Detecting Slums From Quick Bird Data in Pune Using an Object Oriented Approach

Abstract

ABSTRACT:We have been witnessing a gradual and steady transformation from a pre dominantly rural society to an urban society in India and by 2030, it will have more people living in urban than rural areas. Slums formed an integral part of Indian urbanisation as most of the Indian cities lack in basic needs of an acceptable life. Many efforts are being taken to improve their conditions. To car ry out slum renewal programs and monitor its implementation, slum settlements should be recorded to obtain an adequate spatial data base. This can be only achieved through the analysis of remote sensing data with very high spatial resolution. Regarding the occurrences of settlement areas in the remote sensing data pixel-based approach on a high resolution image is unable to represent the heterogeneity of complex urban environments. Hence there is a need for sophisticated method and data for slum analysis. An attempt has been made to detect and discriminate the slums of Pune city by describing typical characteristics of these settlements, by using eCognition software from quick bird data on the basis of object oriented approach. Based on multi resolution segmentation, initial objects were created and further depend on texture, geometry and contextual characteristics of the image objects, they were classified into slums and non-slums. The developed rule base allowed the description of knowledge about phenomena clearly and easily using fuzzy membership functions and the described knowledge stored in the cl assification rule base led to the best classification with more than 80% accuracy.

Detection of Informal Settlements from VHR Images Using Convolutional Neural Networks

“…They include the patch size, m = [65, 99, 129, 165], the number of convolutional layers, g = [2,3,4] the number of the fully connected layers, t = [1,2,3], the number of the filters, k = [8,16,32,64] and the kernel dimension, h = [7,17,25]. We keep the patch size constant at m = 99 while optimizing the hyperparameters in order to keep a low computational cost.…”

Quantifying Slumness with Remote Sensing Data

“…On the contrary, the descriptors of the urban layout, i.e., the structure and texture variables, are easily computed from VHR imagery and polygon boundary data and the use of texture measures for slum detection and mapping has been reported in different cities around the world: Accra, Ghana ; Campinas, Brazil (Barros and Sobreira, 2005); Casablanca, Morocco (Rhinane, 2011); Delhi, India (Niebergall et al, 2007); Guatemala, Guatemala (Owen and Wong, 2013); Hyderabad, India (Kit et al, 2012); Pune, India (Shekhar, 2012); and Rio de Janeiro, Brazil (Barros and Sobreira, 2005). We tested univariate regressions in order to get a better understanding of the explanatory power of each selected structure and texture variables alone for slum index (table 6).…”

“…Previous works have explored the use of remote sensing as a proxy for for socio-economic class differentiation within a city (Avelar et al, 2009;Tapiador et al, 2011), and several works have demonstrated the usefulness of remote sensing to distinguish a slum from its surrounding neighborhoods in the last decade (Barros, 2008;Hofmann et al, 2008;Kit et al, 2012Kit et al, , 2013Kohli et al, 2012;Netzband et al, 2009;Niebergall et al, 2007;Owen and Wong, 2013;Rhinane, 2011;Stow et al, 2007;Shekhar, 2012;Stoler et al, 2012;Weeks et al, 2007). Most previous works of slum detection from remotely sensed data have used high and very high spatial resolution (VHR) imagery from Quickbird, Ikonos, and SPOT 5 sensors, and some approaches have used complementary socio-economic data on housing size and value, while others have used only textural and fractal dimensions derived from satellite imagery to implement statistical regressions (Patino and Duque, 2013).…”

Automated detection of slum area change in Hyderabad, India using multitemporal satellite imagery

“…Successful examples of slum identification from very high resolution imagery include methods based on object-based image analysis (Hofmann et al, 2008), object segmentation and classification (Shekhar, 2012), morphological opening and closing (Rhinane et al, 2011). One of the last successful attempts to automatically identify slums in India has been made by Shekhar, 2012, where eCognition-supported object segmentation and classification has been used to identify slums in Pune.…”

“…One of the last successful attempts to automatically identify slums in India has been made by Shekhar, 2012, where eCognition-supported object segmentation and classification has been used to identify slums in Pune. The author reaches identification accuracy of 87% as benchmarked against the slum survey using classification rules such as structure size and density, street pattern irregularity and vegetation distribution.…”