Yolo V4 for Advanced Traffic Sign Recognition With Synthetic Training Data Generated by Various GAN

• Published in: IEEE access 2021, Vol.9, p.97228-97242
• Main Authors: Dewi, Christine, Chen, Rung-Ching, Liu, Yan-Ting, Jiang, Xiaoyi, Hartomo, Kristoko Dwi
• Format: Article
• Language: English
• Subjects: Accuracy; Artificial neural networks; Data models; Datasets; DCGAN; Detectors; Feature extraction; Generative adversarial networks; Image enhancement; Image recognition; LSGAN; Object recognition; Pictures; Quality assessment; synthetic images; traffic sign; Traffic signs; Training; Training data; WGAN; Yolo V3; Yolo V4
• ISSN: 2169-3536; 2169-3536
• DOI: 10.1109/ACCESS.2021.3094201

Convolutional Neural Networks (CNN) achieves perfection in traffic sign identification with enough annotated training data. The dataset determines the quality of the complete visual system based on CNN. Unfortunately, databases for traffic signs from the majority of the world's nations are few. In this scenario, Generative Adversarial Networks (GAN) may be employed to produce more realistic and varied training pictures to supplement the actual arrangement of images. The purpose of this research is to describe how the quality of synthetic pictures created by DCGAN, LSGAN, and WGAN is determined. Our work combines synthetic images with original images to enhance datasets and verify the effectiveness of synthetic datasets. We use different numbers and sizes of images for training. Likewise, the Structural Similarity Index (SSIM) and Mean Square Error (MSE) were employed to assess picture quality. Our study quantifies the SSIM difference between the synthetic and actual images. When additional images are used for training, the synthetic image exhibits a high degree of resemblance to the genuine image. The highest SSIM value was achieved when using 200 total images as input and 32\times 32 image size. Further, we augment the original picture dataset with synthetic pictures and compare the original image model to the synthesis image model. For this experiment, we are using the latest iterations of Yolo, Yolo V3, and Yolo V4. After mixing the real image with the synthesized image produced by LSGAN, the recognition performance has been improved, achieving an accuracy of 84.9% on Yolo V3 and an accuracy of 89.33% on Yolo V4.