Table of Content

15 October 2023 Volume 36 Issue 10
    Chinese License Plate Detection and Recognition in Unconstrained Scenarios Based on YOLO
    CHEN Ziang,LIU Na,YUAN Ye,LI Qingdu,WAN Lihong
    Electronic Science and Technology. 2023, 36(10):  1-8.  doi:10.16180/j.cnki.issn1007-7820.2023.10.001
    Abstract ( 273 )   HTML ( 24 )   PDF (2014KB) ( 127 )  
    Figures and Tables | References | Related Articles | Metrics

    In view of the problems of traditional Chinese license plate recognition methods, such as the requirement of scenes, poor real-time performance, and inability to deploy on edge devices, this study proposes a Chinese license plate detection and recognition method based on YOLO(You Only Look Once) in unconstrained scenes. This method is divided into two modules: license plate detection and license plate character recognition. In the license plate detection part, the improved YOLOv5 model is used to predict four groups of key points for license plate correction based on the prediction of target candidate regions, and the pre-training model trained on the COCO data set is used for training, which reduces the error detection problem caused by the complex environment and has high real-time performance. In the license plate character recognition part, the CRNN(Convolutional Recurrent Neural Network) model is improved, which greatly reduces the parameters and computation of the algorithm, so that it can be successfully deployed in various edge devices. Experimental results show that the proposed method can efficiently detect and recognize license plates in complex environments. The map value of the proposed license plate detection model is 3.0% higher than that of Retina-face in the license plate detection data set. Compared with LPR-Net, the accuracy of license plate character recognition model in license plate recognition data set is improved by 4.2%.

    Underwater Target Direction Finding Method Based on Vector Hydrophone Single Channel Instantaneous Phase Difference Weighting
    BAI Xingyu,LIU Mingyu,JIANG Yu,WANG Yiqi,LI Shuai
    Electronic Science and Technology. 2023, 36(10):  9-14.  doi:10.16180/j.cnki.issn1007-7820.2023.10.002
    Abstract ( 58 )   HTML ( 3 )   PDF (1220KB) ( 34 )  
    Figures and Tables | References | Related Articles | Metrics

    In order to solve the problem of poor target direction finding accuracy of conventional complex acoustic intensity method under the condition of low SNR(Signal-to-Noise Ratio), an underwater target direction finding method based on vector hydrophone single channel instantaneous phase difference weighting is proposed in this study. According to the characteristics that the instantaneous phase difference corresponding to the line spectrum frequency unit of underwater target is relatively stable and the instantaneous phase difference corresponding to the background noise frequency unit changes randomly, the variance weighting of the instantaneous phase difference of each channel frequency unit of vector hydrophone is carried out to enhance the SNR gain of the line spectrum and effectively restrain the energy interference of the background noise, thus realizing the high precision direction finding of underwater target. Simulation analysis and experimental verifications show that the direction finding accuracy of the proposed method is 15.8% higher than that of the conventional complex acoustic intensity method under the condition of low SNR at -20~10 dB.

    A 3D Object Detection Network Based on Attention Mechanism and Context Awareness
    ZHANG Wuran,LI Feifei
    Electronic Science and Technology. 2023, 36(10):  15-23.  doi:10.16180/j.cnki.issn1007-7820.2023.10.003
    Abstract ( 68 )   HTML ( 4 )   PDF (4391KB) ( 43 )  
    Figures and Tables | References | Related Articles | Metrics

    As research in the field of autonomous driving has attracted much attention, driving safety has become the primary consideration. Because the point cloud scene is cluttered and the background environment interferes greatly, and with the expansion of the acquisition range, the point cloud becomes more sparse, which makes the robustness of the detection algorithm weaker. To alleviate the above problems, this study proposes a 3D object detection network based on attention mechanism and context awareness. In the point cloud processing stage, a double attention mechanism based point cloud is added to generate a point weight matrix, display and mark important point data, and suppress background noise interference. In the pseudo-map feature extraction module, the FPN(Feature Pyramid Network) module is added to reuse multi-scale features, and a Context Awareness Module(CAM) is designed to capture multi-scale context semantics. Furthermore, an Attention Guide Module(AGM) is proposed based on the source features to generate a guidance weight map with clear spatial positions, so as to alleviate the spatial ambiguity caused by redundant features. The experiments in this study are carried out on the KITTI data set test. Compared with the baseline network, the Average Precision(AP) of the proposed method for pedestrians, cars and cyclists is improved by 0.59%, 0.87% and 1.42% respectively under the difficulty index. Compared with the new baseline network, the AP of the proposed method for pedestrians is improved by 3.04%, 3.53% and 3.23% under the three difficulty levels, respectively. The results show that the proposed algorithm can effectively improve the performance of 3D object detection.

    Research on EMI Synergy Analysis of New Energy Vehicle Motor Drive System
    XU Zhijiang,TAN Kangbo,LI Min,ZHAO Ziwen,CHEN Tongshan
    Electronic Science and Technology. 2023, 36(10):  24-31.  doi:10.16180/j.cnki.issn1007-7820.2023.10.004
    Abstract ( 84 )   HTML ( 4 )   PDF (3742KB) ( 32 )  
    Figures and Tables | References | Related Articles | Metrics

    In view of the electromagnetic interference problem of the motor drive system inside the electric vehicle, a prediction model of the electromagnetic interference of the vehicle motor drive system is established in this study. Based on the principle of electromagnetic compatibility, the electromagnetic interference mechanism of the in-vehicle motor drive system module is analyzed. The model of the IGBT power semiconductor device that causes interference in the module is constructed, and the influence of the parasitic parameters of the actual device is considered in the modeling process. Through the Simulink+Maxwell+Simplorer collaborative analysis technology, the electromagnetic interference signal acquisition of the motor drive system is realized. The comparison between the simulation and the measured data verifies the effectiveness of the method proposed in this study, and the errors of both are less than 10 dB.Based on the GB/T 18655-2018 electromagnetic compatibility standard, the results of the parts exceeding the standard are optimized and rectified.

    Line Segment Matching Based on RFNA and Improved LBD of Mirror Image
    GAO Yuke,ZHANG Wei,HU Zhi,JIANG Pengwei
    Electronic Science and Technology. 2023, 36(10):  32-38.  doi:10.16180/j.cnki.issn1007-7820.2023.10.005
    Abstract ( 42 )   HTML ( 1 )   PDF (2511KB) ( 42 )  
    Figures and Tables | References | Related Articles | Metrics

    In view of the matching problem between objects and mirrors in images, the RNFA(Relative Number of False Alarms)edge chain detection method is introduced to obtain richer line segments. An improved LBD(Line Band Descriptor) algorithm is proposed for constructing local invariant feature descriptors, and initial matching pairs are obtained by comparing local invariant feature descriptors. The screening of the global projection angle is adopted and the false matches in the initial matching pairs are eliminated fitting the projection center line. After the selection of the global projection angle and the fitting of the projection median are completed, the screening of the local invariant feature descriptor threshold is relaxed to obtain more matching pairs and improve the recall rate. The experimental results of image set simulation show that the designed algorithm can better identify line segments in the weaker texture regions and obtain more matching pairs on the basis of the guaranteed performance of the original algorithm, which can improve the correct matching rate by about 5% and achieve a recall rate of over 90%.

    A Survey of Text-to-Image Synthesis Based on Generative Adversarial Network
    LI Yueyang,TONG Guoxiang,ZHAO Yingzhi,LUO Qi
    Electronic Science and Technology. 2023, 36(10):  39-55.  doi:10.16180/j.cnki.issn1007-7820.2023.10.006
    Abstract ( 283 )   HTML ( 8 )   PDF (5679KB) ( 62 )  
    Figures and Tables | References | Related Articles | Metrics

    The text-to-image synthesis refers to translating the text description in sentence form into an image with similar semantics to the text. In the early research, the task of image generation is mainly based on keyword or sentence retrieval to align the visual content matched with the text. With the generative adversarial network, the method of text-to-image synthesis has made great progress in visual realism, diversity and semantic similarity. The generative adversarial network generates reasonable and real images through the confrontation between generator and discriminator, and shows strong ability in the fields of image restoration and super-resolution generation. Based on the review and summary of the latest research results in the field of text-to-image synthesis, a new classification method is proposed: Attention enhancement, multi-stage enhancement, scene layout enhancement and universality enhancement. The challenges and future development direction of text-to-image synthesis are also discussed in this study.

    Construction and Exploration of Experimental Teaching Platform of Ultrafast Laser Manufacturing
    YUN Ling,CUI Yudong
    Electronic Science and Technology. 2023, 36(10):  56-61.  doi:10.16180/j.cnki.issn1007-7820.2023.10.007
    Abstract ( 50 )   HTML ( 5 )   PDF (2599KB) ( 26 )  
    Figures and Tables | References | Related Articles | Metrics

    Strengthening practical training and cultivating innovative awareness are indispensable components of higher engineering education, and experimental teaching is needed in the process of cultivating innovative thinking awareness. Considering the high requirements of both theoretical and practical aspects in the major of optoelectronic information science and engineering, as well as the fast updating speed of theoretical knowledge, an experimental teaching platform for ultrafast fiber laser manufacturing technology is designed in this study. The influence of saturable absorbers based on lead sulfide quantum dots on the core performance parameters of ultrafast fiber lasers is deeply studied. This comprehensive experiment covers various aspects such as material preparation and characterization, development of saturable absorbers, construction and testing of ultrafast laser systems, etc. The experimental content is progressive and comprehensive, which enables learners to flexibly grasp various knowledge points and enhance their engineering practice and innovation abilities.

    Security Design of eFlash Controller Based on AES Algorithm
    LIN Yuhong,XIAO Hao
    Electronic Science and Technology. 2023, 36(10):  62-67.  doi:10.16180/j.cnki.issn1007-7820.2023.10.008
    Abstract ( 53 )   HTML ( 6 )   PDF (1139KB) ( 39 )  
    Figures and Tables | References | Related Articles | Metrics

    For the security requirements of ASIC(Application Specific Integrated Circuit) chip data storage, especially eFlash(embed Flash) stores security risks for sensitive data, an eFlash secure storage controller based on AES(Advanced Encryption Standard) algorithm is designed. Compared with traditional encryption designs based on software and hardware platforms, ASIC chip platform has the advantages of high integration and fast operation. By analyzing the principle of AES algorithm, it is proposed to use AES algorithm to encrypt the stored data on eFlash controller. The data transmission rate is a key factor in eflash performance, AES algorithm is implemented by pipeline structure to improve data throughput, the throughput rate can reach 1.4 Gbit·s-1with consumption of 9.96×10-10 m2 logical resources. This encryption scheme enhances eFlash storage security while consuming less logical resources and encryption delay, effectively prevents external attacks on the storage information of ASIC chips.

    Security Detection of Extended Kalman Filter under Injection Attack
    LI Xiuwen,WANG Lei,REN Zhu
    Electronic Science and Technology. 2023, 36(10):  68-73.  doi:10.16180/j.cnki.issn1007-7820.2023.10.009
    Abstract ( 32 )   HTML ( 2 )   PDF (767KB) ( 28 )  
    Figures and Tables | References | Related Articles | Metrics

    Cyber-physical system is an open networked intelligent information system, which is vulnerable to security attack. Injection attack is a kind of communication attack, and injection of false data makes the sensor get wrong measurement value, which leads to system performance deterioration. To solve this problem, this study combines the extended Kalman state estimator and uses the detection scheme of extended Kalman filter state estimation based on the least trace principle to detect the new information sequence in the same scene and judge the system state through hypothesis testing. The simulation experiments on MATLAB show that under the given attack detection rules, the detection scheme can effectively detect the injection attack and reduce the performance loss caused by the wrong measurement value of the sensor.

    Combined Heat and Power Low-Carbon Economic Scheduling Considering Carbon Trading and Price Demand Response
    WANG Yumei,ZHANG Jiqin,ZHOU Yongxin
    Electronic Science and Technology. 2023, 36(10):  74-81.  doi:10.16180/j.cnki.issn1007-7820.2023.10.010
    Abstract ( 20 )   HTML ( 1 )   PDF (2510KB) ( 26 )  
    Figures and Tables | References | Related Articles | Metrics

    In view of the problems of thermoelectric conflict and abandoned wind power during winter heating, a low-carbon economic dispatch model for combined heat and power that takes into account carbon trading costs and price-based demand response is proposed, while electric boilers and heat storage devices are introduced to decouple the coupling limits of thermoelectricity to study the impact of different operation modes on optimal dispatch. Carbon trading costs are considered in the system to limit carbon emissions, and the user-side demand response is used to enhance the peak regulation capability of the grid and optimize the load curve, effectively cutting the peak-to-valley difference in load and providing space for wind power to go online. Finally, the CPLEX solver is used for solving and simulation analysis is carried out based on the improved IEEE-30 node system. The results show that the wind power consumption is increased by 83.72% and carbon emissions are reduced by 2 211 t under the scheduling method proposed in this study, which verifies the reliability of the scheduling model.

    Simulation Analysis of the Shortest Path TAODV Routing Protocol Based on Trust Mechanism
    ZHU Ying,ZHU Jinrong,SHI Zhuangzhuang,XU Siyun,XIA Changquan
    Electronic Science and Technology. 2023, 36(10):  82-86.  doi:10.16180/j.cnki.issn1007-7820.2023.10.011
    Abstract ( 24 )   HTML ( 1 )   PDF (698KB) ( 27 )  
    Figures and Tables | References | Related Articles | Metrics

    In order to compensate for the shortcomings in the security of the AODV(Ad Hoc on Demand Distance Vector) routing protocol while obtaining the shortest paths based on this optimized protocol, a study of a TAODV(Trusted Ad Hoc on Demand Distance Vector Routing Algorithm)routing protocol that improves on the original AODV routing protocol is proposed using the trust mechanism model approach. This protocol performs the relevant operations based on the trust values of each node to determine the path trust values during the operation of the routing protocol. The simulation results show that the improved TAODV routing protocol outperforms the traditional AODV routing protocol in terms of normalized routing overhead, minimum hop count and optimal path, and enhances the robustness and destructiveness of the network.

    Circle Fitting Algorithm Based on Multilevel Optimization
    XU Yongliang,XIE Xiaohui
    Electronic Science and Technology. 2023, 36(10):  87-94.  doi:10.16180/j.cnki.issn1007-7820.2023.10.012
    Abstract ( 76 )   HTML ( 4 )   PDF (1701KB) ( 44 )  
    Figures and Tables | References | Related Articles | Metrics

    In view of the problem that it is difficult to ensure high efficiency and high accuracy in circle fitting algorithm in the same time, a circle fitting algorithm based on multi-level optimization is proposed in this study. The 3σ criterion is used to remove the coarse error points, and the random sampling consistency is improved by reducing the randomness of subset selection, the descending operation of the circle model and the threshold transformation of the number of adaptive iterations, so as to extract the high quality internal group points. The iterative weighted least square method with iteration termination condition is applied to achieve the great processing of point groups. In this study, the effectiveness of the algorithm is verified from the three aspects of defect circle, impurity interference and other noise, and the proposed method is compared with other mainstream circle fitting algorithms. The results show that the fitting accuracy of the proposed algorithm is less than 0.7 pixels under different degrees of circular defect and impurity interference, and the algorithm performs better than other algorithms in fitting effect. In the case of the noise interference of about 20%~265%, the fitting accuracy of the algorithm does not exceed 1 pixel, and the running time is less than 0.7 s. These results indicate that the proposed algorithm can resist a large number of salt-and-pepper noise interference and maintain high accuracy and detection efficiency.

    Deep Completion Based on Multi-Source Data Association Fusion
    WANG Ge,YANG Ruihua,XI Wei,ZHAO Jizhong
    Electronic Science and Technology. 2023, 36(10):  95-102.  doi:10.16180/j.cnki.issn1007-7820.2023.10.013
    Abstract ( 79 )   HTML ( 5 )   PDF (2877KB) ( 40 )  
    Figures and Tables | References | Related Articles | Metrics

    With the acceleration of urbanization, the intelligent transportation has received more and more attention. Among them, the use of depth completion technology to extract the depth information of objects plays an important role in the realization of vehicle target tracking, distance calculation between targets and other tasks. However, multi-source depth completion data collected in practice often have correlation bias, resulting in knotty errors. In this regard, this study studies the depth completion technology based on multi-source data association fusion. The proposed technology enhances the depth map by calculating multi-channel confidence, and performs more accurate point-by-point correlation between the image and the millimeter-wave radar point cloud data. By designing a multi-scale attention fusion module, the adaptive fusion of multi-granularity associated data is realized to generate high-quality depth maps. In this study, a large number of experiments have been carried out in the public nuScenes data set. The experimental results show that the mean absolute error of our method is 1.142 m, which is lower than the 1.472 m of the existing benchmark method.


Monthly,Founded in September 1987
Competent Authorities:
Ministry of Education of the People's Republic of China
Sponsored by:Xidian University
Chief Editor:Liao Guisheng
Executive Editor:Wan Liancheng
Editor:Hei Lei
Editor and Publisher:
The Editorial Department of Electronic Science and Technology
Distribution Abroad:
China Intermational Book Trading Corporation
P.O.BOx 399,Beijing 100044,China
P.O.Box 375,2 Taibai Road(South),Xi'an 710071,China
Unit Price:$20.00