Anomaly detection is a vibrant research direction in controller area networks, which provides the fundamental real-time data transmission underpinning in-vehicle data interaction for the internet of vehicles. However, existing unsupervised learning methods suffer from insufficient temporal and spatial constraints on shallow features, resulting in fragmented feature representations that compromise model stability and accuracy. To improve the extraction of valuable features, this paper investigates the influence of clustering constraints on shallow feature convergence paths at the model level and further proposes an end-to-end intrusion detection system based on efficient deep embedded subspace clustering (EDESC-IDS). Following the standard learning approach, continuous messages are encoded into two-dimensional data frames via a frame builder, which are then input into an extended convolutional autoencoder for extracting shallow features from high-dimensional data. On this basis, the dual constraints of these output features and the embedding clustering module facilitate end-to-end training of the EDESC-IDS in various attack scenarios. Extensive experimental results show that such a system exhibits significant detection performance on four types of attack datasets, including DoS, Gear, Fuzzy, and RPM, with precision, recall, and F1 scores consistently above 97.79%, while maintaining a false negative rate (FNR) and an error rate (ER) below 2.22%.

In person reidentification (PReID) tasks, challenges such as occlusion and small object sizes frequently arise. High-precision object detection methods can accurately locate small objects, while attention mechanisms help focus on the strong feature regions of objects. These approaches mitigate the mismatches caused by occlusion and small objects to some extent. This paper proposes a PReID method based on object detection and attention mechanisms (ODAMs) to achieve enhanced object matching accuracy. In the proposed ODAM-based PReID system, You Only Look Once version 7 (YOLOv7) was utilized as the detection algorithm, and a size attention mechanism was integrated into the backbone network to further improve the detection accuracy of the model. To conduct feature extraction, ResNet-50 was employed as the base network and augmented with residual attention mechanisms (RAMs) for PReID. This network emphasizes the key local information of the target object, enabling the extraction of more effective features. Extensive experimental results demonstrate that the proposed method achieves a mean average precision (mAP) value of 90.1% and a Rank-1 accuracy of 97.2% on the Market-1501 dataset, as well as an mAP of 82.3% and a Rank-1 accuracy of 91.4% on the DukeMTMC-reID dataset. The proposed PReID method offers significant practical value for intelligent surveillance systems. By integrating multiscale attention and RAMs, this method enhances both its object detection accuracy and its feature extraction robustness, enabling a more efficient individual identification process in complex scenes. These improvements are crucial for enhancing the real-time performance and accuracy of video surveillance systems, thus providing effective technical support for intelligent monitoring and security applications.