Automated Fire Extinguishing System Using a Deep Learning Based Framework

Fire accidents occur in every part of the world and cause a large number casualties because risks involved manually extinguishing fire. In most cases, humans cannot detect extinguish fire manually. robots with sophisticated functionalities are being rapidly developed nowadays, these systems use sensors detectors. However, they lack mechanisms for early detection fire, case casualties. To prevent such its stages, deep learning-based automatic mechanism was introduced this work. human presence locations were carried out using convolution neural networks (CNNs), configured to operate on chosen dataset. For detection, custom learning network formed by tweaking layer parameters CNN detecting fires better accuracy. Alex-net architecture employed accident zone. We experimented analyzed proposed model various optimizers, activation functions, rates, based accuracy loss metrics generated The best combination evaluated from an Adam optimizer softmax activation, driven rate 0.001, providing model. Finally, experiments tested mobile robotic system configuring them wireless control modes. mode, robot made patrol around monitor accidents. It automatically extinguished learned features triggered through