| 基于图像感兴趣区域融合特征的PM2.5浓度预测方法 |
| PM2.5 Concentration Prediction Based on Fusion Features of Image Regions of Interest |
| 投稿时间：2021-06-21 修订日期：2021-08-21 |
| DOI：DOI:10.19316/j.issn.1002-6002.2022.04.22 |
| 中文关键词: 大气污染监测 PM2.5浓度 感兴趣区域 传统图像处理 深度学习 融合特征 |
| 英文关键词:air pollution monitoring PM2.5 concentration region of interest traditional image processing deep learning fusion features |
| 基金项目:中国科学院战略性先导科技专项（A类-XDA19040202）；北京信息科技大学其他纵向项目（20190193） |
|
| 通讯作者:王立志* 中国科学院大气物理研究所 中国科学院东亚区域气候-环境重点实验室, 北京 100029 |
| 摘要点击次数: 518 |
| 全文下载次数: 5 |
| 中文摘要: |
| 通过图像预测PM2.5浓度的准确性,在很大程度上取决于模型所选用的特征参数。为丰富特征参数的表达,设计了一种基于图像传统特征与深度特征充分融合的PM2.5浓度预测方法。首先,根据不同PM2.5浓度下的成像差异,选定图像感兴趣区域,解决图像尺寸过大导致的模型运算效率较低问题。然后,针对所选取的局部图像,利用传统图像处理方法手动设计并提取图像浅表视觉特征,同时利用卷积神经网络自动提取图像深层语义特征。最后,将两种特征融合,交由卷积神经网络的全连接层实现对PM2.5浓度的回归预测。预测误差比对结果显示,相比使用单种特征,使用融合特征能够有效提高模型的预测性能。 |
| 英文摘要: |
| The accuracy of PM2.5 concentration prediction using image largely depends on the features extracted by the model.In order to enrich the expression of feature parameters,a PM2.5 concentration prediction method based on the fusion of traditional image features and depth features is designed.Firstly,based on the imaging differences under different PM2.5 concentrations,the region of interest in the image is selected to solve the problem of low model efficiency caused by the excessively large image size.Secondly,the traditional image processing method is used to manually design and extract the superficial visual features of the selected partial image,and the convolutional neural network is used to automatically extract the deep semantic features of the image.Finally,the two features are merged,and the regression prediction of PM2.5 concentration is achieved through the fully connected layer of the convolutional neural network.Comparison on the prediction error of using a single feature and a fusion feature shows that the prediction performance of the model can be effectively improved by the fusion feature. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
