Font Size: a A A

Research On Security Monitoring Strategy Of Power Laser Operation Scene Based On Image Recognition

Posted on:2022-07-28Degree:MasterType:Thesis
Country:ChinaCandidate:C X LiFull Text:PDF
GTID:2480306521455314Subject:Electrical engineering
Abstract/Summary:PDF Full Text Request
For the problem of foreign matter or trees close the transmission line,the foreign matter removal device based on laser technology is generally used.There are not only high-voltage electricity,but also high-energy laser in the power laser operation scene,which is very dangerous.Therefore,it is necessary to make appropriate safety monitoring strategy to monitor and early warn the operation scene.Aiming at the defects of traditional monitoring system,this paper proposes a security monitoring strategy based on image recognition to monitor and warn the whole power laser operation scene.For mobile environment of the monitor,this paper proposes an improved YOLO algorithm for the mobile end,selects the backbone network,and other improvement measures,so that the detection model is more suitable for the mobile end detection environment.Then,the improved algorithm is applied to the scene of power laser operation,which mainly includes the object detection of laser cleaning irradiation range and the object detection of laser construction site.In this paper,choosing the corresponding detection targets and constructing the corresponding Laser Safety Control dataset(LSCD),and two detection models are trained.Experimental results show that in both cases,the improved YOLO algorithm is more faster than other algorithms.On the premise of ensuring the accuracy,it greatly reduces the model size and reasoning time,so that it can be better embedded in the mobile device,which plays a supporting role for the front-end to detect danger and start the alarm strategy.Finally,for the possible security risks in the process of power laser operation,this paper designs two kinds of mobile terminal security monitoring and early warning systems: the security early warning of laser irradiation range and the security early warning of laser construction site.The dangers that may appear in the two cases are monitored in real time at the front-end camera.When the danger is detected,the warning is carried out.Through the simulation of the mobile terminal environment security early warning,it is verified that the two early warning systems perform well in the mobile terminal platform,and can send the alarm in time.Integrating the both systems,a joint safety monitoring strategy of double early warning systems is designed,so can protect the safety control of laser cleaning operation,which has a certain reference value for the intelligent online monitoring under power laser operation.
Keywords/Search Tags:laser foreign body removal, YOLO algorithm, safety monitoring, image recognition
PDF Full Text Request
Related items