Automatic Classification of NVST Short-exposure Data Based on Deep Learning

Abstract

The New Vacuum Solar Telescope is one of the most important solar telescopes in China. However, in the process of reconstructing high-resolution solar data, the data may be distorted by thin film interference fringes. In this paper, an automatic classification method based on deep learning is proposed to distinguish fringe-contained data and fringe-free data, employing the Adaptive Wavelet Transform to construct the sample data set while transfer learning is utilized to train the classification model. The experimental results show that classification accuracy of the proposed method can reach up to 99.3%. This proposed method can make the high-resolution reconstruction pipeline run automatically whether the solar data contains fringes or not.

A Sports Training Video Classification Model Based on Deep Learning

“…On this basis, the radial distortion and tangential distortion in the nonlinear model are fully considered, the Rodrigues rotation equation is used to reduce the number of optimization parameters, and the steepest descent method and LM optimization method are used to solve the accurate parameters, respectively. Because the actual lens in the video is not ideal perspective imaging, with varying degrees of distortion, this kind of distortion can be divided into radial distortion and tangential distortion [3]. In order to describe the imaging model accurately, two parameters are used to describe the lens radial distortion and tangential distortion.…”