Lei Wang

Accurately forecasting air pollutants concentration can reduce health risks and provide an important reference for environmental governance. This study proposes a new deep learning model, GLA-Net, which aims to achieve high-precision prediction of the air quality index (AQI) of monitoring stations. Specifically, GLSTM-Block is designed to use the GAT module to capture dynamic spatial interaction of AQI, generate spatial semantic features, and then use LSTM to capture the temporal correlation characteristics of these spatial characteristics. This paper uses an LSTM network outside of GLSTM-Block to capture the original temporal characteristics of the input data. Then, the temporal characteristics of the LSTM output are added to the dynamic spatiotemporal features of the GLSTM-Block to obtain the final spatiotemporal features as the input of the subsequent temporal attention layer. The temporal attention layer uses a multi-head self-attention mechanism to focus on the impact of the spatiotemporal characteristics of historical air quality data on each prediction time step, and performs AQI prediction through a fully connected layer. Analysis based on measured data from Beijing, Shanghai and Shenzhen shows that the GLA-Net model has significant advantages in predicting single-step and multi-step changes in AQI. The study found that although the model has a large absolute error in predicting concentrations in highly polluted areas, it can better grasp the trend of changes. This feature is particularly evident in Beijing (AQI mean 64.289), with root mean square error (RMSE) of 12.716 and index of agreement (IA) of 0.983.

Inspired by classical works, when constructing local relationships in point clouds, there is always a geometric description of the central point and its neighboring points. However, the basic geometric representation of the central point and its neighborhood is insufficient. Drawing inspiration from local binary pattern algorithms used in image processing, we propose a novel method for representing point cloud neighborhoods, which we call Point Cloud Local Auxiliary Block (PLAB). This module explores useful neighborhood features by learning the relationships between neighboring points, thereby enhancing the learning capability of the model. In addition, we propose a pure Transformer structure that takes into account both local and global features, called Dual Attention Layer (DAL), which enables the network to learn valuable global features as well as local features in the aggregated feature space. Experimental results show that our method performs well on both coarse- and fine-grained point cloud datasets. We will publish the code and all experimental training logs on GitHub.