The detection of surface defects on a printed circuit board (PCB) plays a crucial role in ensuring the quality of PCB products. To address the diversity of defect poses and the challenges associated with detecting small objects on PCB, this paper proposes an improved YOLOv7 model for small objects-oriented PCB defect detection. First, this paper improves the regression loss function of YOLOv7 by incorporating wise-IoU (WIoU) and replacing IoU with an outlier degree to develop a gradient-boosting allocation strategy, thereby increasing the network’s accuracy. Second, this paper proposes a coordinate attention dynamic mechanism that performs convolution operations with deformable convolutional networks v2 (DCNv2) using coordinate attention. This mechanism effectively suppresses redundant information. Finally, this paper proposes a dynamic head diverse (DyHead-d) module that prioritizes spatial awareness over scale awareness, building on DyHead. This module improves the network’s ability to localize small targets. Experimental results show that the WDC-YOLO achieves a mean average precision of 98.4% on public datasets, demonstrating a 3.1% improvement compared to the original network. The significantly enhanced detection accuracy meets the real-time requirements of PCB defect diagnosis, which is of great importance for quality control and cost reduction in PCB industrial production.
In this paper, a neural network time delay prediction method based on phase space reconstruction is presented. This method reconstructs one-dimensional chaotic time series in phase space according to the internal law through phase space reconstruction, and uses BP neural network algorithm to predict the time delay. Simulation experiments show that this method has good prediction performance.
With the rapid development of science and technology and the advent of the information age, the number of components used in electronic devices has increased sharply, making its internal circuit structure increasingly complex. Printed Circuit Boards (PCBs), as part of electronic devices, are becoming smaller and more integrated, resulting in a much greater increase in the probability of failure and the difficulty of detection. Therefore, to reduce the difficulty and cost of PCB fault diagnosis, it is very necessary to explore and study new PCB diagnosis methods. This paper first reconstructs the PCB dataset by ESRGAN, and then the CenterNet based on the center point is introduced and improved. The ResNeSt based on the segmentation attention mechanism is integrated with CenterNet to realize the PCBs fault diagnosis method based on the tiny object detection method. Experiments have proved that the method can achieve 99.42% mAP.
In high power PWM inverter system, EMC problem becomes more and more serious, common mode conducted electromagnetic interference (EMI) becomes the research hotspot. By studying the interference source and interference path of metro traction inverter system, this paper proposes to establish the equivalent model of metro inverter system for the whole circuit. By extracting the high-frequency parasitic parameters of metro inverter system structural parts through simulation software, the whole circuit simulation model is built, and the common mode conducted electromagnetic interference is simulated and measured. At 150KHz-30MHz, the simulated spectrum and the measured spectrum peak value are 92.489dBuA and 94.064dBuA, respectively. The peak error between the simulated spectrum and the measured spectrum is about 1.6dBuA, and there is obvious resonance at 2.1MHz and 2.07MHz, respectively. The simulated spectrum and the measured spectrum peak value and their variation trend are basically the same, the amplitude difference is very small, which proves the correctness of the high-frequency equivalent model and analysis of the whole circuit proposed in this paper. This method can be used as a feasible scheme to calculate and predict common-mode EMI of three-phase metro inverter system.
A time delay prediction method of train network based on wireless transmission is proposed. EMD is used to decompose the time delay series. The decomposed components with large sample entropy are DWT to form new components, in order to reduce the complexity of prediction. The components with similar sample entropy are combined into new components to reduce the amount of model calculation. Finally, each data component is predicted by particle swarm optimization LSSVM model. The simulation results show that the proposed method has high prediction accuracy.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.