Generative Adversarial Networks for anomaly detection in aerial images. (March 2023)

• Record Type: Journal Article
• Authors: Contreras-Cruz, Marco A.; Correa-Tome, Fernando E.; Lopez-Padilla, Rigoberto; Ramirez-Paredes, Juan-Pablo
• Is Part Of: Computers & electrical engineering. Volume 106(2023)
• Publication Date: 2023-03
• Subjects: Generative Adversarial Networks -- Anomaly detection -- Deep learning -- Adversarial learning and inference -- Aerial image analysis -- Automatic inspection; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2022.108470
• Languages: English
• ISSNs: 0045-7906

Abstract

Abstract: Generative Adversarial Networks (GANs) are commonly used as a system able to perform unsupervised learning. We propose and demonstrate the use of a GAN architecture, known as the fast Anomaly Generative Adversarial Network ( f -AnoGAN), to solve the problem of anomaly detection from aerial images. This architecture was previously applied to medical images and, in this work, we adapt it for use on satellite or aerial photographs. To test the effectiveness of this approach, we implemented anomaly detection schemes based on the Bi-directional Generative Adversarial Network (BiGAN), the image - z - image mapping ( izi ), the z - image - z ( ziz ) mapping, and a deep convolutional autoencoder (AE). The results show that the f -AnoGAN outperformed others, achieving AUC (area under the curve) values of 0.99 and 0.92 for urban and rural spaces image sets, respectively. Graphical abstract: Highlights: Identification of anomalies in aerial images with one-class training. The system automatically learns to discern unknown imagery. The selected architecture outperforms other one-class approaches that use generative adversarial networks. The anomaly detection is performed in less than 100 ms with accuracy higher than 90%.