Impact Factor

7.883 Call For Paper

Share on

Author(s):

Sakshi Kharche | Saurabh Bakare | Sima Supekar | Gokul Bhor | Shrikrishna Nimbekar
Keywords:

Video Analysis, Convolutional Neural Networks (CNNs), Image Processing, Deep Learning, Surveillance Systems, False Positives, Smart City Applications, Real-time Detection, And Fire Detection.
Abstract:
In Order To Minimize Damage, Fire Detection Is A Crucial Component Of Early Warning Systems In Both Urban And Rural Areas. And Quickening Response Times. Traditional Fire Detection Methods Have Limited Environmental Detecting Capabilities And Significant False-positive Rates.flexibility Because They Usually Depend On Manually Designed Features In Image Processing Or Sensor-based Systems. Convolutional Neural Networks This Study Proposes An Improved Fire Detection Method That Is Suited For Real-time Picture Processing Using Convolutional Neural Networks (CNNs). The Recommended Technique Uses A Lightweight Deep CNN Architecture That Can Accurately Distinguish Between Areas That Are Burning And Those That Aren't, In Various Lighting And Background Scenario. A Proprietary Dataset With A Range Of Fire Scenarios Was Used To Train And Evaluate The Model. Performance Metrics Such As Precision, Recall, F1-score, And Detection Time Were Significantly Improved In Comparison To Traditional Methods And Baseline CNN Models. The System's Robust Performance In Real-time Video Streams Makes It Suitable For Use In Surveillance Systems, Drones, And Smart City Applications.

Download Article