图像识别技术在螺旋锥齿轮接触区分析中的应用  

Application of Image Recognition to the Analysis of Spiral Bevel Gear's Contact Regions

作　　者：王延忠[1] 张国钢[1] 张俐[1] 雷震天[1] 

机构地区：[1]北京航空航天大学,北京100083

出　　处：《中国机械工程》2007年第16期1984-1988,共5页China Mechanical Engineering

基　　金：国家自然科学基金/航空科学基金联合资助项目(10377003)

摘　　要：提出了应用图像识别技术对轮齿接触迹进行自动识别和切齿修正的方法,建立了螺旋锥齿轮图像识别检测系统。首先采用图像识别方法对轮齿接触区图像进行了预处理,得到了接触区边缘轮廓;接着对接触区边缘轮廓进行了离散与拟合,得到了实际的接触迹图像;最后通过实际接触迹与理论接触迹的逼近,求得了小轮机床调整参数的修正量。A way of automatic recognition and cutting modification of contact trail was proposed. A system for image recognition and identification was established, providing an effective way to achieve automatic analysis and modification. The real bearing contact with the method of image recognition was preprocessed, and the image of the bearing contact profile was attained. Then,the image of the bearing contact profile was dispersed and fitted, and the real contact trail was generated. Finally, when the theoretic contact trail is approximated as closely as possible by the real contact trail, the modifications of machine-settings of pinion are evaluated.

关 键 词：图像识别 螺旋锥齿轮 接触迹 传动误差 

分 类 号：TH132[机械工程—机械制造及自动化]