- Current Issue
- Online First
- Most Downloaded Archive
-
2025,48(22):1-9, DOI:
Abstract:
In emergency communication scenarios, unmanned aerial vehicles (UAVs) serve as aerial data collection platforms that can be rapidly deployed to disaster-stricken areas such as those affected by earthquakes, floods, wildfires, mining accidents and battlefield environments. UAVs are capable of collecting data from wireless IoT devices and transmitting it to the command center, thereby improving the efficiency of rescue decision-making. In post-disaster scenarios, data transmission tasks impose higher requirements on both communication efficiency and data completeness. Meanwhile, the limited energy supply of UAVs makes it challenging to collect large volumes of data efficiently while ensuring the complete transmission of vital information. To address these issues, this paper investigates a wireless communication system assisted by a single UAV, which adopts a multi-user uplink communication mode and a fly-hover-communicate data collection pattern. A joint optimization problem is formulated for device association, UAV hovering location and bandwidth allocation, aiming to maximize the UAV′s coverage utility while minimizing its total energy consumption. First, to enhance the UAV′s coverage utility, a particle swarm optimization (PSO) algorithm initialized with K-means clustering is employed. Then, to minimize energy consumption, we propose a PSO-based two-stage optimization framework that alternately optimizes hovering positions and bandwidth allocation. In particular, a PSO variant incorporating Gaussian perturbation and differential mechanisms is designed for hovering position refinement. Simulation results demonstrate that the proposed method effectively improves both coverage utility and energy efficiency. The coverage utility increased by 13.15% compared to the K-means algorithm, while energy consumption was reduced by 18.24% compared to the approach that only optimizes hovering locations and lower than the scheme where hovering energy and flight energy are optimized separately.
-
Lu Xiang, Gao Xinyue, Wang Du, Kang Qianzhuo, He Sheng
2025,48(22):10-19, DOI:
Abstract:
There are many sensors for civil aviation engine health monitoring. The proper choice of sensors will directly affect the prediction effect of engine remaining useful life. A sensor selection method based on optimal feature selection is proposed and Informer algorithm is used to predict the remaining useful life, which improves the prediction accuracy. Firstly, the differential clustering algorithm is used to classify the real flight conditions, and the health factors are constructed from the degradation mechanism of civil aviation engine, and the regression tree model is established with the data of cruise stage to select important sensors. Finally, the remaining useful life of civil aviation engine is predicted based on Informer algorithm. Using NASA′s newly released civil aviation engine degradation database under real flight conditions, the experimental results show that the root mean square error of prediction results decreases by 14% and the average scoring function decreases by 29% compared with no sensor selection. Compared with the traditional selection method based on sensor degradation trend or sensor data difference, the root mean square error decreases by 10% and 8% respectively, and the average scoring function decreases by 48% and 27% respectively. Compared with CatBoost, LightGBM, XGBoost, BiLSTM and Transformer algorithms, the accuracy of the proposed remaining life prediction method is improved by 36%, 24%, 14%, 6% and 5%, respectively.
-
Wei Yanan, Xie Jun, Lyu Kai, Zhu Rong, Li Cainian
2025,48(22):20-27, DOI:
Abstract:
A fault diagnosis framework for chemical processes is proposed, which combines K-means synthetic minority oversampling techniquewith conditional adversarial domain adaptation to address issues such as feature coupling caused by the temporal dependence of multivariate sensing data, data distribution shift caused by changes in operating conditions, and imbalanced sample data in chemical processes. Firstly, the original one-dimensional data is converted into multiple two-dimensional time window data using time window segmentation technology. Within these windows, the Kmeans SMOTE method is used to expand the minority class fault samples. The expanded samples can retain the complete temporal fault features, and this algorithm can also reduce the number of generated noise samples; then, domain adaptation techniques are used to align the feature distributions of the source domain and the target domain, reducing the distribution differences between the two and enabling the fault diagnosis model trained on the source domain to effectively identify fault categories under new operating conditions; finally, diagnostic experiments were conducted using fault data from the Tennessee Eastman process, and the effectiveness of the proposed method was validated by comparing its diagnostic rates with models such as CDAN, DANN, and JDA.
-
Liu Wenzhong, Lyu Yuhua, Tang Ruixin, Li Yingchun, Zhang Junjie
2025,48(22):28-36, DOI:
Abstract:
Aiming at the challenges of Doppler frequency shift, data bit transition, and wide spreading factor adaptability faced by DSSS signal acquisition technology in satellite-ground TT&C systems, this paper proposes a joint acquisition algorithm based on truncated PN code segmentation. By establishing a truncated PN code segmented parallel correlation architecture combined with an N-segment time-domain aggregation strategy, and implementing two-dimensional joint search through Fast Fourier Transform (FFT)-based frequency offset estimation, the algorithm effectively suppresses correlation peak attenuation caused by Doppler frequency shift (±800 kHz) and bit transitions. In terms of hardware architecture optimization, we analyze the sampling distribution patterns between truncated PN code segments and reconstruct parallel correlator resource allocation, reducing multiplier resource consumption by 73% compared to traditional dual-segment schemes. Simulation results demonstrate that under low spreading factor scenarios (SF=12), when truncation parameter N=32 is selected, the correlation peak-to-noise difference increases ninefold at chip SNR=-18 dB, achieving 89% detection probability (false alarm probability ≤2.5×10-5). FPGA implementation under identical conditions shows stable detection probability exceeding 85%. Featuring dynamically adjustable truncation parameters, this solution overcomes limitations of conventional fixed architectures in high-dynamic rate-switching scenarios, providing an engineering solution with high performance and low complexity for miniaturized spaceborne equipment and high-dynamic weak signal acquisition.
-
2025,48(22):37-47, DOI:
Abstract:
Time series data pose unique challenges for forecasting due to their complex long- and short-term patterns and multi-period characteristics. Traditional fixed-scale patching methods are difficult to effectively capture multi-period information, while periodicity and trend changes further increase the modeling difficulty, affecting forecast accuracy and interpretability. Based on the above problems, this paper proposes a multi-periodic model MDTDNet based on dual time-dependent learning. The model firstly acquires multi-period information adaptively by Fourier transform; then for each period, combined with the seasonal trend enhancement module, it improves the semantic expression of the subsequence through period patch design, frequency domain seasonal enhancement and time domain trend enhancement. A dual time-dependence module is introduced to realize feature extraction and fusion by capturing different time-dependent patterns inter- and intra-patchs by means of a long-term change extractor and a local fluctuation extractor, respectively. The experimental results show that the experimental results of the models in all six datasets outperform the current optimal model, PatchTST, with an average decrease of 9% in the mean square error (MSE) on ETTh1 dataset, with a maximum decrease of 10.14%.
-
Zhang Yimai, Ge Shuangchao, Huang Wentao, Wei Haibo, Feng Kaiqiang
2025,48(22):48-56, DOI:
Abstract:
In view of the problems such as poor adaptability to complex working conditions, insufficient nonlinear error compensation accuracy and low level of instrument intelligence in high oxygen and trace oxygen measurement in the field of gas detection, this paper designs a gas detection of BP neural network (MAPSO-BP) optimized by lightweight modified adaptive particle swarm algorithm. Instrument, the system builds a multi-sensor embedded platform to realize synchronous acquisition and fusion compensation of multi-parameters including temperature, pressure, flow and concentration, uses a microcontroller unit to run MAPSO-BP network in real time for nonlinear error correction, and develops an embedded human-computer interaction system based on Qt, supporting network communication, data storage, real-time alarm and cloud data synchronization functions enhance the intelligence level of the instrument. The system prototype designed in this paper is tested for system stability, anti-interference ability test and comparative experiments with the existing error compensation model. The results show that the error compensation method proposed in this paper and the designed system prototype are compared with the current mainstream error compensation method. The absolute error average of high oxygen and micro-oxygen measurements are respectively reduce by 20% and 25%; effectively solve the problem of low measurement accuracy of sensors under complex working conditions, and provide a feasible solution for the precision and low cost of gas sensors.
-
2025,48(22):57-65, DOI:
Abstract:
Aiming at the recognition failure of traditional 2D code detection methods for assisted navigation in complex industrial, logistics and transport scenarios, this study proposes an improved and lightweight YOLOv8-AR model to further enhance the recognition efficiency by using AR codes that are mature and can provide relative positional information. In terms of the network model, the backbone introduces an ultra-strong lightweight StarNet network to reduce the algorithmic computation of target detection; the C2f-EMSC module is optimised and constructed in the neck network to enhance the extraction of AR code features in complex environments and reduce the computational load at the same time; moreover, a lightweight detail-enhanced shared convolutional detection head LSDECD-H is proposed to improve the detail feature expression capability, so as to improve the detection accuracy of small targets and multi-targets. The experimental results show that the parametric and computational quantities of the model are 1.46M and 4.7GFLOPs, which are only 51% and 42% of the baseline, and the mAP is as high as 0.962 with high robustness in the case that the frame rate meets the real-time detection. It can quickly determine the position before decoding, and improve the recognition effect to achieve precise positioning, making it suitable for application scenarios like 2D code road sign navigation.
-
Zhou Jing, Dong Hui, Liu Xin, Tang Zhenyang
2025,48(22):66-77, DOI:
Abstract:
Ground current in high-voltage cable systems serves as a critical indicator for ensuring operational safety and stability. Accurate ground current prediction is crucial for fault prevention and enhancing grid reliability. To address the limitations of traditional time-series prediction models in terms of prediction accuracy and computational efficiency, this paper proposes a ground current prediction model based on the Mamba architecture, referred to as the Bi-EMamba model. Through a spatiotemporal dependency encoder, the model effectively captures long-term dependencies and spatial correlations in multivariate time series while maintaining high memory efficiency. To address the challenge of non-stationary data, the model incorporates Reversible Instance Normalization for data normalization and employs hyperparameter optimization to further improve prediction accuracy and generalization capability. Experimental results based on a dataset from a high-voltage cable line in Beijing demonstrate that Bi-EMamba outperforms existing benchmark models across various prediction horizons. Notably, in long-term forecasting scenarios, it exhibits superior generalization and computational efficiency. Compared to the current state-of-the-art model, iTransformer, Bi-EMamba achieves a 6.52% reduction in Mean Squared Error, a 3.21% reduction in Mean Absolute Error, and a 29.49% reduction in memory usage.
-
Lu Chang, Li Wenju, Wang Xubin, Yang Kang
2025,48(22):78-88, DOI:
Abstract:
Anomaly detection is an important task in modern industrial manufacturing. Due to the scarcity of abnormal samples, unsupervised detection that only requires normal sample training has attracted widespread attention. Among them, reconstruction based detection has been widely applied due to its concise and universal framework. However, existing algorithms are mostly based on image reconstruction, thus the discrimination between abnormal and normal regions is insufficient. At the same time, due to the strong uncertainty of abnormal positions and sizes in industrial images, existing algorithms cannot capture the overall structural features of samples well. This article proposed an industrial image anomaly detection algorithm based on feature reconstruction to address the above issues. Firstly, the use of pre trained models to extract multi-scale features as reconstruction objects avoids the situation where pixel space reconstruction has insufficient ability to distinguish anomalies; secondly, a global feature extraction module was designed to enhance the perception ability of the reconstruction model towards global features; finally, design a feature recombination strategy to jointly train the reconstruction model, in order to further enhance the model′s understanding of the overall structure of the samples and improve the reconstruction effect. A large number of experiments conducted on MVTec AD have shown that the proposed algorithm achieves an AUROC score of 98.7% in sample level anomaly detection and 98.3% in pixel level anomaly localization, both of which have reached state-of-the-art performance.
-
Li Yao, Huang Daqing, Yin Qiyuan, Xu Wenxiao, Wang Jiarui
2025,48(22):89-97, DOI:
Abstract:
The demand of modern warfare has propelled the application of multi-UAV collaboration in the military field. To address the problem of trajectory planning for multiple UAVs in a multi-threat mountainous environment with radar, artillery, and other threats, an improved Crested Porcupine Optimizer (CPO) algorithm, namely BCPO, is proposed.To tackle the issue of population diversity, the algorithm incorporates an initialization method combining opposition-based learning and good-point set initialization, which enhances the algorithm′s traversal capability. For the development phase of the CPO algorithm, a spiral search strategy based on adaptive small perturbations is introduced to further boost the global search performance. In the exploration phase of the CPO algorithm, a mutation triangle walk strategy based on the optimal random position is added to improve the local convergence efficiency. Additionally, a L-vy flight strategy with dynamic factors is proposed to help the algorithm achieve a better balance between global search and local optimization.Simulations on the CEC2017 test functions demonstrate that the BCPO algorithm has excellent convergence speed and accuracy. In a simulated mountainous environment, the BCPO algorithm shows an average performance improvement of 8.834%, 5.776%, and 21.828% compared to the CPO, GWO and WOA, respectively. Moreover, the stability of the algorithm is significantly enhanced. This method has good application value in solving multi-UAV trajectory planning problems in complex scenarios.
-
Yang Yuan, Chen Mingxia, Lu Junliang, Yan Yichuo
2025,48(22):98-111, DOI:
Abstract:
The artificial lemming algorithm is a newly proposed metaheuristic method that simulates four distinct behaviors of lemmings to effectively explore complex search spaces. However, it still suffers from premature convergence, limited exploration, lack of robustness, and susceptibility to local optima. To address these limitations, a multi-strategy improved artificial lemming algorithm is proposed. First, the Halton sequence is employed to generate a uniformly distributed initial population, enhancing global search capability. Second, an elite pool strategy combined with inertia weights is introduced to reduce excessive reliance on the best individuals and to improve the population′s ability to jump across the search space, thereby suppressing premature convergence. Finally, a nonlinear weighted golden sine strategy, combined with foraging behavior, is incorporated in the later stages of iteration to enhance the precision and stability of local search. To verify the performance of the improved algorithm, experiments are conducted on the CEC2017 benchmark function set, and statistical analysis is performed using the Wilcoxon rank-sum test. Experimental results show that the improved algorithm outperforms five comparative algorithms in terms of convergence speed, optimization accuracy, and stability. Compared to the original algorithm, the improved version achieves an average error reduction of 27.36% and a reduction of 36.99% in the average standard deviation. In three engineering optimization problems, the improved algorithm obtains the minimum objective function values in all cases, demonstrating better applicability and superiority over the comparative methods.
-
Yu Haiyue, Zhang Yingjun, Peng Chuyao, Wang Xiaohui
2025,48(22):112-118, DOI:
Abstract:
To meet the practical demands of autonomous ships for self-identifying navigation scenarios at sea, an adaptive recognition method for navigation scenarios in port waters based on electronic nautical charts is proposed. Firstly, by systematically analyzing the navigation characteristics of port waters, the ship navigation process is divided into eight scenarios: entering the port, leaving the port, navigating in the channel, entering and leaving the anchorage, anchoring, berthing, unberthing and mooring in the port. Secondly, a scene determination rule is established based on the characteristics of objects and their relative positions, and an adaptive recognition model integrating geometric relationships and dynamic parameters is constructed. Finally, the proposed method is verified based on the AIS historical trajectory data of 100 ships entering and leaving the waters of Yantai Port. The results show that the precision rate of this method for port navigation scenarios reaches 95.6%, and the delay is reduced to 12 ms. It can provide real-time necessary navigation scenario information and high-precision navigation environment perception support for the autonomous navigation system of ships sailing in the port area.
-
Li Jiaqi, Zheng Zhanheng, Zeng Qingning, Wang Jian
2025,48(22):119-128, DOI:
Abstract:
To address the limitations of the CAM++ model in feature extraction and recognition performance under complex acoustic conditions, this paper proposes TF-DCAM, a speaker verification model integrating dilated convolution and temporal-frequency multi-scale attention mechanisms. The model enhances feature representation through dilated residual convolution and a time-frequency adaptive refocusing unit to suppress redundant information. A temporal-frequency multi-scale attention module is introduced to improve sensitivity to key information via channel attention and cross-dimensional interaction. An adaptive masking temporal convolution module is further incorporated to model long-term dependencies effectively. Finally, a combination of contrastive loss functions is applied to jointly optimize the speaker embedding space. Experiments conducted on the CN-Celeb dataset show that TF-DCAM reduces EER and minDCF by 14.98% and 10.98% respectively, compared with the baseline. The model also demonstrates strong cross-lingual generalization on the VoxCeleb1 dataset. Results indicate that the proposed method significantly improves speaker verification performance and robustness while maintaining model efficiency.
-
Zhuang Jingying, Liu Lei, Yan Dongmei, Liang Chengqing
2025,48(22):129-140, DOI:
Abstract:
This paper proposed an Event-triggered Curriculum DDPG algorithm to improve the efficiency and accuracy of dynamic target tracking for UAVs. The algorithm combined Deep Deterministic Policy Gradient （DDPG） and YOLO object detection technology. It introduced an event-triggered mechanism to dynamically adjust the policy update frequency, enhancing decision-making efficiency. Additionally, it incorporated curriculum learning to create a staged training framework, gradually improving the UAV′s adaptability to complex tasks. Experimental results showed that the ETC-DDPG algorithm effectively improved the tracking efficiency of dynamic target tracking task and the stability of training process, and reduced the demand for computing resources, achieving a success rate of 93.357%. Compared with the original-DDPG algorithm and ETC-TD3 algorithm, the success rate is improved by 56.175% and 37.1% respectively. The collaborative effect of the event-triggered mechanism and curriculum learning was verified by ablation experiment, providing a reference for autonomous task execution in UAVs.
-
Xie Jinpeng, Teng Yulin, Li Hui, Wang Chenshan, Zhao Chaoyou
2025,48(22):141-151, DOI:
Abstract:
This paper, based on the principle of eddy electromagnetic induction, specifically the eddy current method and the magnetic method, combined with finite element simulation technology, simulates and analyzes the coil impedance and magnetic induction strength of the substrate surface, and verifies the feasibility of the inspection method. According to the study, the galvanized layer and coating thickness of the substrate surface have a significant effect on the coil impedance and magnetic induction strength. Based on this, this paper designs and develops the measurement circuit, algorithm and detection equipment for the thickness of galvanized layer and coating on the surface of ferromagnetic substrate. The experimental results show that the error of this equipment is less than 1% in the measurement of coating thickness on aluminum base surface, less than 6% in the measurement of coating thickness on iron base surface, and less than 6% in the measurement of insulator cap thickness. The coating layer detector developed in this paper can measure the thickness of the galvanized layer and the coating layer simultaneously, efficiently and accurately.
-
2025,48(22):152-165, DOI:
Abstract:
NdFeB (neodymium-iron-boron) permanent magnetic materials have been widely applied in modern industry and electronics due to their exceptionally high magnetic energy product and coercivity. However, in practical production, the compaction process— a critical stage in NdFeB manufacturing— still relies primarily on operator experience for setting process parameters. Variations in operator expertise and the inherent complexity of the production process often lead to unstable parameter settings, which adversely affect product quality and result in resource wastage. To accurately predict the process parameters during the powder compaction stage, this study proposes a Dynamic Layered Adjustment CatBoost (DLA-CatBoost) multi-output prediction model. Furthermore, an innovative hybrid optimization strategy, PSO-DSS-NSGA-III, which integrates particle swarm optimization to guide dynamic search space adjustment, is introduced to achieve multi-objective cooperative optimization of the model′s hyperparameters. Experimental results demonstrate that the DLA-CatBoost model optimized with the PSO-DSS-NSGA-III strategy exhibits excellent performance in multi-output prediction tasks, with a root mean square error (RMSE) ranging from 0.5 to 0.9, a mean absolute error (MAE) between 0.2 and 0.5, and a coefficient of determination (R2) between 0.96 and 0.99, thereby demonstrating its superior predictive capability and establishing it as an effective new approach for optimizing the process parameters in NdFeB compaction.
-
Li Qiang, Nan Xinyuan, Cai Xin, Yang Shiwei
2025,48(22):166-176, DOI:
Abstract:
Addressing the issue of false negatives and positives in non-motor vehicle irregular driving behavior detection with the current detection algorithm, an improved target detection algorithm, YOLO-CSSM, was proposed based on YOLOv8. The Backbone and Neck were enhanced with an SPD-Conv network module, which improved the model′s ability to learn from small targets and extract features under complex backgrounds. Subsequently, DCNv2 and SegNext Attention modules were integrated into the Backbone and Neck networks, respectively, to emphasize important feature information of non-motor vehicles and drivers, enhancing the model′s feature fusion capability. The MPDIoU was improved using the concept of the WIoU loss function, replacing the original CIoU loss function with Wise-MPDIoU to mitigate the imbalance between positive and negative samples. Validated on a self-built dataset of non-motor vehicle irregular driving behaviors, the improved YOLOv8 algorithm demonstrated precision, recall and mean average precision (mAP@0.5) of 89.4%, 90.0% and 93.6%, respectively, showing improvements of 3.3%, 5.4% and 4.5% over the traditional YOLOv8 algorithm, achieving better detection accuracy and effectiveness.And Based on the non-motorized vehicle violation detection algorithm, a non-motorized vehicle violation recognition and detection system was designed and developed using PyQT5.
-
Fu Pengfei, Xu Wei, Liu Huaiguang, Liu Yuanjiong, Liu Jinwei
2025,48(22):177-186, DOI:
Abstract:
This paper proposes a robust recognition method based on dual attention calibration to address the issues of insufficient multi-dimensional dynamic collaboration and fine-grained suppression in the attention mechanism under mask occlusion. The method dynamically calibrates the occlusion area in both channel and spatial dimensions. The channel dimension is based on global statistics to suppress abnormal responses of polluted channels, while the spatial dimension locates occluded areas and weakens their gradient propagation, achieving dynamic calibration from coarse-grained screening to fine-grained enhancement. On this basis, the weighted cross entropy loss and triplet loss are used to further guide the model to focus on the feature expression of locally unobstructed areas, thereby expanding the inter class feature distance interval. The experimental results show that the dual attention calibration mechanism proposed in this paper, through the synergistic effect of channel dimension feature screening and spatial dimension region enhancement, has improved accuracy by 6% and 7.2% respectively compared to the ArcFace algorithm in mask scenes of LFW and AgeDB-30, and by 7.3% on the real occlusion dataset MAFA dataset, verifying its recognition robustness in complex occlusion scenes.
-
Zhang Shuqing, Xiao Fan, Ge Chao
2025,48(22):187-197, DOI:
Abstract:
In response to the challenges posed by small target volumes and complex backgrounds in aerial remote sensing images, a lightweight object detection algorithm named ELS-RTDETR, based on enhancements to RT-DETR, has been proposed. This algorithm introduces and utilizes a new backbone network called LOB-Vovnet, which is an improved version based on the Vovnet network, to replace the original backbone network.Within the LOB-Vovnet architecture, a novel feature enhancement module named LRFF (Lightweight receptive field focus) has been designed to enhance the detection accuracy of small targets. To address complex background interference, an attention mechanism called SE (Squeeze-and-Excitation) based on adaptive channel extraction has been introduced.To strike a balance between model accuracy and size, LOB-Vovnet replaces some convolutions with depthwise separable convolutions. Extensive ablation experiments have been conducted to readjust the depth and width of the backbone network. In the AIFI section, a Cascaded Group Attention （CGA） mechanism has been introduced to effectively reduce computational redundancy in multi-head attention mechanisms.Regarding datasets, the RSOD dataset and NWPU VHR-10 dataset have been merged. Additionally, offline data augmentation techniques such as affine transformations and camera noise have been applied to the original data to make the training dataset more closely aligned with real-world applications.Experimental results indicate that the improved ELS-RTDETR model has shown a 2.7% increase in mAP@50 compared to the original model, with a reduction in model parameters by 32.9%. It has demonstrated good detection performance for challenging targets. Further validation of the enhanced method has been conducted on the SIMD dataset to verify its effectiveness.
-
Ji Xiaofei, Sun Yingchao, Song Jinghao
2025,48(22):198-205, DOI:
Abstract:
Existing pedestrian re-identification algorithms heavily rely on convolutional neural networks as the backbone, which often leads to an overemphasis on regions with prominent features while neglecting broader foreground features. This results in insufficiently rich global feature representations and inadequate attention to subtle discriminative features. To address these issues, we propose a feature-enhanced pedestrian ReID algorithm. The global branch utilizes position encoding and a multi-layer, multi-head attention structure to better leverage spatial context information, enhance the understanding of relative spatial positions, and effectively capture spatial structural information, thereby improving feature representation and global feature extraction capability. The local branch optimizes spatial attention using feature matrices associated with spatial vectors, enabling the capture of more compact general appearance features. Furthermore, by modeling the relationships between different channels, it strengthens feature expression in the channel dimension, highlighting distinctive features and improving the attention to discriminative characteristics. Finally, the model is trained using softmax loss, triplet loss, and center loss on the Market-1501 and DukeMTMC-ReID datasets. Experimental results demonstrate the effectiveness and superior performance of the proposed algorithm.
-
Xia Zhenghong, Zhong Jifei, Zhang Jun, Zhao Liang
2025,48(22):206-213, DOI:
Abstract:
To address the insufficient intelligent detection of surface damage on general aviation aircraft skins, an improved YOLOv11n-based detection algorithm is proposed. Firstly, the Adown downsampling mechanism is replaced to construct a multi-scale feature fusion architecture, achieving dynamic compression of redundant information through cross-level feature interaction and lightweight kernel design, thereby reducing model parameters and computational complexity. Secondly, a DySample dynamic upsampling strategy is designed, enhancing the model′s generalization across different scenarios via variable convolutional kernel deformation perception and multi-task gradient collaborative optimization. Furthermore, the FASSHead feature aggregation module is introduced, improving the algorithm′s recognition capability for complex damage areas through progressive semantic fusion and edge-aware constraints. Finally, a P2 small object detection layer is added, embedding high-resolution detection branches in shallow networks to enhance the capture of small objects and detailed damages. The improved algorithm was validated using a self-built dataset of general aviation skin surface damages. Results show that the precision reached 87.4%, recall reached 80.4%, and mAP attained 86.6%. Compared with the baseline model YOLOv11n, these metrics improved by 2.0%, 9.4% and 6.7% respectively, significantly enhancing the detection performance of skin surface damage and laying a theoretical foundation for an intelligent detection and maintenance system for general aviation aircraft.
-
Song Zhiqiang, Li Mingyang, Zhou Peng
2025,48(22):214-223, DOI:
Abstract:
Aiming at the limitation of face detection accuracy degradation when facing light changes or complex background in driver fatigue detection methods, an improved MTCNN network is proposed. By optimising the MTCNN network, the coordinate attention mechanism and batch normalisation algorithm are introduced in all three sub-networks to improve the model′s localisation accuracy of the driver′s face, enhance the convergence speed and stability of the network, and enhance the suppression of overfitting. The experimental results show that the accuracy of the improved MTCNN model on the fatigue driving dataset reaches 98.78%, which is 2.43% higher than that of the original model, and the number of parameters of the model is only 0.5 M, which has good face detection accuracy and deployability. In addition, combining the improved MTCNN model with the PFLD model, a reasonable fatigue parameter threshold is set based on the experiments, and a more accurate fatigue driving detection is achieved.
-
Cheng Rong, Zhu Wenzhong, Wang Wen
2025,48(22):224-234, DOI:
Abstract:
Crack detection is crucial in the maintenance of civil infrastructure. The many drawbacks of traditional manual visual inspection methods have led to the continuous development of crack detection methods. However, existing crack detection techniques face the challenges of complex backgrounds and feature diversity interference, and the high computational resource requirements. This study exploits the potential of Mamba for visual tasks and proposes an UltraLight CrackNet, which consists of a parallel lightweight visual Mamba block for efficiently modelling long-distance dependencies and extracting deep semantic features, a multi-scale residual visual state space block for enhanced multi-scale feature representation, and an enhanced semantics and detail infusion module for optimising skip connections within the encoder-decoder architecture. The experimental results show that our method requires only 0.13 M parameters and 1.96 G FLOPs, and achieves the optimal performance on DeepCrack and Crack500 datasets with ultra-lightweight model design, with the mean intersection over union (mIoU) of 87.85% and 77.92%, respectively, and obtains comparable results on SteelCrack dataset, and the number of parameters is 87.85% lower than that of the model with the smallest number of parameters among the available comparison models.
Volume 48, 2025 Issue 22
Research&Design
Application of Artificial Intelligence in Electronic Measurement
Theory and Algorithms
Information Technology & Image Processing
-
Xue Xianbin, Tan Beihai, Yu Rong, Zhong Wuchang
2024,47(6):1-7, DOI:
Abstract:
Urban intersections are accident-prone sections. For intelligent networked vehicles, it is very important to carry out risk detection and collision warning during driving to ensure the safety of driving. This paper proposes a traffic risk field model considering traffic signal constraints for urban intersections with traffic lights, and designs a three-level collision warning method based on this model. Firstly, a functional scenario is constructed according to the potential conflict risk points of urban intersections, and the vehicle risk field model is carried out considering the constraint effect of traffic signal. In order to solve the problem of collision warning, a three-level conflict area is proposed to be divided by the index, and the collision risk of the main vehicle is measured according to the position of the potential energy field around the main vehicle by calculating the corresponding field strength around the main vehicle. The experimental results show that the designed model can accurately warn the interfering vehicles entering the potential energy field of the main vehicle, the warning success rate can reach 100%, and the false alarm rate is only 3.4%, which proves the reliability and effectiveness of the proposed method.
-
Wei Jinwen, Tan Longming, Guo Zhijun, Tan Jingyuan, Hou Yanchen
2024,47(6):8-13, DOI:
Abstract:
To address the issue of low accuracy in indoor static target positioning with existing single-antenna ultra-high frequency RFID technology, this paper proposes a new RFID localization method based on an antenna boresight signal propagation model. The method first determines the height position of the target through vertical antenna scanning; secondly, it adjusts the antenna height to match that of the target and then performs stepwise rotational scanning to identify the target′s azimuth angle; furthermore, it utilizes a Sparrow Search Algorithm optimized back propagation neural network to establish a path loss model for ranging purposes; finally, it integrates the height, azimuth angle, and distance data to complete the target positioning. Experimental results show that in indoor environment testing, the proposed method has an average positioning error of 7.2 cm, which meets the positioning requirements for items in general indoor scenarios.
-
Zhang Fubao, Wu Ting, Zhao Chunfeng, Wei Xianliang, Liu Susu
2024,47(6):100-108, DOI:
Abstract:
In real-time detection of saw chain defects based on machine vision, factors like oil contamination and dust impact image brightness and quality, leading to a decrease in the feature extraction capability of the object detection network. In this paper, an automated saw chain defect detection method that combines low-light enhancement and the YOLOv3 algorithm is proposed to ensure the accuracy of saw chain defect detection in complex environments. In the system, the RRDNet network is used to adaptively enhance the brightness of the saw chain image and restore the detailed features in the dark areas of the image. The improved YOLOv3 algorithm is used for defect detection. FPN structure is added with a feature output layer, the a priori bounding box parameters are re-clustered using the K-means clustering algorithm, and the GIoU loss function is introduced to improve the object defect detection accuracy. Experimental results demonstrate that this approach significantly improve image illumination and recover image details. The mAP value of the improved YOLOv3 algorithm is 92.88%, which is a 14% improvement over the original YOLOv3. The overall leakage rate of the system eventually reduces to 3.2%, and the over-detection rate also reduces to 9.1%. The method proposed in this paper enables online detection of saw chain defects in low-light scenarios and exhibits high detection accuracy for various defects.
-
Zhan Huiqiang, Zhang Qi, Mei Jianing, Sun Xiaoyu, Lin Mu, Yao Shunyu
2024,47(6):123-130, DOI:
Abstract:
Aiming at the force test in low-speed pressurized wind tunnel, the original data source of aerodynamic characteristic curve is analyzed. With the balance signal, flow field state and model attitude as the main objects, combined with the test control process, the abnormal detection methods and strategies of the test data are studied from the dimensions of single point data vector, single test data matrix and multi-test data set in the same period, and an expert system for abnormal data detection is designed and developed based on this core knowledge base. The system inference engine automatically detects online during the test, and realizes the pre-detection and pre-diagnosis of the original data through data identification, rule reasoning, logical reasoning and knowledge iteration. The experimental application results show that the expert system is highly sensitive to the detection of abnormal types such as abnormal bridge pressure, linear segment jump point and zero point detection, which guides the direction of abnormal data analysis and improves the efficiency of problem data investigation.
-
Zhang Huimin, Li Feng, Huang Weijia, Peng Shanshan
2024,47(6):86-93, DOI:
Abstract:
A lightweight improved model CAM-YOLOX is designed based on YOLOX to address the issues of false alarms of land targets and missed detections of shore targets encountered in ship target detection in large scene Synthetic Aperture Radar(SAR)images in near-shore scenes. Firstly, embed Coordinate Attention Mechanism in the backbone to enhance ship feature extraction and maintain high detection performance; Secondly, add a shallow branch to the Feature Pyramid Network structure to enhance the ability to extract small target features; Finally, in the feature fusion network, Shuffle unit was used to replace CBS and stacked Bottleneck structures in CSPLayer, achieving model compression. Experiments are carried out on the LS-SSDD-v1.0 remote sensing dataset. The experimental results show that compared with the original algorithm, the improved algorithm in this paper has the precision increased by 5.51%, the recall increased by 3.68%, and the number of model parameters decreased by 16.33% in the near-shore scene ship detection. The proposed algorithm can effectively suppress false alarms on land and reduce the missed detection rate of ships on shore without increasing the number of model parameters.
-
Li Ya, Wang Weigang, Zhang Yuan, Liu Ruipeng
2024,47(6):64-70, DOI:
Abstract:
A task offloading strategy based on Vehicle Edge Computing (VEC) is designed to meet the requirements of complex vehicular tasks in terms of latency, energy consumption, and computational performance, while reducing network resource competition and consumption. The goal is to minimize the long-term cost balancing between task processing latency and energy consumption. The task offloading problem in vehicular networks is modeled as a Markov Decision Process (MDP). An improved algorithm, named LN-TD3, is proposed building upon the traditional Twin Delayed Deep Deterministic Policy Gradient (TD3). This improvement incorporates Long Short-Term Memory (LSTM) networks to approximate the policy and value functions. The system state is normalized to accelerate network convergence and enhance training stability. Simulation results demonstrate that LN-TD3 outperforms both fully local computation and fully offloaded computation by more than two times. In terms of convergence speed, LN-TD3 exhibits approximately a 20% improvement compared to DDPG and TD3.
-
Zhang Bian, Tian Ruyun, Han Weiru, Peng Yuxin
2024,47(6):109-115, DOI:
Abstract:
In order to solve the problems that the traditional SPD life alarm characterization method can not clearly correspond to the real life state of SPD, and the remaining life model characterized by a single degradation related parameter has poor predictability, a multi-parameter SPD life remote monitoring system based on STM32 is designed. With STM32 as the main controller, the important parameters such as surge current, leakage current, surface temperature and tripping status of SPD are collected in real time, and the status information is uploaded to the One net cloud platform through the BC20 wireless communication module. The One net cloud platform displays and stores the multi-parameter data of SPD in real time, and provides data management and analysis. The SVM classification model is used to judge whether SPD is damaged and the BO-LSTM prediction model is used to predict the remaining life of SPD. Based on the positioning function of BC20, the real-time geographic location of SPD can be viewed on the host computer. The results show that the root mean square error and average absolute error of the BO-LSTM prediction model are 0.001 3 and 0.001 8, and the system can monitor the SPD status in real time, effectively predict the remaining life value of SPD, and give early warning in time.
-
Chen Haoan, Li Hui, Huang Rui, Fu Pingbo, Zhang Jian
2024,47(6):182-189, DOI:
Abstract:
Facing the challenges of regulating unmanned aerial vehicles (UAV), and based on an YOLOv5-Lite improved model, this paper incorporates an exponential moving sample weight function that dynamically allocates loss function weights to the model during the training iteration. Through model computations, we achieve real-time UAV tracking using a two-degree-of-freedom servo platform. Furthermore, video capture, model calculations, and servo control are all performed locally on a Raspberry Pi 4B.The optimized model maintains the original model's parameter count while achieving a mAP@.5:.95 score of 70.2%, representing a 1.5% improvement over the baseline model. Real-time inference on the Raspberry Pi yields an average speed of 2.1 frames per second (FPS), demonstrating increased processing efficiency. Simultaneously, the Raspberry Pi controls a servo platform via the I2C protocol to track UAV targets, ensuring real-time dynamic monitoring of UAVs. This optimization enhances system reliability and offers superior practical value.
-
Zhou Jianxin, Zhang Lihong, Sun Tenghao
2024,47(6):79-85, DOI:
Abstract:
Aiming at the problems that the standard honey badger algorithm (HBA) is easy to fall into local optimum, low search accuracy and slow convergence speed, a honey badger algorithm based on elite differential mutation (EDVHBA) is proposed. The elite solution searched by the two optimization strategies in the standard HBA is combined with differential mutation to generate a new elite solution. The use of three elite solutions to guide the next iteration of the population can increase the diversity of the algorithm solution and prevent the algorithm from falling into premature convergence. At the same time, the nonlinear density factor is improved and a new position update strategy is introduced to improve the convergence speed and optimization accuracy of the algorithm. In order to verify the performance of the algorithm, simulation experiments are carried out on eight classical test functions. The results show that compared with other swarm intelligence algorithms and improved HBA, EDVHBA can find the optimal value 0 in the unimodal function, and converge to the ideal optimal value in the multimodal function after about 50 iterations, which verifies that EDVHBA has better optimization performance.
-
Wang Huiquan, Wei Zhipeng, Ma Xin, Xing Haiying
2024,47(6):14-19, DOI:
Abstract:
To solve the problem of low control accuracy of the tidal volume emergency ventilation for lower air pressure at high altitudes, we propose a dual-loop PID tidal volume control system, which utilizes a pressure-compensated PID controller to adjust fan speed, supplemented by an integral-separate PID controller in order to achieve precise control of airflow velocity.Compared with single-loop PID control, the rapid response and no overshooting are observed in the performance tests of the dual-loop control system at an altitude of 4 370 m and atmospheric pressure of 59 kPa, in addition, the output error of the average airflow velocity decrease to 3.19% (the maximum error is 4.1%), which is superior to that of current clinical equipment. Our work offers an effective solution for high-altitude emergency ventilator tidal volume control, and contributes important insights to the development of ventilation control technology in special environments.
-
Fang Xin, Shen Lan, Li Fei, Lyu Fangxing
2024,47(6):20-27, DOI:
Abstract:
The high-frequency measurement data of underground vibration signals can record more specific details about the dynamic response of drilling tools, which is helpful for analyzing and diagnosing abnormal vibrations underground. However, the high-frequency measurement generates a large amount of measurement data, resulting in significant storage pressure for underground vibration measurement equipment. The proposed method uses compressed sensing technology to selectively collect and store sparse underground vibration data and then recover high-frequency measurement results through a signal reconstruction algorithm. In the process of realizing this method, an innovative method of constructing a layered Fourier dictionary against spectrum leakage is proposed, and an improved OMP signal reconstruction algorithm based on layered tracking is researched and realized, which greatly reduces the time required for signal recovery. Simulation and experimental test results demonstrate the method′s effectiveness, achieving a system compression ratio of 18.9 and a reconstruction error of 52.1 dB. The proposed method may greatly reduce the data storage pressure of the measuring equipment in the underground, and provides a new way to obtain high-frequency measurement data of underground vibration.
-
Cheng Dongxu, Wang Ruizhen, Zhou Junyang, Zhang Kai, Zhang Pengfei
2024,47(6):137-142, DOI:
Abstract:
For the tobacco industry, there is currently no detection device and method for detecting the heating temperature and temperature uniformity of heated cigarette smoking sets. In order to solve the temperature measurement needs of micro rod-shaped heating sheets in a narrow space, this article developed a cigarette heating rod thermometer, and designed a new structure suitable for temperature measurement of cigarette heating rods. In order to verify the accuracy and reliability of the measurement results of the cigarette heating rod thermometer, uncertainty analysis of the thermometer was performed. The analysis results are based on the "GB/T 13283-2008 Accuracy Level of Detection Instruments and Display Instruments for Industrial Process Measurement and Control" standard. The measurement range is 100 ℃~400 ℃, meeting the requirements of level 0.1. The final experiment verified that the heating temperature field of different cigarettes can be effectively measured.
-
Peng Duo, Luo Bei, Chen Jiangxu
2024,47(6):50-57, DOI:
Abstract:
Aiming at the non-range-ranging location problem of multi-storey WSN structures, a three-dimensional indoor multi-storey structure location algorithm IAODV-HOP algorithm based on improved Tianying is proposed in the field of large-scale indoor multi-storey structure location for some large commercial supermarkets, hospitals, teaching buildings and so on. Firstly, the nodes are divided into three types of communication radius to refine the number of hops, and the average hop distance of the nodes is modified by using the minimum mean square error and the weight factor. Secondly, the IAO algorithm is used to optimize the coordinates of unknown nodes, and the population is initialized by the best point set strategy, which solves the problem that the quality and diversity of the population are difficult to guarantee due to the random distribution of the initial population in the Tianying algorithm. In addition, the golden sine search strategy is added to the local search to improve the position update mode of the population, and enhance the local search ability of the algorithm. Through simulation experiments, compared with traditional 3D-DV-Hop, PSO-3DDV-Hop, N3-3DDV-Hop and N3-ACO-3DDV-Hop, the normalized average positioning error of the proposed algorithm IAODV-HOP is reduced by 70.33%, 62.67%, 64% and 53.67%, respectively. It has better performance, better stability and higher positioning accuracy.
-
Feng Zhibo, Zhu Yanming, Liu Wenzhong, Zhang Junjie, Li Yingchun
2024,47(6):34-40, DOI:
Abstract:
The data bits and spread spectrum codes of the spaceborne spread-spectrum transponder are asynchronous. Due to the influence of transmission system noise and Doppler frequency shift, it can cause attenuation of peak values related to receiving and transmitting spread spectrum codes, leading to a decrease in capture performance. Traditional capture techniques often have problems such as high algorithm complexity, slow capture speed, and difficulty adapting to the requirements of large frequency offsets of hundreds of kilohertz. This article proposes a spread spectrum sequence search method that truncates the spread spectrum sequence into two segments for correlation operations, and combines the signal squared sum FFT loop for a large frequency offset locking, effectively suppressing the attenuation of correlation peaks and improving pseudocode capture performance. MATLAB simulation and FPGA board level testing show that the proposed spread spectrum signal capture scheme can resist Doppler frequency shifts of up to ±300 kHz, with an average capture time of about 95 ms. In addition, the FPGA implementation of this algorithm saves about 47% of LUT, 43% of Register, and more than half of DSP and BRAM resources compared to traditional structures, making it of great application value in resource limited real-time communication systems.
-
Ma Zhewei, Zhou Fuqiang, Wang Shaohong
2024,47(6):94-99, DOI:
Abstract:
A feature point extraction algorithm based on adaptive threshold and an improved quadtree homogenization strategy are proposed to address the issue of low positioning accuracy or low matching logarithms of the SLAM system caused by the ORB-SLAM2 algorithm extracting fewer feature points in dark environments or environments with fewer textures, resulting in system crashes. Firstly, based on the brightness of the image, FAST (Features from Accelerated Seed Test) feature points are extracted using adaptive thresholds. Then, an improved quadtree homogenization strategy is used to eliminate and compensate the feature points of the image, completing feature point selection. The experimental results show that the improved feature point extraction algorithm increases the number of matching pairs by 17.6% and SLAM trajectory accuracy by 49.8% compared to the original algorithm in dark and textured environments, effectively improving the robustness and accuracy of the SLAM system.
-
2024,47(6):28-33, DOI:
Abstract:
With the increasing demand for satellite network, vehicle-connected network, industrial network and other service simulation, this paper proposes a multi-session delay damage simulation method based on delay range strategy to build flexible software network damage simulation, aiming at the problems of small number of analog links, low flexibility and high resource occupation of traditional dedicated channel damage instruments. In this method, the delay damage of each session flow is identified and controlled independently, and the multi-queue merging architecture based on time delay strategy is adopted to reduce the resource consumption. The experimental results show that compared with the traditional dedicated device and simulation software NetEm, the proposed method supports the independent delay configuration of million-level links, increases the number of session streams from ten to one million, and reduces the memory consumption by at least 85% under each bandwidth, which meets the requirements of large scale and accuracy, and greatly reduces the system cost.
-
Ma Dongyin, Wang Xinping, Li Weidong
2024,47(6):58-63, DOI:
Abstract:
Aiming at the Automatic Train Operation of high-speed train,an algorithm based on BAS-PSO optimized auto disturbance rejection control (ADRC) is used to design speed tracking controller.The ADRC is designed based on the train dynamics model,ITAE is used as the objective function,and the parameters are tuned by BAS-PSO.CRH380A train parameters are selected, The tracking effect of BAS-PSO, PSO and improved shark optimized ADRC algorithm on the target speed curve of the train is compared by MATLAB simulation,The tracking error of the train target speed curve based on the BAS-PSO optimized ADRC algorithm is kept in the range of ±0.4 km/h,which is closer to the target speed curve than the other two algorithms.The results show that the ADRC based on BAS-PSO optimization has the advantages of small tracking error and strong anti-interference ability.
-
Long Biao, Yang Jun, Chen Huiping, Chen Guangrun, Zhao Peiyang
2024,47(6):157-163, DOI:
Abstract:
In order to solve the problem that the audio signal processing in the voice communication system has a large amount of data, a lot of stray signals, and the received audio signals of the frequency modulation receiver are large and small, a lightweight audio signal processing algorithm is proposed, and based on this algorithm, the audio signal receiving and automatic gain control are realized on the field programmable gate array(FPGA) platform. The algorithm combines digital down conversion technology, multistage extraction filtering technology and automatic gain control technology (AGC) technology, and is applied to the audio signal processing system. The RF analog signal received from the upper antenna is converted into baseband audio signal through analog-to-digital conversion and digital down-conversion, and the stray signal in the baseband signal is filtered through four-stage extraction filtering, reducing the complexity and power consumption of the system. At the same time, the digital AGC controls and adjusts the baseband audio signal to output a more stable audio signal. The experimental results show that the algorithm can effectively reduce the information rate from 102.4 MHz to 32 kHz, reduce the computation burden, improve the signal quality, and reduce the resource utilization of FPGA. And the automatic gain control adjustment of audio signal is realized, and the adjustment time is only 12.8 μs, which meets the power stability time of the receiver.
-
Shi Shujie, Zhao Fengqiang, Wang Bo, Yang Chenhao, Zhou Shuai
2024,47(6):116-122, DOI:
Abstract:
Rolling bearings play an important role in rotating machinery. If a fault occurs, it can cause equipment shutdown, and in severe cases, endanger the safety of on-site personnel. Therefore, it is necessary to diagnose the fault. In response to the difficulty in extracting fault features of rolling bearings and the low accuracy of traditional classification methods, this paper proposes a fault diagnosis method based on Set Empirical Mode Decomposition (EEMD) energy entropy and Golden Jackal Optimization Algorithm (GJO) optimized Kernel Extreme Learning Machine (KELM), achieving the goal of extracting fault features of rolling bearings and correctly classifying them. Through experimental data validation, this method can extract the fault information features hidden in the original signal of rolling bearings, with a diagnostic accuracy of up to 98.47%.
-
Zhou Guoliang, Zhang Daohui, Guo Xiaoping
2024,47(6):190-196, DOI:
Abstract:
The gesture recognition method based on surface electromyography and pattern recognition has a broad application prospect in the field of rehabilitation hand. In this paper, a hand gesture recognition method based on surface electromyography (sEMG) is proposed to predict 52 hand movements. In order to solve the problem that surface EMG signals are easily disturbed and improve the classification effect of surface EMG signals, TiCNN-DRSN network is proposed, whose main function is to better identify the noise and reduce the time for filtering the noise. Ti is a TiCNN network, in which convolutional kernel Dropout and minimal batch training are used to introduce training interference to the convolutional neural network and increase the generalization of the model; DRSN is a deep residual shrinkage network, which can effectively eliminate redundant signals in sEMG signals and reduce signal noise interference. TiCNN-DRSN has achieved high anti-noise and adaptive performance without any noise reduction pretreatment. The recognition rate of this model on Ninapro database reaches 97.43% 0.8%.
