Dai Huafeng, Wang Jyunrong, Zhong Quan, Chen Taogen, Liu Hao, Zhang Xuegang, Lu Rongsheng
Tsinghua University, Beijing, China.
LCFC (Hefei) Electronics Technology Co., Ltd., Hefei, Anhui, China.
Sci Rep. 2024 Mar 4;14(1):5259. doi: 10.1038/s41598-024-52378-9.
In numerous applications, abnormal samples are hard to collect, limiting the use of well-established supervised learning methods. GAN-based models which trained in an unsupervised and single feature set manner have been proposed by simultaneously considering the reconstruction error and the latent space deviation between normal samples and abnormal samples. However, the ability to capture the input distribution of each feature set is limited. Hence, we propose an unsupervised and multi-feature model, Wave-GANomaly, trained only on normal samples to learn the distribution of these normal samples. The model predicts whether a given sample is normal or not by its deviation from the distribution of normal samples. Wave-GANomaly fuses and selects from the wave-based features extracted by the WaveBlock module and the convolution-based features. The WaveBlock has proven to efficiently improve the performance on image classification, object detection, and segmentation tasks. As a result, Wave-GANomaly achieves the best average area under the curve (AUC) on the Canadian Institute for Advanced Research (CIFAR)-10 dataset (94.3%) and on the Modified National Institute of Standards and Technology (MNIST) dataset (91.0%) when compared to existing state-of-the-art anomaly detectors such as GANomaly, Skip-GANomaly, and the skip-attention generative adversarial network (SAGAN). We further verify our method by the self-curated real-world dataset, the result show that our method is better than GANomaly which only use single feature set for training the model.
在众多应用中，异常样本难以收集，这限制了成熟的监督学习方法的使用。基于生成对抗网络（GAN）的模型以无监督且单一特征集的方式进行训练，通过同时考虑正常样本与异常样本之间的重构误差和潜在空间偏差而被提出。然而，其捕捉每个特征集输入分布的能力有限。因此，我们提出一种无监督多特征模型Wave-GANomaly，仅在正常样本上进行训练以学习这些正常样本的分布。该模型通过给定样本与正常样本分布的偏差来预测其是否正常。Wave-GANomaly融合并从WaveBlock模块提取的基于小波的特征和基于卷积的特征中进行选择。WaveBlock已被证明能有效提升图像分类、目标检测和分割任务的性能。结果表明，与现有诸如GANomaly、Skip-GANomaly和跳跃注意力生成对抗网络（SAGAN）等先进异常检测器相比，Wave-GANomaly在加拿大高级研究所（CIFAR）-10数据集（94.3%）和改进的美国国家标准与技术研究院（MNIST）数据集（91.0%）上实现了最佳的平均曲线下面积（AUC）。我们通过自行整理的真实世界数据集进一步验证了我们的方法，结果表明我们的方法优于仅使用单一特征集训练模型的GANomaly。
