Metal corrosion in high-risk areas, such as high-altitude cables and chemical factories, is very complex and inaccessible to people, which can be a hazard and compromise people’s safety. Embedding deep learning models into edge computing devices is urgently needed to conduct corrosion inspections. However, the parameters of current state-of-the-art models are too large to meet the computation and storage requirements of mobile devices, while lightweight models perform poorly in complex corrosion environments. To address these issues, a lightweight residual deep-learning model based on an encoder–decoder structure is proposed in this paper. We designed small and large kernels to extract local detailed information and capture distant dependencies at all stages of the encoder. A sequential operation consisting of a channel split, depthwise separable convolution, and channel shuffling were implemented to reduce the size of the model. We proposed a simple, efficient decoder structure by fusing multi-scale features to augment feature representation. In extensive experiments, our proposed model, with only 2.41 MB of parameters, demonstrated superior performance over state-of-the-art segmentation methods: 75.64% mean intersection over union (IoU), 86.07% mean pixel accuracy and a 0.838 F1-score. Moreover, a larger version was designed by increasing the number of output channels, and model accuracy improved further: 79.06% mean IoU, 88.07% mean pixel accuracy, and 0.891 F1-score. The size of the model remained competitive at 8.25 MB. Comparison work with other networks and visualized results were used for validation and to determine the accuracy of metal corrosion surface segmentation with limited resources.

Addressing the optimal path planning problem encountered by swarm of unmanned aerial vehicle (UAV) in three-dimensional space under multiple constraints, the Multi-population Adaptive Cuckoo Search and Grey Wolf Optimizer (MACSGWO) integrates Multi-Population (MP) strategies and adaptive evolutionary optimizer including the enhanced Adaptive Grey Wolf Optimizer (AGWO) and adaptive Cuckoo search (ACS). The optimizer strategically divides the initial population into multiple sub-groups, enabling each sub-group to independently iterate. During the iteration process, the algorithm adaptively adjusts parameters based on the optimal fitness values obtained by each sub-group after each iteration. The iteration cycle is divided into two stages: during the global exploration phase, each sub-group autonomously executes AGWO and periodically shares the fitness information of the Alpha wolf with other sub-groups, accelerating convergence. In the subsequent local optimization phase, MACSGWO dynamically decides whether to initiate ACS based on the disparity in the best fitness of each sub-group after each iteration, assisting the algorithm in escaping local optima. In experiments involving various complex benchmark functions and swarm path planning scenarios, MACSGWO demonstrated significant superiority in solution stability, convergence speed, and optimal convergence value compared to multiple existing variant algorithms. The integration of MACSGWO with the best relay UAV selection strategy further optimized the communication efficiency within the swarm. MACSGWO ensures the efficient resolution of UAV swarm path planning problems, providing robust support for optimization challenges in complex, multi-constraint scenarios.

Printed Circuit Boards (PCBs) are essential components in electronic devices, making defect detection crucial. PCB surface defects are diverse, complex, low in feature resolution, and often resemble the background, leading to detection challenges. This paper proposes the YOLOv8_DSM algorithm for PCB surface defect detection, optimized based on the three major characteristics of defect targets and feature map visualization. First, to address the complexity and variety of defect shapes, we introduce CSPLayer_2DCNv3, which incorporates deformable convolution into the backbone network. This enhances adaptive defect feature extraction, effectively capturing diverse defect characteristics. Second, to handle low feature resolution and background resemblance, we design a Shallow-layer Low-semantic Feature Fusion Module (SLFFM). By visualizing the last four downsampling convolution layers of the YOLOv8 backbone, we incorporate feature information from the second downsampling layer into SLFFM. We apply feature map separation-based SPDConv for downsampling, providing PAN-FPN with rich, fine-grained shallow-layer features. Additionally, SLFFM employs the bi-level routing attention (BRA) mechanism as a feature aggregation module, mitigating defect-background similarity issues. Lastly, MPDIoU is used as the bounding box loss regression function, improving training efficiency by enhancing convergence speed and accuracy. Experimental results show that YOLOv8_DSM achieves a mAP (0.5:0.9) of 63.4%, representing a 5.14% improvement over the original model. The model’s Frames Per Second (FPS) reaches 144.6. To meet practical engineering requirements, the designed PCB defect detection model is deployed in a PCB quality inspection system on a PC platform.

Crowd stability is a research hotspot in the area of emergency management, while crowd density is the most direct and effective basis for crowd stability analysis. Therefore, it is significant to study the trend of crowd density and make some reliable predictions. This paper introduces advanced image processing technology, increasing the number of columns in a traditional convolutional neural network (CNN) to improve the accuracy of crowd counting, and further obtains high-precision crowd density values through image correction. Then, combined with the mature support vector regression (SVR) algorithm, this paper explores the kinematic characteristics of pedestrians, and proposes a novel crowd density prediction model. By predicting the position of group pedestrians in next few moments, the prediction of crowd density changes in public places can be achieved, which provides data support for crowd stability analysis. Finally, case study is carried out with actual scenario of Shanghai Hongqiao Airport, and the feasibility of the model is verified.

This paper proposes and demonstrates a single-line discontinuous track recognition system by associating the track recognition problem of a humanoid robot with the lane detection problem. The proposal enables the robot to achieve stable running on the single-line discontinuous track. The system consists of two parts: the robot end and the graphics computing end. The robot end is responsible for collecting track information and the graphics computing end is responsible for high-performance computing. These two parts use the TCP for communication. The graphics computing side uses PolyLaneNet lane detection algorithm to train the track image captured from the first perspective of the darwin-op2 robot as the data set. In the inference, the robot end sends the collected tracking images to the graphics calculation end and uses the graphics processor to accelerate the calculation. After obtaining the motion vector, it is transmitted back to the robot end. The robot end parses the motion vector to obtain the motion information of the robot so that the robot can achieve stable running on the single-line discontinuous track. The proposed system realizes the direct recognition of the first perspective image of the robot and avoids the problems of poor stability, inability of identifying curves and discontinuous lines, and other problems in the traditional line detection method. At the same time, this system adopts the method of cooperative work between the PC side and the robot by deploying the algorithm with high computational requirements on the PC side. The data transmission is carried out by stable TCP communication, which makes it possible for the robot equipped with weak computational controllers to use deep-learning-related algorithms. It also provides ideas and solutions for deploying deep-learning-related algorithms on similar low computational robots.

In this letter, we propose a novel lightweight X-ray image contraband segmentation network, XSNet, which integrates State Space Models (SSM) with Convolutional Neural Networks (CNNs) to achieve a significant trade-off between segmentation accuracy and lightweight design for computer-aided X-ray security check. The model is built based on the encoder-decoder framework. Specifically, we design an Multi-scale Convolution Fusion (MCF) block for multi-scale information extraction and a Dual-branch State Space Model (DSSM) block to relieve the bias caused by the imbalance of single branch structure in feature extraction and maintain the capabilities of SSM in modeling long range pixel dependencies. In addition, we present two versions of the model in two different sizes called XSNet-s and XSNet-l respectively. The quantitative and qualitative evaluations on the public PIDray and PIXray datasets both show the superiority of two models in terms of mean Intersection over Union (mIoU) and FLOPs.