Fire Detection and Analysis in Video Surveillance Application using CNN

Fire detection based on computer vision is one amongst the key technologies of the trendy surveillance systems. Convolutional Neural Networks (CNN) has yielded cutting edge execution in picture characterization, and diverse computer vision undertakings. The fire detection systems increases the accuracy, thereby reduces the fire disasters and minimizes the overall threats to the society. However, how to reliably and accurately detect the fire is still a tricky problem. This article proposes a computationally economical CNN design for fireplace detection, localization and spread rate estimation of the fireplace. Experimental results shows that the proposed method has improved accuracy and effectiveness of the localization and spread rate estimation.