融合HOG与SVM算法的智能船机油液监测方法探究

打开文本图片集

中图分类号：U676.4文献标志码：A

Research on Intelligent Ship Oil Monitoring Method IntegratingHOG and SVM Algorithms

GAO Bing¹²，WANG Lin1²，LI Wei²， LIU Guodong4 （1.MaritimeColege，Guangzhou510800，China;2.Guangdong ShipAutomationEnginering TechnologyResearch Center， Guangzhou51800Cina;3.arieEgingCllege，Dalian66，Cha;4.CHangpuWechongSiBlgCompaied Guangzhou 510715，China）

Abstract： This paper proposes a ship oil monitoring method that integrates the Histogram of Oriented Gradient （HOG） and Support Vector Machine （SVM） algorithms.Through algorithm optimization， improvement， and application，it achieves stableand intelligent online monitoring of wear particles in shipoil under various dierent states to resist bubble interference.Analyze and studytheabrasive particle recognition method integrating HOGand SVM from the aspects of image sample collection， image sample preprocessing， image feature extraction fused with HOG algorithm，and construction and training ofclassification model fused with SVM algorithm.Building anonline monitoring test bench for ship cylinder lubricating oil system，conducting comparative analysis of diffrent algorithm tests， the results showed that the classification accuracy of test samples using HOG + SVM scheme was significantly improved， with a classification accuracy of up to 84.35%

Key words： oil monitoring；ship engine room； histogram of oriented gradient； support vector machine; intelligent ship

1 引言

船海航运业界高度重视数字化转型升级，世界范围内开展了各种船舶智能化关键技术攻关研究，以期提升航运船舶智能化、设备安全可靠性、管理简易化。（剩余5209字）