The project proposes an approach for the development of deep learning based classifiers for LiDAR point cloud, based on methodologies and tools that have been studied.
This is an image acquired in one of the tests in the Puerto Rico terrain data collection
Natural disasters can be defined as phenomena generated by the planet's own processes, that is, by nature. These phenomena are extremely destructive and can have catastrophic consequences for humans. Thus, places frequently affected by natural disasters are always developing and applying technologies to protect their population.
Still, many natural disasters are unpredictable and unfortunately their frequency is significantly increasing because of humans. Human actions directly affect the planet, as can be seen in phenomena such as global warming and the greenhouse effect. Some of the most common and destructive natural disasters are earthquakes, hurricanes and volcanic eruptions. Usually, more than one natural disaster is linked to another, for example: hurricanes can culminate in floods, large landslides can be an indication of earthquakes.
For this reason this project is being developed in this hackathon, it was done with the aim of locating, understanding and solving problems related to some natural disasters using the Python language for coding, aid of the QT Modeler software and the use of data. provided by the Hackathon itself for a more realistic analysis of a land that at first suffered losses to natural disasters.
Several challenges were proposed for solving problems with natural disasters, among them some were made to analyze specifically, they are:
- Detection of buildings and constructions;
- Damaged roads and buildings;
- Isolated areas;
- show_images.py is our script made using Python to visualize and select point cloud data to segment. It also has a filter application to extract the main points of a selected area.
- hackaton.ipynb is our notebook for identifying households from footprints data.
With our project that was developed in 3 weeks, we got the first place in this great challenge!!
