Most of the existing vehicle-loaded video surveillance systems aim at recoding and analyzing the outer circumstance of the vehicles. The inside drivers' and passengers' motion and status, however, are also greatly related to vehicles' running safety. In this paper, we combined Spatial-Temporal Entropy Image (STEI) with Independent Component Analysis (ICA) to deal with in-vehicle scenes, which are usually contain dynamic background and subtle motions. The entropy which denotes the temporal discrimination of local areas is computed. Meanwhile, the short-term accumulation of the independent dynamic component is subtracted to extract drivers and passengers from an in-vehicle scene. The quantitative evaluation from experimental results show that the improved STEI achieves a better detection performance with a reasonable cost of tiny increase on false alarms.

There are many children with autism in today's society who have social interaction difficulties and language impairments. To further understand the neural basis of autism, in this paper, we investigate the similarities and differences of EEG characteristics between children with autism spectrum disorder (ASD) and typically developing (TD) children on a brain-computer interface system based on steady-state visual evoked potentials (SSVEP-BCI). We acquired the SSVEP signals elicited at different frequencies (8-11Hz) in 9 channels from three children with ASD and two TD children. For SSVEP signals analysis, we extracted the power spectral density (PSD), the canonical correlation coefficient (CCA), and the extended canonical correlation coefficient (eCCA). To further analyzed the difference between signals of children, we calculated the SSVEP recognition rates in different channels using the one-dimensional convolutional neural networks (CNN). The experiment result shows that children with ASD responded more significantly to SSVEP stimulation than children with TD, and most strongly at the Oz channel.

Background: Integrating human neural signals with computational vision systems offers a promising route toward more robust visual recognition, yet supporting mixed-granularity recognition, where both coarse- and fine-grained categories must be distinguished within a unified system, remains challenging due to the heterogeneous feature spaces of electroencephalography (EEG) and visual data. Methods: We propose “Align and Fuse,” a two-stage Transformer-based framework. Stage 1 constructs a shared semantic space using a hardness-aware multimodal supervised contrastive loss with Hard Negative Weighting to explicitly target confusable class pairs. Stage 2 employs a multimodal Transformer with co-attention to fuse the aligned features for classification. Results: On the 80-class EEG-ImageNet benchmark, our framework achieved 91.12% Top-1 accuracy under a temporally separated control protocol, improving over the corresponding vision-only (89.08%) and Standard Transformer (89.95%) baselines. Under the original stratified random split, it achieved 92.56% Top-1 accuracy; on the 40-class EEGCVPR dataset, accuracy reaches 95.82%. Cross-subject experiments yield 90.92% average Top-1 accuracy on four unseen subjects, and Grad-CAM analysis suggests that aligned EEG signals shift the model’s attention toward semantically relevant regions. Conclusions: Coupling hardness-aware alignment with decoupled multimodal fusion supports EEG-augmented recognition by leveraging complementary stimulus-related information under the evaluated protocols. Because EEG features are required at inference time, the framework is positioned as a human-in-the-loop EEG-augmented recognition system rather than a standalone vision model.