激光与光电子学进展, 2021, 58 (4): 0410016, 网络出版: 2021-02-24
基于改进生成对抗网络和MobileNetV3的带钢缺陷分类 下载: 1010次
Strip Defect Classification Based on Improved Generative Adversarial Networks and MobileNetV3
图像处理 缺陷检测 图像分类 生成对抗网络 数据增强 image processing defect detection image classification generative adversarial networks data augmentation
摘要
针对数据集样本数量较少会影响深度学习检测效果的问题,提出了一种基于改进生成对抗网络和MobileNetV3的带钢缺陷分类方法。首先,引入生成对抗网络并对生成器和判别器进行改进,解决了类别错乱问题并实现了带钢缺陷数据集的扩充。然后,对轻量级图像分类网络MobileNetV3进行改进。最后,在扩充后的数据集上训练,实现了带钢缺陷的分类。实验结果表明,改进的生成对抗网络可生成比较真实的带钢缺陷图像,同时解决深度学习中样本不足的问题;且改进的MobileNetV3参数量是原有参数量的1/14左右,准确率为94.67%,比改进前提高了2.62个百分点,可在工业现场对带钢缺陷进行实时准确的分类。
Abstract
Aiming at the problem that the small number of samples in the dataset will affect the effect of deep learning detection, a strip defect classification method based on improved generative adversarial networks and MobileNetV3 is proposed in this paper. First, a generative adversarial network is introduced, and the generator and discriminator are improved to solve the problem of category confusion and realize the expansion of the strip defect data set. Then, the lightweight image classification network MobileNetV3 is improved. Finally, it is trained on the expanded data set to realize the classification of strip defects. Experimental results show that the improved generative adversarial network can generate more real strip steel defect images and solve the problem of insufficient samples in deep learning. And the parameter amount of the improved MobileNetV3 is about 1/14 of that before improvement, and the accuracy is 94.67%, which is 2.62 percentage points higher than that before improvement. It can be used for accurate and real-time classification of strip steel defects in industrial field.
常江, 管声启, 师红宇, 胡璐萍, 倪奕棋. 基于改进生成对抗网络和MobileNetV3的带钢缺陷分类[J]. 激光与光电子学进展, 2021, 58(4): 0410016. Jiang Chang, Shengqi Guan, Hongyu Shi, Luping Hu, Yiqi Ni. Strip Defect Classification Based on Improved Generative Adversarial Networks and MobileNetV3[J]. Laser & Optoelectronics Progress, 2021, 58(4): 0410016.