Identification of Impervious Built-Up Surface Features Using ResourceSat-2 LISS-III-Based Novel Optical Built-Up Index
In the context of urban planning, the increasing urban concentration and growth result in changes from natural landscape to impervious surface features. Remote sensing provides an efficient method in automated identification of land use/cover classes. How
- PDF / 814,612 Bytes
- 14 Pages / 439.37 x 666.142 pts Page_size
- 83 Downloads / 177 Views
Abstract In the context of urban planning, the increasing urban concentration and growth result in changes from natural landscape to impervious surface features. Remote sensing provides an efficient method in automated identification of land use/cover classes. However, a common challenge is the accurate extraction of builtup features from satellite images. The conventional Normalized Difference Built-up Index (NDBI) has been modified by several researchers in the anticipation of improvement of the built-up area classification. The indices adopted in the study are Index-based Built-up Index (IBI), Built-up Index (BUI), NDBI, and the newly developed Impervious Built-up Index (IBUI). These indices work on automated kernel-based probabilistic thresholding algorithm to group the index values into built-up and non-built-up areas. This study investigates the performance of the abovementioned spectral indices on ResourceSat-2 Linear Imaging Self-ScannerIII (LISS III) imageries of the city of Kolkata, India, and its adjoining areas in the delineation of built-up areas and compares them based on spectral feature space correlation and classification approach. Although all the built-up indices showed high mutual correlation, the performance varied greatly as showed by the accuracy in the classification. Overall accuracy values of built-up feature extraction using IBUI, IBI, BUI, and NDBI are 92.33%, 89%, 86%, and 80.67% respectively. Keywords Built-up · Feature extraction · Spectral indices · Classification · Correlation
A. Santra (*) · S. S. Mitra · S. Sinha · S. Routh · A. Kumar Department of Civil Engineering, Haldia Institute of Technology, Haldia, India e-mail: [email protected] © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 P. Kumar et al. (eds.), Remote Sensing and GIScience, https://doi.org/10.1007/978-3-030-55092-9_7
113
114
A. Santra et al.
1 Introduction The world has witnessed a few natural and anthropogenic pulverizations which had a direct antagonistic impact on sustainable livelihood. The heinous effect of such catastrophe profoundly affects the land cover features; however, the impact proliferates more colossally on the built-up areas, causing alterations in the land cover dynamics. This results in the essential requirement of rapid identification of urban built-up areas, as the maximum economic loss occurs within the urban built-up areas (Varshney and Rajesh 2014). Under such emergency scenario, remote sensing techniques offered complex as well as diverse characteristics of data (Sajjad and Kumar 2019). Therefore, the traditional classification systems detecting specific land use/cover (LULC) changes may yield erroneous results sometimes. The index-based automatic extraction of land features from satellite imagery yields quick and accurate results and thus advantageous for disaster management and spatial mitigation response. However, among all the land cover features, accurate extraction of builtup features is a common challenge from the sa
Data Loading...
