search

tobybreckon / fire-detection-cnn

real-time fire detection in video imagery using a convolutional neural network (deep learning) - from our ICIP 2018 paper (Dunnings / Breckon) + ICMLA 2019 paper (Samarth / Bhowmik / Breckon)
MIT License
533 stars 171 forks source link

Please tell us about your usage ... #28

Open tobybreckon opened 4 years ago

tobybreckon commented 4 years ago

If you find our fire detection software, CNN models or dataset useful - please post to let us know how you are making use of it. It's great to hear about it.

e.g.

I am using this for my {high school, college, university} project at ... for ... I am using this for ... in collaboration with .... I am using at {company, organization, .... } to ... ... [If you can't post publicly, consider just emailing - toby.breckon@durham.ac.uk]

I know it has already been used for this hackathon and by this team with thermal imagery.

How have you made use of this in your work ? let us know

It helps keep up our motivation to maintain and update the repo.

Just reply to this issue/thread.

tobybreckon commented 4 years ago

A similar CNN architecture construction (i.e. 3 convolutions with pooling followed by 2 dense layers) is used in this work: FireNet: A Specialized Lightweight Fire & Smoke Detection Model for Real-Time IoT Applications Arpit Jadon, Mohd. Omama, Akshay Varshney, Mohammad Samar Ansari, Rishabh Sharm

.... smaller network taking a 64 x 64 input, runs at 24fps on a Raspberry Pi.

Afsalmc commented 2 years ago

.... smaller network taking a 64 x 64 input, runs at 24fps on a Raspberry Pi.

raspberry pi ? i thought a powerful rig is necessary for this model to run . thankyou