| 磨粒轮廓特征变换研究 |
| |
| 引用本文： | 汪家道,孔宪梅.磨粒轮廓特征变换研究[J].润滑与密封,1997(3):8-10. |
| |
| 作者姓名： | 汪家道 孔宪梅 |
| |
| 作者单位： | 摩擦学国家重点实验室，摩擦学国家重点实验室，摩擦学国家重点实验室，摩擦学国家重点实验室 清华大学，北京100084，清华大学，北京100084，清华大学，北京100084，清华大学，北京100084 |
| |
| 基金项目： | 本课题由国家自然科学基金资助 |
| |
| 摘 要： | 本文提出了一个磨粒轮廓特征变换的方法，该方法解决了磨粒图像的平移、旋转和大小不变性，同时，该变换所得的结果能较为全面地反映了磨粒的轮廓特征，这使磨粒识别具有较高的准确率成为可能。文中给出了一个利用此变换方法所得的轮廓特征作为神经网络输入的磨粒识别实例。
|
| 关 键 词： | 磨粒 磨粒识别 神经网络 摩擦学 图像识别 |
Study on Translation for the Profile Feature of Wear Particles |
| |
| Wang Jiadao Kong Xianmei Zhang Song Chen Darong.Study on Translation for the Profile Feature of Wear Particles[J].Lubrication Engineering,1997(3):8-10. |
| |
| Authors: | Wang Jiadao Kong Xianmei Zhang Song Chen Darong |
| |
| Affiliation: | National Tribology Laboratory |
| |
| Abstract: | This paper describes a method on translation for the profile of wear particles, which solve sinvariabiliry on displacement, rotation and scale of wear particles image. The results given by this method can reflect accurately the profile features of wear particles, which is the base to raise the accuracy for the wear particles recognition. An example using the method to get the input of a neural network for wear particles recognition is given in this paper. |
| |
| Keywords: | Wear Particle Wear Particles Recognition Neural Network |
| 本文献已被 CNKI 维普 等数据库收录！ |
