| 基于BP神经网络的粘接质量检测判别技术研究 |
| |
| 引用本文： | 孙秋菊,钟莉娟. 基于BP神经网络的粘接质量检测判别技术研究[J]. 数字社区&智能家居, 2009, 0(18) |
| |
| 作者姓名： | 孙秋菊 钟莉娟 |
| |
| 作者单位： | 信阳师范学院物理电子工程学院; |
| |
| 摘 要： | 该文针对粘接质量检测,运用BP算法和MATLAB软件,设计了用于粘接质量模式识别的神经网络。该网络能够对粘接质量合格与否进行比较精确的判别,判别正确率达到了98%。神经网络训练成功后,用未经训练的数据检验神经网络的检测能力,识别正确率达到了91.3%,充分证明了该网络模型的准确有效性。
|
| 关 键 词： | 模式识别 神经网络 粘接质量 MATLAB |
The Research of Recognition Technology of Adhesion Quality Test Based on BP Neural Network |
| |
| SUN Qiu-ju,ZHONG Li-juan. The Research of Recognition Technology of Adhesion Quality Test Based on BP Neural Network[J]. Digital Community & Smart Home, 2009, 0(18) |
| |
| Authors: | SUN Qiu-ju ZHONG Li-juan |
| |
| Affiliation: | SUN Qiu-ju,ZHONG Li-juan (Xinyang Normal University,College of Physics , Electronic Engineering,Xinyang 464000,China) |
| |
| Abstract: | Aiming at the need of the classification in the adhesion quality test, the text designs a neural network that can be used for adhesion quality identification by using the BP lgorithm and MATLAB software. The net designed in the text can distinguish the adhesion quality is eligible or not accurately, the correct ratio is 98 percent. After the neural network is trained successfully, we verify its identification ability using the data that is not used in the training; the correct identification ratio is reach ... |
| |
| Keywords: | pattern recognition neural network adhesion quality MATLAB |
| 本文献已被 CNKI 等数据库收录！ |
|
