Wasserstein Generative Adversarial Networks Based Data Augmentation for Radar Data Analysis

Abstract

Ground-based weather radar can observe a wide range with a high spatial and temporal resolution. They are beneficial to meteorological research and services by providing valuable information. Recent weather radar data related research has focused on applying machine learning and deep learning to solve complicated problems. It is a well-known fact that an adequate amount of data is a positively necessary condition in machine learning and deep learning. Generative adversarial networks (GANs) have received extensive attention for their remarkable data generation capacity, with a fascinating competitive structure having been proposed since. Consequently, a massive number of variants have been proposed; which model is adequate to solve the given problem is an inevitable concern. In this paper, we propose exploring the problem of radar image synthesis and evaluating different GANs with authentic radar observation results. The experimental results showed that the improved Wasserstein GAN is more capable of generating similar radar images while achieving higher structural similarity results.

A radar waveform recognition method based on ambiguity function generative adversarial network data enhancement under the condition of small samples

“…Literature [37] shows that WGAN-GP in a deep generation network is ideal for radar data enhancement. It uses a Wasserstein range and GP mechanism to solve the problem of gradient disappearance, and can generate high-quality radar signals.…”

“…In the experiment, AFCAN adopts Adam optimiser, the learning rate is 0.0002, each batch has 24 samples, and the signal-to-noise ratio is 0 dB. The number of samples at four levels is set as 50, 200, 800 and 3000 samples [37], which can be regarded as a process from insufficient to sufficient samples. At each level, the effectiveness of the AFGAN data enhancement method when using small samples is compared and analysed through experiments.…”

Spatial Prediction Using Random Forest Spatial Interpalation with Sample Augmentation: A Case Study for Precipitation Mapping

“…Generative adversarial network(GAN) has been practical and effective in HSIs classi cation (Zhu et al, 2018). And improved Wasserstein GAN is morecapable of generating similar radar images while achieving higher structural similarity results (Lee et al, 2020) To sum up, the combination of data augmentation and ML has been applied to the classi cation and prediction aspects of geosciences, especially in the classi cation aspcets. However, it is relatively less used in the prediction aspects.…”