In recent years, the popularity and growth of Internet of Things (IoT) devices have led to an increase in network traffic. Detecting anomalous traffic in these networks is critical to maintaining network security and preventing attacks. In this paper, we propose a deep learning-based approach to detect anomalous traffic in IoT networks. Our approach combines convolutional neural networks (CNN), density-based spatial clustering of applications with noise (DBSCAN), and transfer learning. We evaluate our approach on a real-world dataset and achieve promising results.