are important in urban planning and urban development. Multi-spectral images have largely enabled only coarse-grained land-use classification of urban areas. By contrast, hyperspectral images have shown potentials to facilitate fine-grained classification of urban areas. In this paper, we evaluated and analyzed the classification of fine-grained UFZ in the central city of Wuhan, Hubei, China, using GaoFen-5 (GF-5) hyperspectral satellite imagery. We first compared the performance of hyperspectral data (GF-5) and multispectral data (Landsat 8) for the classification of fine-grained functional areas of cities by employing two classical classification algorithms and two deep learning methods. We also propose the deep learning-based SSUN-CRF algorithm, to enable better landscape pattern analysis. We then analyzed the landscape pattern of the seven administrative districts in the main urban areas of Wuhan by combining ten landscape indicators, based on the precise UFZ classification results obtained from the hyperspectral data. Experimental results illustrated the following: (1) Compared to multispectral images, hyperspectral images can allow for a more accurate UFZ classification. (2) Deep learning classification algorithms can better exploit hyperspectral image data, with the SSUN-CRF algorithm, in particular, being able to achieve an overall accuracy of 93.86% and a Kappa coefficient of 92.08%. (3) We were able to analyze the landscape pattern of Wuhans main urban areas based on the results of the UFZ classification. The results indicated that hyperspectral remote sensing imagery shows significant potentials in mapping fine-grained urban functional zones.