25 Jul 2023
 | 25 Jul 2023
Status: this preprint is open for discussion and under review for Atmospheric Measurement Techniques (AMT).

Improved RepVGG Ground-Based Cloud Image Classification with Attention Convolution

Chaojun Shi, Leile Han, Ke Zhang, Hongyin Xiang, Xingkuan Li, Zibo Su, and Xian Zheng

Abstract. Clouds greatly impact the earth's radiation prediction, hydrological cycle, and climate change. Accurate automatic recognition of cloud shape based on ground-based cloud image is helpful to analyze solar irradiance, water vapor content, and atmospheric motion, and then predict photovoltaic power, weather trends, and severe weather changes. However, the appearance of clouds is changeable and diverse, and its classification is still challenging. In recent years, convolution neural network(CNN) has made great achievements in ground-based cloud image classification. However, traditional CNN has a poor ability to associate long-distance clouds, so extracting the global features of cloud images is difficult. Therefore, a ground-based cloud image classification method based on improved convolution neural network RepVGG and attention mechanism is proposed in this paper. Firstly, the proposed method increases the RepVGG residual branch and obtains more local detail features of cloud images through small convolution kernels. Secondly, an improved channel attention module is embedded after the residual branch fusion, which can effectively extract the global features of the cloud images. Finally, the linear classifier is used to classify the ground cloud images. In addition, the warm-up method is introduced to optimize the learning rate in the training stage of the proposed method, and it is lightweight in the inference stage, which can avoid over-fitting and accelerate the convergence speed of the model. The proposed method in this paper is evaluated on MGCD and GRSCD ground-based cloud image datasets, and the experimental results show that the accuracy of this method reaches 98.15 % and 98.07 %, respectively, which exceeds other most advanced methods, and proves that CNN still has room for improvement in ground-based cloud image classification task.

Chaojun Shi et al.

Status: open (until 23 Oct 2023)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on egusphere-2023-1094', Anonymous Referee #1, 04 Aug 2023 reply
    • AC1: 'Reply on RC1', Hongyin Xiang, 29 Aug 2023 reply
    • AC2: 'Reply on RC1', Hongyin Xiang, 29 Aug 2023 reply

Chaojun Shi et al.

Chaojun Shi et al.


Total article views: 272 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
211 49 12 272 3 2
  • HTML: 211
  • PDF: 49
  • XML: 12
  • Total: 272
  • BibTeX: 3
  • EndNote: 2
Views and downloads (calculated since 25 Jul 2023)
Cumulative views and downloads (calculated since 25 Jul 2023)

Viewed (geographical distribution)

Total article views: 263 (including HTML, PDF, and XML) Thereof 263 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
Latest update: 30 Sep 2023
Short summary
We propose a method CloudRVE, which improves the convolution neural network model RepVGG, and applies it to the ground-based cloud image classification task. In this paper, we introduce the CloudRVE network framework and its composition in detail through words and diagrams, and verify that CloudRVE is superior to the previous advanced methods.