Fisher准则函数在机械刀具切削磨损检测中的应用  

Application of Fisher Criterion Function in Mechanical Tool Cutting Wear Detection

作　　者：任慧[1] 马金萍 REN Hui;MA Jinping(School of Statistics and Big Data,Zhengzhou College of Finance and Economics,He’nan Zhengzhou 450000,China;School of Statistics,Xi’an University of Finance and Economics,Shannxi Xi’an 710100,China)

机构地区：[1]郑州财经学院统计与大数据学院,河南郑州450000 [2]西安财经大学统计学院,陕西西安710100

出　　处：《机械设计与制造》2025年第1期154-159,165,共7页Machinery Design & Manufacture

基　　金：2023年国家社科基金年度项目(23BTJ024)。

摘　　要：在高精度机床道具切削磨损无接触图像检测中,磨损区域边缘特征容易发生混淆,区分其像素边缘区域成为一个难点。Fisher判别函数可以用于特征提取,即通过选择最具有区分力的特征来改善分类性能。针对这一特点,提出基于Fisher准则函数的机械刀具切削磨损检测方法。首先,对采集到的机械刀具切削磨损图像实施图像增强处理,利用实施修正线性拉伸操作解决光照不均匀问题,凸出刀具磨损区域的细节特征并提高图像质量;其次,利用基于Fisher准则函数的图像阈值分割方法,将刀体区域和其工作背景区域分割出来,使刀具磨损区域更加清晰,提高磨损区域的可识别性;最后,通过Roberts边缘检测确定刀具切削磨损边缘,并使用Zernike正交矩确定刀具切削磨损边缘的精确值以获取磨损区域,提取其磨损信息实现机械刀具切削磨损的检测。实验结果表明:所提方法具有较高图像处理能力以及磨损检测精度。In the non-contact image detection of cutting wear of high-precision machine tool props,the edge features of the wear area are prone to confusion,and distinguishing its pixel edge areas becomes a difficulty.The Fisher discriminant function can be used for feature extraction,which improves classification performance by selecting the most discriminative features.To address this characteristic,a mechanical tool cutting wear detection method based on Fisher criterion function is proposed.Firstly,im⁃age enhancement is applied to the collected images of mechanical tool cutting wear,and the problem of uneven lighting is solved by implementing a modified linear stretching operation,highlighting the detailed features of the tool wear area and improving image quality.Secondly,using the image threshold segmentation method based on the Fisher criterion function,the tool body area and its working background area are segmented to make the tool wear area clearer and improve the recognizability of the wear area.Finally,the cutting wear edge of the tool is determined through Roberts edge detection,and the precise value of the cutting wear edge of the tool is determined using Zernike orthogonal moments to obtain the wear area.The wear information is ex⁃tracted to achieve the detection of mechanical tool cutting wear.The experimental results show that the proposed method has high image processing ability and wear detection accuracy.

关 键 词：图像增强 Fisher准则函数 图像阈值分割 Roberts边缘检测 Zernike正交矩 磨损检测 

分 类 号：TH16[机械工程—机械制造及自动化] TP391[自动化与计算机技术—计算机应用技术]