Socio-Economic Status Classification for Poverty Mitigation

Abstract

The concept of satellite images available from various sources using image processing technique provides an insight to predict or determine the poverty at various levels based on certain features like roads, buildings, schools and other factors. In the present scenario we are making use of Convolutional Neural Networks (CNN) mainly used for image recognition and processing to predict the poverty at various levels. The efficiency and accuracy of the model depends upon the number of pre-processing steps employed, the number of datasets used, the type of welfare indicator targeted and the choice of AI model. Leveraging advanced algorithms together with the ample amount of geographic information which satellites capture offers an appropriate approach for delicate socio-economic indicator evaluation. Leveraging advanced algorithms together with the ample amount of geographic information which satellites capture offers an appropriate approach for delicate socio-economic indicator evaluation. Key Words: Convolutional neural networks, ReLU architecture, Gaussian Blur