ORIGINAL PAPER Extraction of impervious features from spectral indices using artificial neural network Nilanchal Patel & Rohit Mukherjee Received: 23 July 2013 /Accepted: 2 June 2014 # Saudi Society for Geosciences 2014 Abstract An urban area comprises a complex mix of diverse land cover types and materials. Urban ecology and environ- ment is significantly influenced by the proportion of impervi- ous cover that is increasing considerably with time due to the continuous influx of people into urban areas. Therefore, it is of vital importance to determine the spatiotemporal pattern and magnitude of urbanization. In the present study, we have employed a supervised backpropagation neural network in order to extract the impervious features using five spectral indices, such as one vegetation index—Soil-Adjusted Vegeta- tion Index (SAVI), one water index—Modified Normalized Water Index (MNDWI), and three urban indices—Normal- ized Difference Built-up Index (NDBI), Built-up Index (BUI), and Index-Based Built-up Index (IBI). The study has been performed using Landsat Thematic Mapper data of Novem- ber, 2011, of the rapidly urbanizing city of Ranchi, capital of Jharkhand state, India. Using different combinations of these spectral indices while keeping SAVI and MNDWI constant, seven composite images were built, and from each of these composites, impervious features were classified and its accu- racy assessed with reference to high-resolution images pro- vided by Microsoft Bing Imagery and adequate ground truthing. It was observed that along with SAVI and MNDWI, whenever IBI was used in any combination, it decreased the classification efficiency. On the other hand, NDBI and BUI, individually or when used together, discriminated the imper- vious features from the others with high accuracy with the combination of SAVI, MNDWI, and BUI achieving the highest accuracy of 90.14 %. Keywords Impervious feature . Spectral indices . Backpropagation . Feature extraction Introduction Urban areas consist of a complex mix of diverse land cover types and materials, such as roads, rooftops, vegetated areas, soil surfaces, and water bodies, and are quite different from rural and natural environments, which lead to a substantial difference in physical processes between these land covers (Schueler 1994) and also to variations in local urban climate (Oke 1987). A typical urban environment is characterized by its vegetation density and coverage, percent area of impervi- ous surfaces, sky view factor, and the structure and composi- tion of buildings (Oke 1987). Urban areas are dominated by built-up land with impervious surfaces; therefore, the ecosystem, hydrological system, biodiversity, and local climate maybe significantly impacted due to the conver- sion of natural lands into impervious built-up lands. This can result in negative aspects such as the urban heat island phenomenon (Xu 2007). Rapid development of urban areas, which leads to replacement of vegetation cover with buildings and paved road, has brought about negative repercussions on the world environment, such as less precipitation, dryness, and higher temperatures, which lead to global warming (Xu 2008). Over the past two decades, researchers have become increasingly in- terested in using remotely sensed imagery to address urban and suburban problems (Jacquin et al. 2008). Impervious features are closely associated with many urban- or environment-related studies, for example, urban land use classification, urbanization, etc. (Madhavan et al. 2001; Phinn et al. 2002; Arnold and Gibbons 1996). Arnold and Gibbons (1996) defined impervious features as anthropo- genic features through which water cannot infiltrate into the N. Patel (*) : R. Mukherjee Department of Remote Sensing, Birla Institute of Technology Mesra, Ranchi, Jharkhand, India e-mail: nilanchal.patel@gmail.com R. Mukherjee e-mail: rohitmukherjee@live.com Arab J Geosci DOI 10.1007/s12517-014-1492-x