Abstract
Fire in power system is easy to cause heavy casualties and a large number of property losses. Because of the complex and changeable environment, it is important to accurate early fire detection in current research hotspot issues. Aiming at the fire safety detection of power system transmission, transformation and distribution equipment, a smoke and fire target detection method based on YOLOV5 model is proposed according to the questions of small target size, which is easy to be confused with the actual situation. According to the characteristics of smoke and fire movement changes, the network is improved, and a detection layer is added to the original detection layer, so that the feature map can continue to be up-sampled and the detection accuracy can be improved.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
