电梯检验视频的缓冲器行为AI跟踪判定方法研究

  • 打印
  • 收藏
收藏成功


打开文本图片集

中图分类号:TP391.4;TU857 文献标识码:A 文章编号:2096-4706(2025)22-0023-07

Research on Al Tracking and Judgment Method for Buffer Behavior in Elevator Inspection Videos

LUO Weili’, CHEN Guilong,LIU Guixiong,CHEN Jianxun (1.Zhuhai Anyue Technology Co.,Ltd., Zhuhai 519ooo, China; 2.SchoolofMechanical andAutomotiveEngineering,South China UniversityofTechnology,Guangzhou ,China; 3.Guangdong Institute of Special Equipment Inspection and Research Zhuhai Branch, Zhuhai 519ooo, China)

Abstract:Thenumberof elevators inservice is huge,and their periodic inspections formamasive video database.Manual reviewofvideocompliancehasproblemsoflowefciencyandlowreliability.Toaddressthis,thispaperstudiesabuferbehavior trackingand judgment method basedonMask R-CNN + SORT+HeightSeqAnalyzer. It uses Mask R-CNN and SORT algorithms respectivelytoachievehigh-precisiondetectionofbuferobjects insingle framesandobjecttracking inthetemporaldimnsion. ItcalculatestereferenceheightofbjectstroughthepixelheightsequenceoftheHeightSeqAnalyzeralgorithm,judgesobject compressionbehavioraccordingtothecompressiondegree thresholdandcompression duration threshold,and finallymerges compresion eventsbasedontheoverlapofcompression frame sets tooutput the numberofbufercompresson events inthe video. Atraining setandavalidationsetareconstructedthroughorthogonal experimental design totrain the Object Detection model. Experiments show that in image-level detection,the method in this paper achieves a precision P=0.81 ,recall R=0.78 ,mAP@0.75-0.72, and mAP@0.5:0.95=0.63 when both the IoU threshold and confidence threshold are 0.75.In video-level judgment, the precision P (2号 and specificity Sp both reach 1.00,and the accuracy A=0.84 ,completely eliminating FP samples,which verifies the effectiveness and engineering applicabilityof the method in this paper for automatic review ofelevator inspectionvideo compliance.

Keywords: behavior determination; Object Detection; object tracking; elevator inspection

0 引言

电梯作为城市基础设施重要组成部分,关系到公共民生保障,其运行安全受到高度关注[。(剩余12879字)

目录
monitor
客服机器人