| 基于深度卷积神经网络的塑料垃圾分类研究 |
| |
| 引用本文： | 吴晓玲,黄金雪,何文海.基于深度卷积神经网络的塑料垃圾分类研究[J].塑料科技,2020,48(4):86-89. |
| |
| 作者姓名： | 吴晓玲 黄金雪 何文海 |
| |
| 作者单位： | 广州商学院,广东 广州 511363 |
| |
| 基金项目： | 广东省普通高校重点科研平台和科研项目;广东省创新强校工程项目 |
| |
| 摘 要： | 针对塑料垃圾识别分类问题,设计标准卷积层、残差卷积层1和残差卷积层2。通过实验仿真验证各卷积层的优势,将其应用于塑料垃圾分类模型的不同部位中。实验表明,塑料垃圾分类模型的分类准确率高于传统的HOG、LBP分类模型约2%~3%,可为塑料垃圾分类工作提供一定的参考。
|
| 关 键 词： | 塑料 卷积层 残差网络 神经网络 |
Research on Plastic Waste Classification Based on Deep Convolutional Neural Network |
| |
| WU Xiao-ling,HUANG Jin-xue,HE Wen-hai.Research on Plastic Waste Classification Based on Deep Convolutional Neural Network[J].Plastics Science and Technology,2020,48(4):86-89. |
| |
| Authors: | WU Xiao-ling HUANG Jin-xue HE Wen-hai |
| |
| Affiliation: | (Guangzhou College of Commerce,Guangzhou 511363,China) |
| |
| Abstract: | For the problem of plastic waste identification and classification,standard convolutional layer,residual convolutional layer 1 and residual convolutional layer 2 are designed.The advantages of each convolutional layer are verified by experimental simulation,and they are applied to different parts of the plastic waste classification model.Experiments show that the classification accuracy of the plastic waste classification model is about 2%~3%higher than the traditional HOG and LBP classification models,which can provide a certain reference for the classification of plastic waste. |
| |
| Keywords: | Plastic Convolutional layer Residual network Neural network |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
