The manufacturing processes of Printed Circuit Boards (PCBs) have become increasingly complex, and even minor defects can significantly impair product performance and yield. Accurate identification of PCB defects is therefore crucial but remains challenging. Given the low resolution, small target size, and diverse nature of PCB surface defects, this study proposes a novel YOLO MSES-SPDConv (MS-YOLO) network based on an improved YOLOv10 framework to achieve more accurate and efficient detection of minor PCB defects with a smaller model size. Firstly, to address the performance bottleneck of the C2F module in the head network of YOLOv10 when processing complex PCB features, we introduce the Multiscale Edge Strengthening (MSES) structure to replace the traditional C2F module. The MSES structure employs a multi-branch design for multidimensional learning of minor PCB features enhanced by edge information. During inference, it is simplified to a single branch to retain high-frequency information and reduce memory consumption. Secondly, in the neck of the network, we incorporate the Improved SPDConv module to minimize the loss of fine-grained information and enable multiscale feature extraction for PCB defect images. Lastly, we optimize the network structure using the Layer Adaptive Magnitude-based Pruning (LAMP) method and design a dual-distillation strategy to further refine the improved YOLOv10 algorithm. This strategy leverages two distinct network heads for knowledge transfer, enhancing the learning effectiveness of the student model and improving accuracy while reducing model size. Experimental results demonstrate that the proposed MS-YOLO outperforms several state-of-the-art models, achieving a mean Average Precision (mAP@50) of 98. 9%. This validates the significant improvements of MS-YOLO in detection accuracy, operational efficiency, and practical applicability.

The efficient and accurate detection of foreign objects invading railway tracks is of paramount importance in safeguarding the safety of train operations. Addressing the issue of the problem of the low efficiency of the existing foreign objects detection methods, this work proposes a fast railway foreign objects intrusion detection method based on cascaded convolution neural network and Overhaul Knowledge Distillation. First, a two-stage cascaded convolution neural network is built.The first stage can identify whether the railway images are intruded by foreign objects or not. This is achieved by a light weight image classification network. The use of lightweight classification network can reduce the use of the object detection network, thus improving the overall efficiency of the railway foreign objects intrusion detection method in this paper. Secondly, this paper employs the Overhaul Knowledge Distillation algorithm to train a lightweight network that is supervised by a larger network, so that the lightweight network constructed in this paper also has satisfactory image classification performance. Finally, the YOLOv3 object detection network is used to detect the foreign object image classified by the first level network. The experimental results demonstrate that the accuracy of the image classification network proposed in this paper is competitive to the classical backbone network, and the FPS is about 50-70 higher than the comparison method.

Printed circuit board (PCB) surface defect detection is an essential part of the PCB manufacturing process. Currently, advanced CCD or CMOS sensors can capture high-resolution PCB images. However, the existing computer vision approaches for PCB surface defect detection require high computing effort, leading to insufficient efficiency. To this end, this article proposes a local and global context-enhanced lightweight CenterNet (LGCL-CenterNet) to detect PCB surface defects in real time. Specifically, we propose a two-branch lightweight vision transformer module with local and global attention, named LGT, as a complement to extract high-dimension features and leverage context-aware local enhancement after the backbone network. In the local branch, we utilize coordinate attention to aggregate more powerful features of PCB defects with different shapes. In the global branch, Bi-Level Routing Attention with pooling is used to capture long-distance pixel interactions with limited computational cost. Furthermore, a Path Aggregation Network (PANet) feature fusion structure is incorporated to mitigate the loss of shallow features caused by the increase in model depth. Then, we design a lightweight prediction head by using depthwise separable convolutions, which further compresses the computational complexity and parameters while maintaining the detection capability of the model. In the experiment, the LGCL-CenterNet increased the mAP@0.5 by 2% and 1.4%, respectively, in comparison to CenterNet-ResNet18 and YOLOv8s. Meanwhile, our approach requires fewer model parameters (0.542M) than existing techniques. The results show that the proposed method improves both detection accuracy and inference speed and indicate that the LGCL-CenterNet has better real-time performance and robustness.

Abstract The efficient and accurate detection of foreign objects invading railway tracks holds paramount importance in safeguarding the safety of train operations. Focusing on the problem of the low efficiency of the existing foreign objects detection methods, this work proposes a fast railway foreign objects intrusion detection method based on cascaded convolution neural network and knowledge distillation. First, a two-stage cascade convolution neural network is built.The first stage can identify whether the railway images are intruded by foreign objects or not. This is achieved by a light weight image classification network.In the second stage,YOLOv3 is employed to classify and locate the objects in the intruded railway image. The use of lightweight classification network can reduce the use of the object detection network, thus improving the overall efficiency of the railway foreign objects intrusion detection method in this paper. Secondly, this paper employs the Overhaul knowledge distillation algorithm to train a lightweight network that is supervised by a larger network, so that the lightweight network constructed in this paper also has satisfying image classification performance. Finally, the YOLOv3 object detection network is used to detect the foreign object image classified by the first level network. The experimental results demonstrate that the accuracy of the image classification network proposed in this paper is competitive to the classical backbone network, and the FPS is about 50–70 higher than the comparison method.

Foreign object intrusion is one of the main causes of train accidents that threaten human life and public property. Thus, the real-time detection of foreign objects intruding on the railway is important to prevent the train from colliding with foreign objects. Currently, the detection of railway foreign objects is mainly performed manually, which is prone to negligence and inefficient. In this study, an efficient two-stage framework is proposed for foreign object detection in railway images. In the first stage, a lightweight railway image classification network is established to classify any input railway images into one of two classes: normal or intruded. To enable real-time and accurate classification, we propose an improved inverted residual unit by introducing two improvements to the original inverted residual unit. First, the selective kernel convolution is used to dynamically select kernel size and learn multiscale features from railway images. Second, we employ a lightweight attention mechanism, called the convolutional block attention module, to exploit both spatial and channel-wise relationships between feature maps. In the second stage of our framework, the intruded image is fed to the foreign object detection network to further detect the location and class of the objects in the image. Experimental results confirm that the performance of our classification network is comparable to the widely used baselines, and it obtains outperforming efficiency. Moreover, the performances of the second-stage object detection are satisfying.

中国特色高水平高职学校和专业建设计划,简称"双高计划",被业内人士称之为高职院校版的"双一流大学工程",是我国对职业教育的一次重要制度设计.在教育部遴选国家级双高校和双高专业群的同时,各省教育厅也启动了省级双高计划的培育工作,将在技术技能人才培养高地、高水平专业群和高水平双师队伍等多方面全面打造,以提升高职院校的办学条件和办学水平.本文以广州铁路职业技术学院为例,分析"双高计划"建设为教师发展带来的难得的发展机遇与挑战.

Noise pollution in the closed space such as railway carriage is an important problem because the noise pollution seriously affects comfort and health of people in the closed space. We propose the method to detection, integration, and optimization of acoustic field simulation in the closed space. First, we analyze the acoustic field distribution in the virtual 3D close space. We use harmonic sound wave propagation in the closed space and present the distribution according to geometric analysis. Second, we introduce Delaunay triangulation and k-means clustering into visualization to form the quiet zone and show it in 3D perspective. Our method used acoustic simulation to develop the sound barrier system. The simulation results show that our method can improve the analysis of the noise problem in the closed space.

27.5kV交联聚乙烯单芯电力电缆在我国高铁牵引供电系统中应用广泛.目前27.5kV高压电缆及附件故障率比较高,难以对电缆缺陷和故障准确定位以及实时监测,影响到牵引供电设施的安全运行.通过研究运用物联网技术、地理信息系统(GIS)技术、空间索引技术等信息技术,对电缆信息有效、实时、自动化的采集,采用数字化运维和智能化监控手段,实现对电缆全路径在线监测、故障判断和精准定位.

随着我国高速铁路迅猛发展，为人们出行提供了很大的便捷，但是当列车高速行驶穿越城镇时，会产生严重的环境噪声问题。对于声屏障等现有的传统高速铁路噪音隔离措施，存在着装置相对固定、噪音消除强度不够与环境变化要求未能很好适应等不足，本文提出的一种基于主动噪音监测和干扰的铁路噪音屏蔽装置，具备了高适应性、安装方便、易于应用推广等优势，是一种有效运用人工智能计算模型的非传统降噪措施。

数字孪生是将物理空间和赛博空间相互影射的新一代技术.在5G、物联网、大数据和人工智能技术的支撑下,数字孪生首先被应用到工业制造领域,将物理空间中的实体和流程实时转化为赛博空间中的数字孪生体.数字孪生体可用来对物理空间实体和流程进行仿真分析和预测演化.将数字孪生技术应用到铁路交通运输应用场景中,研究构建了铁路交通智慧管理数字孪生模型,为铁路交通智慧管理系统提出新思路.

At present, the image mining is mainly based on its local and key features, which focuses on its texture and statistical grayscale features, but it focuses on its edge and shape features rarely. However, the contour is also an important feature for image shape recognition. In this paper, a good target image contour coding algorithm was adopted, and an LCV segmentation model with good image boundary acquisition capability that can reflect the target image contour features was selected for the original image contour segmentation. The detailed features analysis of the contour coding algorithm was carried out through the experiments; the experimental results showed that the algorithm was a significant technological breakthrough in image feature extraction and recognition.

Active Appearance Model (AAM) is a valid statistical algorithm for human face alignment, which composes of two parts, namely AAM sub-space model and AAM searching process. AAM has three sub-spaces, namely form sub-space, texture sub-space and surface sub-face. As it is stated above, AAM is built on the dot distribution model. Different from ASM, it not only conducts statistical analysis of textures through the shape information, but also explores the connection between the shape and the texture. During the training period, it mainly targets at finding out the connection between model parameter changes and changes of shapes and textures. In terms of new image searching, model parameters can be continuously adjusted according to the connection so as to make the composite image approximate the new one as much as possible. The shape and the texture at the moment are regarded at the shape and the texture of the new image.

2D-to-3D conversion that would be a solution of the lack of 3D contents has been a worthy and challenging research field. In this paper, we propose a computer interactive conversion method to capture components which is used to generate 3D sequences. First, we divide the key frame into foreground and background, and then label the objects by convenient computer interactive operation. Depth information of objects is labeled after segmentation. Second, we use object tracking technique which synthesizes the advantages of kernel-based mean shift tracker and contour tracker to accomplish object depth capture for non-key frame. Finally, all the 3D information is prepared to render 3D sequences. After all, we propose our future work direction: a 2D-to-3D system which can generate 3D sequence interactively.

2D-to-3D conversion that would be a solution of the lack of 3D contents has been a worthy and challenging research field. In this paper, we propose a computer interactive conversion method to capture components which is used to generate 3D sequences. First, we divide the key frame into foreground and background, and then label the objects by convenient computer interactive operation. Depth information of objects is labeled after segmentation. Second, we use the Multi-Feature Object Tracking method which synthesizes the advantages of kernel-based mean shift tracker and contour tracker to accomplish object depth capture for non-key frame. Finally, all the 3D information is prepared to render 3D sequences.

China has the largest number of hearing-impaired people in the worldover millions of Chinese suffered from hearing-impaired. However, most television programs don't offer Chinese subtitles or the sign language, leaving the hearing disabled people unable to obtain the information and enjoy recreational activities as ordinary people do. Therefore, the sign language on television is very important. Comparing to the ordinary computer, the calculation and processing power of the digital TV terminals is nowhere near a computer. Small storage space and low performance restrict the ability of displaying 3D graphics on the television. Therefore, this article uses a 3D graphics mesh simplification method which based on edge collapse, to simplified 3D graphics, and apply it to the sign language on the digital television, to achieve the goal of displaying the 3D sign language graphics smoothly and intuitively on the digital TV terminals.

The invention discloses a system for identifying hands based on complexion detection and background elimination. The system is divided into three parts, namely video input, gesture analytic and gesture identification, and in most cases, the gesture analysis comprises the following three phases: detecting, namely determining feature data to be extracted by selecting a gesture model which is suitable for study contents, and detecting gestures from an input image by a detection algorithm; tracking, namely tracking the motion of the hands in real time by using a tracing algorithm; and performing feature extraction, namely extracting relevant feature data which are separated from the image. In addition, the invention also discloses a method for identifying the hands based on the complexion detection and the background elimination. By synthetic sign languages, deaf people can use various kinds of information and participate in various social activities, so that the system and the method have significant application value and social meaning.

The embodiment of the invention discloses a service network node configuration method under a district end-based network. The method comprises the following steps: detecting the average demand of each street in a region in a time period, and judging whether the average demand of each street in the time period is larger than a preset threshold or not; when the average demand of a certain street in the region in the time period is larger than the preset threshold, setting a service network node server for the certain street; when the average demand of the certain street in the region in the time period is smaller than the preset threshold, calculating to determine which streets in the region need to be served by the service network node server according to the genetic algorithm; and after calculating a plurality of the streets and configuring the service network node server, calculating the position of the center point of the plurality of the streets according to the absolute center point algorithm in the graph theory, wherein the position of the center point is the position for configuring the service network node server. By adopting the algorithm of the embodiment, service network nodes can be rapidly calculated and configured.

To establish the human body model to analyze the heat and moisture transfer on body surface, a new explicit definition of rational L-recursion surface is given and the L-recursion surfaces, in Grassmann spaces, are constructed by using blossom method of the homogeneous normal pyramid form. Based on our human body model, the balance theory of garment simulation, the heat and moisture transfer balance equations, called ICAD-balance equations are obtained. The balance theory of garment simulation integrally studies the complex system of human body–fabric–environment. At the same time, the method of obtaining the heat and moisture transfer balance equations is also based on the mass conservation law, the energy conservation law and the Fish law of capillarity. A finite volume method is employed to solve the ICAD-balance equations.

In this paper, we propose a novel depth map generation method. After a series of pre-treatment process, image quality capture and bilateral filtering, K-means clustering method has been used for classification of background and front objects. Then the depth map could be generated directly depend on the predeterminate model which is given a forehand, finally the correct depth map can be vividly created base on the layer Stratifying. The experiment result shows that the depth map directly represent the depth information and also earn good subjective evaluation.