Skip to content

Project on Automated Detection of Hazards for Space Apps Challenge 2020

License

Notifications You must be signed in to change notification settings

KorKux1/Automated-Detection-of-Hazards

Repository files navigation

AQUDA

We propose a data integration and prediction system (AQUDA), that has state-of-the-art machine learning algorithms for novelty detection and a web app that includes dashboards that allow any researcher to make real time decisions, moreover the system is designed that any user regardless of his or her academic background can receive some information about the actual situation of the geographic zone with the purpose to bring awareness of climate change. We consider it a very important problem to solve, because we want to make people conscious that pollution is getting worse due to vehicles and industries that throw out harmful gases to the atmosphere, and if we don't act now, there will be an irreversible damage.

PRs Welcome GitHub license GitHub contributors GitHub issues GitHub forks GitHub stars GitHub language count

Space Apps Challenge - AQUDA Solution

AQUA was born from the hackathon space apps 2020. You can find more information HERE.

Requirements

  • Languages:

python

  • Libraries: scikit-learn, numpy, matplotlib, dash, torchvision, PIL, Django, bootstrap, pytorch, pandas, seaborn.

Dashboard

We build a dashboard for the visualization of data on air quality.

AQUDA Index

Visualization of statistics for the metrics of PM, RH, UGRD, VGRD, HPBL, TMP, GOES per year.

AQUDA Index

Visualization of the behavior of the measures with respect to the others

AQUDA Measurament Relationship

Visualization of the averages of the different measurements per month.

AQUDA Measurament Relationship

We raise awareness for all people. It consists of knowing how many cigarettes smoked per day the air you breathe is equivalent.

AQUDA awareness

We created a system based on the detection of anomalies to be able to identify data of harmful particles and make predictions of future records.

AQUDA Novelty 1 AQUDA Novelty 2

We developed a deep neural network based on ResNet-50 to perform dust detection in the air by satellite images.

AQUDA Dust

Installation

Clone the repository

$ git clone https://github.com/KorKux1/Automated-Detection-of-Hazards.git

Create virtual enviroment:

$ python -m venv env

Active your enviroment

Install dependencies

$ pip install -r requirements.txt

Run the project

$ python manage.py runserver

Contributors

  • Christian Urcuqui.
  • Cristhian Eduardo Castillo.
  • Jhoan Delgado.

License

AGPL-3.0


⌨️ with the ❤️

Releases

No releases published

Packages

No packages published