基于改进BP网络的机械零件图纸信息识别
|
摘要
在通过对ＢＰ算法加以改进；使其性能有所提高，收敛速度加快的基础上，提出了一种新的ＣＡＰＰ系统零件信息的输入方法；即零件的图纸经过扫描仪扫描获得数字信息、力。载神经网络模式识别，同时将数字信息还原成ＣＡＰＰ系统中计算机直接读懂的信息，并加载网络测试加以验证．
It is one of the important methods of pattern recognition to apply neural networks to target classification. Forward propagation multi-layers neural networks and its BP algorithm are used widely. In this article, some measures are taken to improve BP network, and to make its performance better and its convergence speed quicker. A new input method of mechanical parts information in CAPP system is raised. That is, the contents of the parts drawing are scanned into digital information by scanner and loaded on neural network recognizing the pattern, at the same time, the digital information is reverted which the computer can understand directly, furthermore, tested by loading neural network.
It is one of the important methods of pattern recognition to apply neural networks to target classification. Forward propagation multi-layers neural networks and its BP algorithm are used widely. In this article, some measures are taken to improve BP network, and to make its performance better and its convergence speed quicker. A new input method of mechanical parts information in CAPP system is raised. That is, the contents of the parts drawing are scanned into digital information by scanner and loaded on neural network recognizing the pattern, at the same time, the digital information is reverted which the computer can understand directly, furthermore, tested by loading neural network.
引文
[1]聂伟荣.基于改进BP网络的地震动信号目标识别[J].南京理工大学学报(自然科学版),2000,24(1):40-41.
[2]郭子华.基于神经网络和遗传算法地图自动输入新算法[J].模式识别与人工智能,1998,11(2):48-50
[3] Don D. A Syntactic/Geometric Approach to Recognition of Dimension in Engineering Machine Drawing [J]. Computer Vision. Graphics andlmageProcessing, 1989, 47 (2): 106-110.
[4] Suzuki S. Symbol Recognition of Equipment Location Maps With Raster Operation [A], Proceeding of Lot. Workshop on Industrial Application of Machine Vision and Machine Intelligence [C]. 1987, 198-203.
[5] Joseph S H. Processing of Engineering Line Drawing for Automatic Inputting to CAD [J]. Pattern Recognition 1992, 1: 36-42.
[2]郭子华.基于神经网络和遗传算法地图自动输入新算法[J].模式识别与人工智能,1998,11(2):48-50
[3] Don D. A Syntactic/Geometric Approach to Recognition of Dimension in Engineering Machine Drawing [J]. Computer Vision. Graphics andlmageProcessing, 1989, 47 (2): 106-110.
[4] Suzuki S. Symbol Recognition of Equipment Location Maps With Raster Operation [A], Proceeding of Lot. Workshop on Industrial Application of Machine Vision and Machine Intelligence [C]. 1987, 198-203.
[5] Joseph S H. Processing of Engineering Line Drawing for Automatic Inputting to CAD [J]. Pattern Recognition 1992, 1: 36-42.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心