Machine Learning-Based Image Processing For Urban Development
Project: Automatic recognition of urban patterns for massive social simulations to evaluate the implementation of public policies
The analysis of images of urban regions is crucial for the development of public policies oriented to improve life quality in densely populated cities. The detection and analysis of organic growth requires of the development of advanced techniques aimed to recognize different types of structures such as buildings, roads, and vegetation.
During 2023, I was the Leader of the CONICET Research Team of a Fundación Sadosky Project that used drones and the latest machine learning techniques for image recognition. Learn more about it watching this video.
The developments of the team allow for a fast learning of structures and semantic content associated to images taken with drones, allowing for a reduction of the cost of the characterization task. These techniques can be applied to any kind of populated areas. In particular, it is relevant for the characterization of informal settlements where, usually, no official information is available.