Exploring the Potential of Machine Learning for Automatic Slum Identification from VHE Imagery
Duque, Juan Carlos
Patiño, Jorge Eduardo
Slum identification in urban settlements is a crucial step in the process of formulation of propoor policies. However, the use of conventional methods for slums detection such as field surveys may result time consuming and costly. This paper explores the possibility of implementing a low-cost standardized method for slum detection. We use spectral, texture and structural features extracted from very high spatial resolution imagery as input data and evaluate the capability of three machine learning algorithms (Logistic Regression, Support Vector Machine and Random Forest) to classify urban areas as slum or no-slum. Using data from Buenos Aires (Argentina), Medellin (Colombia), and Recife (Brazil), we found that Support Vector Machine with radial basis kernel deliver the best performance (over 0.81). We also found that singularities within cities preclude the use of a unified classification model.
Puede citar el documento con el texto siguiente:
Item perteneciente a la Colección
Items similares por titulo, autores y/o materias.
Amodio, Francesco (CAF; Caracas, 2013)This paper studies spillover effects among potential crime victims from investment in observable protection technologies. Criminals and potential victims interact in a frictional market for offenses. Externalities within ...
Freire, María Emilia; Hoornweg, Daniel; Slack, Enid; Stren, Richard; The Global Dialogue Institute (CAF, 2016-10-14)Every day around the world our cities add 180,000 new citizens. By 2050, there will be a total of 3.5 billion new urban residents, which will represent a two-fold increase from current numbers. To understand each city´s ...
Brassiolo, Pablo (CAF; Caracas, 2014-10-01)This paper investigates whether lowering the cost of divorce can reduce domestic violence. The cost of divorce influences the bargaining position of spouses, and thus, their behavior within the marriage. This study takes ...