Social network analysis has important research significance in sociology, business analysis, public security, and other fields. The traditional Louvain algorithm is a fast community detection algorithm with reliable results. The scale of complex networks is expanding larger all the time, and the efficiency of the Louvain algorithm will become lower. To improve the detection efficiency of large-scale networks, an improved Fast Louvain algorithm is proposed. The algorithm optimizes the iterative logic from the cyclic iteration to dynamic iteration, which speeds up the convergence speed and splits the local tree structure in the network. The split network is divided iteratively, then the tree structure is added to the partition results, and the results are optimized to reduce the computation. It has higher community aggregation, and the effect of community detection is improved. Through the experimental test of several groups of data, the Fast Louvain algorithm is superior to the traditional Louvain algorithm in partition effect and operation efficiency.

In real-world scenarios, the facial expression recognition task faces several challenges, including lighting variations, image noise, face occlusion, and other factors, which limit the performance of existing models in dealing with complex situations. To cope with these problems, we introduce the CoT module between the CNN and ViT frameworks, which improves the ability to perceive subtle differences by learning the correlations between local area features at a fine-grained level, helping to maintain the consistency between the local area features and the global expression, and making the model more adaptable to complex lighting conditions. Meanwhile, we adopt an adaptive learning method to effectively eliminate the interference of noise and occlusion by dynamically adjusting the parameters of the Transformer Encoder’s self-attention weight matrix. Experiments demonstrate the accuracy of our CoT_AdaViT model in the Oulu-CASIA dataset as (NIR: 87.94%, VL: strong: 89.47%, weak: 84.76%, dark: 82.28%). As well as, CK+, RAF-DB, and FERPlus datasets achieved 99.20%, 91.07%, and 90.57% recognition results, which achieved excellent performance and verified that the model has strong recognition accuracy and robustness in complex scenes.

It is an important part of security inspection to carry out security and safety screening with X-ray scanners. Computer vision plays an important role in detection, recognition, and location analysis in intelligent manufacturing. The object detection algorithm is an important part of the intelligent X-ray machine. Existing threat object detection algorithms in X-ray images have low detection precision and are prone to missed and false detection. In order to increase the precision, a new improved Mask R-CNN algorithm is proposed in this paper. In the feature extraction network, an enhancement path is added to fuse the features of the lower layer into the higher layer, which reduces the loss of feature information. By adding an edge detection module, the training effect of the sample model can be improved without accurate labeling. The distance, overlap rate, and scale difference between objects and region proposals are solved using DIoU to improve the stability of the region proposal’s regression, thus improving the accuracy of object detection; SoftNMS algorithm is used to overcome the problem of missed detection when the objects to be detected overlap each other. The experimental results indicate that the mean Average Precision (mAP) of the improved algorithm is 9.32% higher than that of the Mask R-CNN algorithm, especially for knife and portable batteries, which are small in size, simple in shape, and easy to be mistakenly detected, and the Average Precision (AP) is increased by 13.41% and 15.92%, respectively. The results of the study have important implications for the practical application of threat object detection in X-ray images.

Deploying prohibited-item detectors on resource-constrained X-ray security inspection terminals is not equivalent to selecting the smallest available YOLO scale: excessive compression can reduce feature capacity, whereas medium-scale detectors may retain avoidable computational redundancy. This study therefore investigates a deployment-oriented operating point through coordinated backbone compression, feature compensation, and class-sensitive optimization. YOLO-SMV and YOLO-EMV are developed from YOLOv8m and YOLO11m by combining MobileNetV3-Small backbone reconstruction, SE/ECA-based channel recalibration, and a VF-BCE objective for difficult and underrepresented categories. A three-seed full-factorial study on YOLO11m shows that VF-BCE provides the largest individual accuracy gain and that ECA repeatedly recovers part of the performance lost in the compressed VF-BCE pathway. The accuracy-oriented ECA+VF-BCE configuration reaches $0.93842\pm 0.00720$ mAP50, whereas the deployment-oriented YOLO-EMV reaches $0.92744\pm 0.00768$; the latter trades $1.098\pm 0.061$ percentage points of mAP50 for a reduction from 20.03 M to 12.04 M parameters and from 67.9 G to 28.8 G FLOPs. Under the common seed-41 SIXray protocol, YOLO-EMV also achieves higher mAP50 than standard YOLO11n, YOLO11s, and YOLO11m, demonstrating that the selected operating point is not reproduced simply by choosing a smaller baseline. Published SIXray results are reported separately as protocol-aware literature context rather than as a cross-paper ranking. Additional OPIXray and PIDray experiments provide multi-benchmark evidence for the component interactions under heavy occlusion and long-tailed class distributions. The deployment-oriented YOLO-EMV model has also been integrated into customs security inspection equipment.

Transformer structures have demonstrated outstanding skills in the deep learning space recently, significantly increasing the accuracy of models across a variety of domains. Researchers have started to question whether such a sophisticated network structure is actually necessary and whether equally outstanding results can be reached with reduced inference cost due to its complicated network topology and high inference cost. In order to prove the Mixer's efficacy on three datasets Speech Commands, UrbanSound8k, and CASIA Chinese Sentiment Corpus this paper applies amore condensed version of the Mixer to an audio classification task and conducts comparative experiments with the Transformer-based Audio Spectrogram Transformer (AST)model. In addition, this paper conducts comparative experiments on the application of several activation functions in Mixer, namely GeLU, Mish, Swish and Acon-C. Further-more, the use of various activation functions in Mixer, including GeLU, Mish, Swish, and Acon-C, is compared in this research through comparison experiments. Additionally, some AST model flaws are highlighted, and the model suggested in this study is improved as a result. In conclusion, a model called the Audio Spectrogram Mixer, which is the first model for audio classification with Mixer, is suggested in this study and the model's future directions for improvement are examined.