Leveraging Artificial Intelligence for Slum Mapping and Upgrading Programs Using Drones: The Case of Bangalore City, India

Introduction

In recent decades, India has experienced a sharp increase in urbanization as a result of a large-scale migration of people from rural to urban areas in search of better job and living conditions. Slums have sprung up all over urban areas as a result of this fast migration since urban infrastructure could not keep up. Bangalore, one of the Indian States that receives the most foreign direct investment (FDI), has also become a leader in this area. These encouraging changes are, however, accompanied by certain unsettling characteristics, such as a stagnant job market, increasing urban poverty, the breakdown of the organized sector, and a significant increase in the unorganized labor force. According to some estimates, 65 million people live in India's slums today. There are a lot of issues that come with living in these slums. One of the obstacles to reaching this goal is the dearth of reliable baseline data on slum regions. The government needs these kinds of data to help in prioritizing areas, tracking program implementation, and calculating areas before and after upgrades.

To address these problems AI-based drone mapping of slums can be a useful tool for comprehending the infrastructure, living conditions, and spatial distribution of these informal settlements. Generally speaking, computer vision, machine learning, and geographic information systems (GIS) algorithms and approaches are used while mapping slums using AI. The following list of popular AI algorithms for slum mapping Various algorithms and methods from the fields of computer vision, machine learning, and geographic information systems (GIS) are used in the process of mapping slums using artificial intelligence. The following are a few popular AI algorithms for slum mapping: CNNs, or convolutional neural networks: CNNs are a popular choice for applications involving object detection and picture classification, which makes them appropriate for spotting signs of slums in satellite or aerial imagery.

Semantic Segmentation: By labeling every pixel in a picture, semantic segmentation algorithms make it possible to identify and distinguish impoverished areas from satellite photography. Object Detection Algorithms: Within impoverished areas, algorithms like Faster R-CNN and YOLO (You Only Look Once) can locate and identify particular objects or buildings, assisting in the delineation of their boundaries and features. These adaptable devices today have cutting-edge technologies installed, artificial intelligence being one of the most revolutionary (AI). Drones equipped with artificial intelligence (AI) have the potential to address the problem of slum mapping by utilizing technology to produce useful insights and promote constructive change in informal settlements.