RFID multi-tag dynamic detection measurement based on conditional generative adversarial nets and YOLOv3

Abstract

The quality of multi-tag imaging greatly affects the effective detection of multi-tag. When multi-tag moves rapidly, the image may have serious dynamic blur, and tags can not be detected efficiently. In previous work, it is generally assumed that blur kernel and noise stationary to improve image quality. However, the dynamic deblurring of Radio Frequency Identification (RFID) multi-tag imaging is an ill-posed inverse problem. In this paper, firstly, blur-sharp multi-tag image pairs are made by superimposing and averaging the adjoin random frames. Then, we propose blind deblurring for dynamic RFID multi-tag imaging based on conditional generative adversarial nets (CGANs), which adds perceptual loss and content loss to generator to make image sharper. Finally, tags are detected by YOLOv3 in real time in end-to-end manner. Experimental results demonstrate that PSNR is at least 0.56dB higher and speed is at least 31.25 % faster than that of the current improved convolution neural networks (CNN). CGANs can remove image blur better, which has great superiority in the field of dynamic multi-tag imaging. In addition, YOLOv3 detects multi-tag quickly, thereby improving the detection accuracy.

Generative adversarial network–assisted image classification for imbalanced tire X-ray defect detection

“…Recently, generative adversarial network (GAN) becomes a promising data augmentation method and attracts increasing attention (Goodfellow et al, 2014; Li et al, 2021b). Various variants of GAN have been developed such as conditional GAN (Li et al, 2021a), Wasserstein generative adversarial network (WGAN) (Xiong et al, 2020; Zhang et al, 2019), info generative adversarial network (Gong et al, 2020), variational autoencoder generative adversarial network (VAE GAN) (Hu and Sun, 2021), and bidirectional generative adversarial network (BiGAN) (Zhou et al, 2020). The GAN structure includes a generator network and a discriminator network.…”

“…The synthesized samples are used to expand the minority class, thus changing the imbalanced dataset to be balanced (Liu et al, 2022). Due to its good data augmentation performance, GAN has been recently applied to insufficient and imbalanced problems such as reliability assessment (Li et al, 2019a), prediction model (Li and Mao 2021a; Zhang et al, 2022), fault diagnosis (Li et al, 2019b; Zareapoor et al, 2021), and defect detection (Liu et al, 2020, 2021a, 2021b; Xuan et al, 2019).…”