Inspiration
The devastating wildfires that swept across Algeria in 2021 have left a lasting impact on the country, underscoring the urgent need for innovative solutions to prevent and detect fires. If you're looking for inspiration to create a fire detection system that could make a real difference, look no further than this important issue. By leveraging cutting-edge technology like AI and machine learning, you could create a system that can detect fires in their earliest stages and alert emergency services, potentially saving countless lives and preserving natural habitats. Imagine being part of a movement to protect Algeria's beautiful landscapes and communities from the ravages of wildfire - your work could make a real impact and potentially save lives. The recent wildfires in Algeria and other parts of the world have highlighted the urgent need for effective fire detection systems that can protect both human life and the natural environment. As we become more aware of the impact of climate change and other environmental factors on the frequency and severity of wildfires, it's clear that we must take action to protect ourselves and the planet we call home. By creating innovative fire detection technologies that use the latest advancements in AI, machine learning, and sensor technologies, Not only can this save lives and property, but it can also help to preserve the natural habitats and greenspaces that are so vital to our planet's health and sustainability. Whether you're a developer, an environmentalist, or simply someone who cares about protecting our world, a fire detection system that can help to prevent and respond to wildfires is a powerful way to make a difference.
What it does
Our solution has two services: The fire tracking service which utilizes cameras and AI to detect fires in real-time. Cameras are installed in strategic locations where they can effectively monitor for signs of fire. As the cameras capture footage, the video feed is sent to an AI model for analysis.
Our Solution also has a software which allows to check on surey camera on diffrent area and recieve an alert if fire is detected also provide the location relative the cameras that detected it which is the second service. This information is critical for emergency responders to quickly assess the situation and take appropriate action.
How we built it
• we Trained our model from scratch using TensorFlow Framework, Classes we used our Architecture Based on AlexNet Architecture .
• For the front-end we used both ReactJs and VanillaJS, jquery. - ReactJS for the landing page which identifies our project . -VanillaJS, jquery and bootstrap, for the dashboard page that allow to check on survey camera, and recieve alerts .
Challenges we ran into :
Trying to understand wildfires , and how to use camera to detect fire , its integration with Raspberry pi
which allows real-time fire and smoke detection .
Finding the write data to train our AI model .
concepting the idea to be benifitiel has an impact on the real world wildfires .
Data used and researched considered :
• Research conducted by enp students about how to predicts fire spread using diffrent variables and mathematical equation : www.researchgate.net/publication/354678516_Applying_semi-empirical_simulation_of_wildfire_on_real_world_satellite_imagery_data
• For the DataSet used : https://drive.google.com/drive/u/1/folders/11QAgQVgfhKetW3H8k_T4NNKbht6B391q
Accomplishments that we're proud of
In a very short time , manage to learn about fire and smoke and making an AI model that detects them , also coming up with a solution that has both hardware and software parts, and binding the diffrent parts of the solution together .
What we learned
• That wildfires, forestry, are very complicated topics •but we also learnt of the possible solutions to wildfires and fire in general and the huge cost it takes but also the loss it will prevent or at least reduce the damage .
What's next for IgnitionGuard
We thought of things that could help our system expands and provides more services .
• In the future we intend to make an accurate Ai models That predicts WildFire spread and use it to help reduce the damage our earth would take .
• We can integrate drones to survey forest every specific period, and set a location for it to land and get charged using solar panels,as drones are already being used to help firefighters, their integration can be easy .
Built With
- adobe-illustrator
- api
- bootstrap
- canvas
- javascript
- jquiry
- machine-learning
- python
- react
- tensorflow
Log in or sign up for Devpost to join the conversation.