Proposed System of Enhancement in Accuracy for Fire Detection System using Machine Learning Algorithm
Anjali Pathak1, S.M. Chaware2
1Anjali M Pathak*, Computer Engineering ,Savitribai Phule Pune University, India.
2Dr. S.M. Chaware, Computer Engineering ,Savitribai Phule Pune University, India.
Manuscript received on January 23, 2020. | Revised Manuscript received on January 30, 2020. | Manuscript published on February 10, 2020. | PP: 2093-2097 | Volume-9 Issue-4, February 2020. | Retrieval Number: D1446029420/2020©BEIESP | DOI: 10.35940/ijitee.D1446.029420
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The video surveillance system has become a important part in the security and protection of cities. The Video surveillance has become an important factor in the cities, since smart monitoring cameras mounted with intelligent video analytics techniques can monitor and pre-alert system by capturing abnormal activity such as fire events. The current world is completely under CCTV for make the various areas secure. The video recorded is unable to find out fire detection at early stage of fire event. After event happened this video sequence is used to find out causes of an event/fire but problem is after event happened system are unable to save loss by that event or accident, so there is need to such system is able to help us in early event detection and pre-alert generation system. Motive behind this proposed work is to invent pre-alert generation system without any hardware as well as sensor. Accuracy of this proposed system may be approx.85-90% or more which is better than existing system.
Keywords: Closed Circuit Television (CCTV), Intelligent Video Surveillance (IVS), Conventional Neural Network.
Scope of the Article: Machine Learning