Research On Laboratory Equipment Condition Monitoring Application Based On Infrared And Visible Light Image Fusio

As China’s higher education enters a stage of widespread and high-quality development,the government is continuously increasing funding for university laboratory construction.As universities acquire a large quantity of various types of experimental equipment,the management of this equipment faces numerous challenges.Problems include incidents where improper use of electricity causes laboratory fires,or cases where equipment is left on when personnel leave.There are also cases where equipment starts automatically when no one is using it,resulting in temporal and spatial mismatches.These phenomena pose safety hazards and can even cause accidents.This paper proposes to apply image fusion technology and object detection technology to equipment status monitoring to address the shortcomings of traditional sensorbased laboratory equipment status monitoring technology,such as small detection coverage and single-use scenarios.By doing this,the accuracy and universality of detection are improved.The laboratory electrical equipment is used as the object,and a modular design scheme is adopted to verify and develop the laboratory equipment status monitoring system.The primary research work is as follows:(1)Regarding the initial stages of laboratory equipment fires,the YOLOX object detection algorithm has issues with low recognition accuracy and imprecise identification of locations.An improved YOLOX algorithm is proposed to address these issues.The improved algorithm was validated using photographs taken in a laboratory,and the experimental results showed that the improved algorithm has better recognition performance.(2)Single target detection technology based on visible light has certain limitations in laboratory equipment monitoring.It can only detect significant defects or abnormal situations,but it cannot effectively detect safety hazards such as equipment abnormal startup or sustained heat generation.In response to such hidden dangers with infrared characteristics,this paper proposes an improved image fusion algorithm that combines Non-Subsampled Shearlet Transform(NSST),Discrete Wavelet Transform(DWT),and Pulse Coupled Neural Network(PCNN).The algorithm first fuses the images using the image fusion algorithm,and then identifies the fused image using the object detection algorithm.By combining object detection technology and image fusion technology,the effectiveness and accuracy of monitoring the abnormal working states of laboratory electrical equipment have been improved.(3)A modular design scheme was adopted to develop a laboratory equipment status monitoring system.The system was tested using laboratory equipment as the object,and the test results verified the effectiveness of the laboratory equipment status monitoring system.