ANALYSIS FOR FIRE DETECTION SYSTEM FOR IMPROVEMENT IN ACCURACY

The video surveillance machine has come to be an essential element within the security and safety of towns. The video surveillance machines has become an essential element within the security and safety of towns. Reliable fire detection systems with high accuracy and speed are essential for the safety of smart city services. A fire detection system requires precise and firm mechanisms to make the right decision in a fire situation. Since maximum commercial fire detection systems use a sensor, their fire recognition accuracy is poor because of the limitations of the detection capability of the sensor. After event happen this video sequence is used to find out causes of an occasion/fire but trouble is after occasion passed off we are not able to keep loss by way of that event .So there is need to such system which is able to assist us in early fire event detection and pre-alert generation. Purpose behind these proposed work is to invent pre-alert technology machine modern deep learning networks using without any hardware in addition to sensor. Accuracy of this proposed device approx. 85% or extra which is better than current machine.