摘要
为了改善传统钢丝绳横向振动图像识别算法的检测效果,提高检测准确率与实时性,降低拍摄中残影现象和背景干扰带来的误差,提出一种基于倾斜矩形拟合的钢丝绳横向振动图像识别算法,并结合Visual Studio编程平台与OpenCV软件库使用C++语言进行算法设计。经过对比验证,在较低的输入图像分辨率(640×480)和帧率(30fps)下,本算法能准确识别钢丝绳横向振动幅值与振动频率,提高了钢丝绳横向振动的检测精度:振幅误差小于0.8%、频率检测误差小于0.15%,降低了由昂贵的高速相机带来的检测成本。
In order to improve the detection effect of the traditional image recognition algorithm of wire rope’s transverse vibra⁃tion,promote the detection accuracy and real-time ability,and reduce the error caused by afterimage and background interfer⁃ence in shooting,it proposes an image recognition algorithm which is based on inclined rectangle fitting for the detection of wire rope’s transverse vibration,and uses C++language to design the algorithm combined with Visual Studio platform and OpenCV software library.After comparison and verification,at a low input image resolution(640×480)and frame rate(30fps),the algo⁃rithm can accurately identify the amplitude and frequency of wire rope’s transverse vibration.Therefore,it can improve the detec⁃tion accuracy of wire rope’s transverse vibration:the amplitude error is less than 0.8%and the frequency detection error is less than 0.15%.It reduces the detection cost caused by expensive high-speed camera.
作者
任一男
彭玉兴
常向东
卢昊
REN Yinan;PENG Yuxing;CHANG Xiangdong;LU Hao(School of Mechanical and Electrical Engineering,China University of Mining and Technology,Jiangsu Xuzhou 221116,China;Jiangsu Intelligent Mining Equipment Collaborative Innovation Center(Jointly Built by the Province and the Ministry),Jiangsu Xuzhou 221116,China;Jiangsu Key Laboratory of Mining Electromechanical Equipment,Jiangsu Xuzhou 221116,China)
出处
《机械设计与制造》
北大核心
2025年第7期6-10,共5页
Machinery Design & Manufacture
基金
国家自然科学基金项目(51975572)
江苏高校优势学科建设工程项目(PAPD)
江苏高校品牌专业建设工程项目(TAPP)。
关键词
钢丝绳振动
图像识别
特征提取
轮廓拟合
Wire Rope Vibration
Image Recognition
Feature Extraction
Contour Fitting
