| 引用本文: | 何智恒;柴欣生,陈春霞;陈润权.生活用纸纤维原生状态识别程序的界面设计和算法改进[J].造纸科学与技术,2018,37(4):. |
| Hezhiheng;Chaixinsheng,Chenchunxia;Chenrunquan.Interface Design and Algorithm Improvement of Fiber Original State identification Program for Tissue Paper[J].Paper Science and Technology,2018,37(4):. |
|
| 摘要: |
| 基于PCA-BP神经网络(Principal Component Analysis-Back Propagation Neural Network)的算法，采用LabView编写了程序界面交互友好的操作软件系统，用于对生活用纸的纤维原生状态识别。界面设计主要赋予了程序简便操作，人机友好的特点；在算法中通过引入国质检总局提供的条件判定方法对前期的纤维原生状态识别模型进行了改进。结果表明，程序界面具有清晰简洁、交互友好、可操作性强的特点，可对处于模糊区域的样品的纤维原生状态进行准确的识别。本程序对于市售卫生纸使用的安全性，防止假冒伪劣和制假售假，具有重要的意义。 |
| 关键词: 界面设计;主成分分析;BP神经网络;纤维原生状态 |
| DOI： |
| 分类号: |
| 基金项目:制浆造纸工程国家重点实验室开放基金资助项目 |
|
| Interface Design and Algorithm Improvement of Fiber Original State identification Program for Tissue Paper |
|
Hezhiheng;Chaixinsheng and Chenchunxia;Chenrunquan
|
|
State Key Laboratory of Pulp and Paper Engineering, South China University of Technology;State Key Laboratory of Pulp and Paper Engineering, South China University of Technology,National Paper Products Quality Supervision and Inspection Center
|
| Abstract: |
| Based on the algorithm of PCA-BP neural network (Principal Component Analysis-Back Propagation Neural Network), a program-interactive friendly operation software system was programmed by LabView to identify the fiber''s original state of tissue paper. The interface design mainly provides the advantages of concise operation and friendly interaction between the computer and operator. The conditional identification rule from the National Quality Inspection Center is integrated in the algorithm to improve the performance of the identification model. The results show that the program interface is clear and concise, with friendly interaction and strong operability. It can accurately identify the fiber''s original state of the sample in the fuzzy area. This procedure is of great significance for the safety of tissue paper on the market and prevention of counterfeiting. |
| Key words: Interface Design; Principal Component Analysis; Back-Propagation Neural Network; Fiber Original State |