Content of Research Hotspots and Reviews in our journal

        Published in last 1 year |  In last 2 years |  In last 3 years |  All
    Please wait a minute...
    For Selected: Toggle Thumbnails
    Review of Object Detection Algorithm Improvement in Deep Learning
    YANG Feng, DING Zhitong, XING Mengmeng, DING Bo
    Computer Engineering and Applications    2023, 59 (11): 1-15.   DOI: 10.3778/j.issn.1002-8331.2209-0312
    Abstract21)      PDF(pc) (691KB)(17)       Save
    Object detection is currently a research hotspot in the field of computer vision. With the development of deep learning, object detection algorithms based on deep learning are increasingly applied and their performance is constantly improved. This paper summarizes the latest research progress of object detection methods based on deep learning by summarizing common problems encountered in the process of object detection and corresponding improvement methods. This paper focuses on two types of object detection algorithms based on deep learning. In addition, the latest improvement ideas of target detection algorithms are summarized from the aspects of attention mechanism, lightweight network, multi-scale detection. Finally, in view of the current problems in the field of target detection, the future development trend is prospected. And the feasible solution is put forward in order to provide reference ideas and directions for the follow-up research work in this field.
    Reference | Related Articles | Metrics
    Review of Small Object Detection Algorithms Based on Deep Learning
    DONG Gang, XIE Weicheng, HUANG Xiaolong, QIAO Yitian, MAO Qian
    Computer Engineering and Applications    2023, 59 (11): 16-27.   DOI: 10.3778/j.issn.1002-8331.2211-0377
    Abstract21)      PDF(pc) (646KB)(16)       Save
    The existing object detection algorithms have high accuracy for the detection of large objects and medium objects, but due to the few pixels in the image and the available features of small objects, the detection accuracy of small objects is too low compared with that of large objects. By fusing the feature layer, the detection of small objects has achieved good results, but there are still problems such as the localization of small objects. Based on this, the definition of small objects is first explained, and five reasons for the low detection accuracy of small objects are pointed out. Subsequently, the latest progress in recent years and the classic small object detection optimization method in the past are described from multi-scale features, novel metric, and super-resolution according to the general principle. Secondly, the detection methods of small objects for specific scenes:aerial images, faces, and pedestrians are summarized. Finally, the possible research directions of small object detection in the future are summarized and proposed.
    Reference | Related Articles | Metrics
    Review of WebAssembly Application Research for Edge Serverless Computing
    WANG Xin, ZHAO Kai, QIN Bin
    Computer Engineering and Applications    2023, 59 (11): 28-36.   DOI: 10.3778/j.issn.1002-8331.2210-0308
    Abstract19)      PDF(pc) (575KB)(4)       Save
    WebAssembly(Wasm) is a new binary format that is portable, small, fast to load and compatible with the Web. It has the characteristics of high efficiency, security, and openness. The basic concept of edge computing is to run computing tasks on computing resources close to the data source. However, the performance and resources of devices deployed on the edge are usually very limited. In this resource-constrained environment, how to provide low-latency and secure services is an important research direction of edge computing. Serverless is a new way to host applications on infrastructure. At present, it is mainly based on container technology to realize program hosting. Serverless computing is currently the most suitable architecture for edge computing due to its lightweight, function as a service(FaaS), automatic scaling and so on, but it always has problems such as cold start and large memory consumption. Wasm can replace the traditional container and provide an updated, faster, less resource consuming and secure isolation implementation for edge serverless computing. In this paper, the characteristics of edge serverless computing and its application scenarios, as well as the development trend of Wasm are introduced at first. The current research on Wasm-based edge serverless computing is analyzed and the development direction of Wasm runtime alternative containers as the carrier of edge serverless computing is illustrated. Moreover, the problems of edge Wasm serverless computing platform are discussed and the future optimization directions based on deep reinforcement learning and other artificial intelligence algorithms are summarized.
    Reference | Related Articles | Metrics
    Application and Progress of Deep Learning in 2D Virtual Try-on Technology
    HUA Ailing, YU Feng, CHEN Ziyi, WANG Hua, JIANG Minghua
    Computer Engineering and Applications    2023, 59 (11): 37-45.   DOI: 10.3778/j.issn.1002-8331.2209-0352
    Abstract16)      PDF(pc) (597KB)(7)       Save
    Virtual try-on technology has wide application and research value in promoting the informatization and intelligence of the clothing industry, and is one of the research hotspots of artificial intelligence in the field of intelligent clothing manufacturing. At present, virtual try-on mainly focuses on two-dimensional virtual try-on based on image generation. This paper provides a comprehensive overview of two-dimensional virtual try-on technology. It introduces and analyzes traditional virtual try-on, and categorizes and summarizes the main tasks, types, development process, and models of existing two-dimensional virtual try-on technology. It also discusses in detail the principles and related improvements of representative algorithms of various types. It summarizes the application of traditional virtual try-on and two-dimensional virtual try-on technology, and discusses the extension technology of two-dimensional virtual try-on technology. It summarizes and compares the benefits and drawbacks of using traditional and current virtual try-on technology, as well as summarizes and forecasts the field’s future development.
    Reference | Related Articles | Metrics
    Survey of Adaptive MAC Protocols in Mobile Ad Hoc Networks
    YAN Tao, ZHAO Yifan, GAO Minghu, YANG Lujin, ZHOU Sida
    Computer Engineering and Applications    2023, 59 (11): 46-56.   DOI: 10.3778/j.issn.1002-8331.2211-0090
    Abstract15)      PDF(pc) (576KB)(8)       Save
    Mobile Ad Hoc Network(MANET) has the characteristics of frequent node movement and dynamic topology, and the media access control(MAC) protocol, as the basic communication protocol followed by nodes to access wireless channel resources, is critical to network performance. With the diversification of application tasks and the increasing demand for quality of service(QoS), it is critical to design an adaptive MAC protocol that can maintain good performance in a dynamic environment such as MANET. Firstly, the network characteristics and basic MAC protocol classification of MANET are summarized, and the challenges faced in MAC protocol design are analyzed. Then, starting from the improvement basis of different MAC protocols, the adaptive MAC protocols applied to complex network environments in recent years are expounded, and the characteristics, performance and limitations of each protocol are summarized and compared. Finally, the research progress of adaptive MAC protocol is summarized, and the next research direction is prospected.
    Reference | Related Articles | Metrics
    Construction and Application of Discipline Knowledge Graph in Personalized Learning
    ZHAO Yubo, ZHANG Liping, YAN Sheng, HOU Min, GAO Mao
    Computer Engineering and Applications    2023, 59 (10): 1-21.   DOI: 10.3778/j.issn.1002-8331.2209-0345
    Abstract133)      PDF(pc) (929KB)(143)       Save
    The discipline knowledge graph is an important tool to support teaching activities based on big data, artificial intelligence and other technologies. As a kind of discipline knowledge semantic network, it contributes to the development of personalized learning systems and the promotion of new infrastructure for digital education resources. Firstly, this paper outlines the concept and classification of knowledge graph. Secondly, this paper summarizes the concept, characteristics, advantages, connotation and the support for personalized learning of discipline knowledge graph. Nextly, this paper focuses on the sorting of construction process of discipline knowledge graph:discipline ontology construction, discipline knowledge extraction, discipline knowledge fusion and discipline knowledge processing, and it also introduces the application of discipline knowledge graph in personalized learning situations and the challenges. Finally, this paper prospects the future tendency of discipline knowledge graph and personalized learning. It provides the reference and inspiration for the organization of educational resources and the innovative development of personalized learning.
    Reference | Related Articles | Metrics
    Research of Deep Learning-Based Computational Spectral Imaging for Single RGB Image
    JIANG Zhongmin, ZHANG Wanyan, WANG Wenju
    Computer Engineering and Applications    2023, 59 (10): 22-34.   DOI: 10.3778/j.issn.1002-8331.2211-0082
    Abstract66)      PDF(pc) (662KB)(64)       Save
    Deep learning is introduced into computational spectral imaging to address the high cost and long image acquisition time of traditional spectral imaging methods to investigate how spectral information can be recovered from a single RGB image to provide assistance for various computer vision applications. Computational spectral imaging methods for single RGB images based on deep learning lack comprehensive and systematic research. Deep learning algorithms and network models used for computational spectral imaging are summarised, analyzed and compared. The four categories of CNN(convolutional neural networks), GAN(generative adversarial networks), Attention and Transformer are used to sort out supervised learning methods with excellent reconstruction performance in recent years. The unsupervised learning methods are discussed in terms of both self-encoders and domain adaptation. Datasets and evaluation metrics commonly used for the algorithms are listed, and future research trends and development directions are given.
    Reference | Related Articles | Metrics
    Application Progress of Deep Learning in Epilepsy Detection
    ZHANG Hanming, MA Jingang, ZHANG Ningning, ZHAO Zhenzhen, LI Ming
    Computer Engineering and Applications    2023, 59 (10): 35-47.   DOI: 10.3778/j.issn.1002-8331.2207-0062
    Abstract62)      PDF(pc) (623KB)(36)       Save
    With the increasing number of epileptic patients year by year, it is of great practical significance to detect epileptic diseases timely and accurately. Nowadays, deep learning develops rapidly and is widely used in the medical field. Epilepsy detection task based on deep learning has also become a research hotspot. After combing the relevant literature in recent years, this paper systematically summarizes the application of deep learning algorithm in epilepsy detection. Firstly, the pathogenesis, etiology and treatment of epilepsy are introduced. Secondly, the EEG used in epilepsy detection and the division of the overall process of epilepsy are explained. Then, the differences between traditional machine learning and deep learning in this field are simply compared. This paper focuses on the research progress of detecting EEG signals in various stages of epilepsy by deep learning, including two-stage, three-stage and multi-stage EEG detection, and compares the detection algorithms in various stages of epilepsy. Finally, this paper summarizes and prospects the research status and future development direction in this field.
    Reference | Related Articles | Metrics
    Survey of Transformer-Based Object Detection Algorithms
    LI Jian, DU Jianqiang, ZHU Yanchen, GUO Yongkun
    Computer Engineering and Applications    2023, 59 (10): 48-64.   DOI: 10.3778/j.issn.1002-8331.2211-0133
    Abstract104)      PDF(pc) (875KB)(80)       Save
    Transformer is a kind of deep learning framework with strong modeling and parallel computing capabilities. At present, object detection algorithm based on Transformer has become a hotspot. In order to further explore new ideas and directions, this paper summarizes the existing object detection algorithm based on Transformer as well as a variety of object detection data sets and their application scenarios. This paper describes the correlation algorithms for Transformer based object detection from four aspects, i.e. feature extraction, object estimation, label matching policy and application of algorithm, compares the Transformer algorithm with the object detection algorithm based on convolutional neural network, analyzes the advantages and disadvantages of Transformer in object detection task, and proposes a general framework for Transformer based object detection model. Finally, the prospect of development trend of Transformer in the field of object detection is put forward.
    Reference | Related Articles | Metrics
    Review and Prospect of Underwater Environmental Monitoring System
    ZHU Yipu, DU Xiujuan
    Computer Engineering and Applications    2023, 59 (10): 65-74.   DOI: 10.3778/j.issn.1002-8331.2210-0414
    Abstract40)      PDF(pc) (605KB)(22)       Save
    Abundant underwater resources are closely related to human production and life, and the real-time monitoring of underwater environment is an important means to protect and develop underwater resources. The underwater environment monitoring system mainly realizes the acquisition and transmission of underwater environment data, as well as the storage, processing and analysis management of the data. According to the technical characteristics, the development history and application of the underwater environment monitoring system at home and abroad are described. The key technologies and their advantages and disadvantages of the underwater environment monitoring system are summarized, the system architecture is classified and analyzed, and the environment monitoring system based on the underwater wireless sensor networks and the functions of each module are introduced emphatically. Finally, the future development of underwater environmental monitoring is proposed.
    Reference | Related Articles | Metrics
    Survey of Few-Shot Relation Classification
    LIU Tao, KE Zunwang, Wushour·Silamu
    Computer Engineering and Applications    2023, 59 (9): 1-12.   DOI: 10.3778/j.issn.1002-8331.2208-0027
    Abstract122)      PDF(pc) (687KB)(120)       Save
    Few-shot relation classification aims to mine the semantic relationship between target entities in natural language texts with limited labeled training examples, so as to deal with the resource shortage problem faced by the traditional relation classification methods, so that it can be better applied to medicine, finance and ethnic language processing and other data scarce fields. At present, the relevant research work on few-shot relation classification all learns prior knowledge under the training strategy of meta learning, and to quickly adapt to new tasks. Generally, it can be divided into four classes method:prototype network based, pre-training language model based, parameter optimization based, and graph neural network based. This paper reviews the development of few-shot relation classification, analyzes and summarizes the advantages and limitations of different research methods. On this basis, the paper analyzes the current problems and challenge faced by few-shot relation classification, and prospects the future research directions of this field.
    Reference | Related Articles | Metrics
    Research Progress on Vision System and Manipulator of Fruit Picking Robot
    GOU Yuanmin, YAN Jianwei, ZHANG Fugui, SUN Chengyu, XU Yong
    Computer Engineering and Applications    2023, 59 (9): 13-26.   DOI: 10.3778/j.issn.1002-8331.2209-0183
    Abstract125)      PDF(pc) (787KB)(92)       Save
    Fruit picking robot is of great significance to the realization of automatic intelligence of fruit equipment. In this paper, the research work on the key technologies of fruit-picking robot at home and abroad in recent years is summarized, firstly, the key technologies of fruit-picking robot vision system, such as traditional image segmentation methods based on fruit features, such as threshold method, edge detection method, clustering algorithm based on color features and region-based image segmentation algorithm, are discussed, the object recognition algorithm based on depth learning and the target fruit location are analyzed and compared, and the state-of-the-art of fruit picking robot manipulator and end-effector is summarized, finally, the development trend and direction of fruit-picking robot in the future are prospected, which can provide reference for the related research of fruit-picking robot.
    Reference | Related Articles | Metrics
    Review of Application of Deep Learning in Symbolic Music Generation
    CHEN Jishang, Abudukelimu Halidanmu, LIANG Yunze, Abulizi Abudukelimu, Aishan Mikelayi, GUO Wenqiang
    Computer Engineering and Applications    2023, 59 (9): 27-45.   DOI: 10.3778/j.issn.1002-8331.2209-0305
    Abstract94)      PDF(pc) (938KB)(58)       Save
    Symbolic music generation is an important task in the field of music information retrieval. This paper provides a comprehensive summary of deep learning-based symbolic music generation, and existing methods are classified, analyzed, as well as compared. The research status and tasks of symbolic music generation are introduced in detail. It expounds the representation and coding methods of symbolic music, and focuses on the induction, comparison and analysis of deep learning-based models, which are divided into three categories according to different basic structures. It expounds and summarizes the evaluation criteria and datasets in the field of symbolic music generation, and evaluates the performance of representative models. The existing problems in this field are pointed out and corresponding prospects are put forward.
    Reference | Related Articles | Metrics
    Application and Prospect of Python Language in Field of Hydrology and Water Resources
    JIANG Qiuxiang, GUO Weipeng, WANG Zilong, OUYANG Xingtao, LONG Ruirui
    Computer Engineering and Applications    2023, 59 (9): 46-58.   DOI: 10.3778/j.issn.1002-8331.2212-0008
    Abstract92)      PDF(pc) (752KB)(58)       Save
    Python programming language has gradually become a promising data analysis tool that can be widely used in various fields. However, in the field of hydrology and water resources, there is little research on using Python language for scientific analysis. Firstly, in this thesis, the common Python libraries used in the field of hydrology and water resources are introduced, based on the main research direction and application scenarios of Python language, in this thesis, the main research contents of it in the field of hydrology and water resources are summarized from four aspects:web crawler, data analysis, in-depth learning and Web development. Then, the common algorithms of deep learning used in the field are also summarized. Finally, from the prospects of automatic prediction, edge computation, virtual and augmented reality, reinforcement learning and transfer learning, it is expected to promote the rapid development of the field of hydrology and water resources with the advanced computer technology realized by Python language.
    Reference | Related Articles | Metrics
    Survey on Application and Research of Blockchain Incentive Mechanism in Internet of Vehicles
    ZHANG Tianxiang, LI Leixiao, LIU Dongjiang, GAO Haoyu
    Computer Engineering and Applications    2023, 59 (9): 59-74.   DOI: 10.3778/j.issn.1002-8331.2210-0298
    Abstract57)      PDF(pc) (825KB)(49)       Save
    In recent years, blockchain has become a popular choice to solve security problems in the Internet of vehicles(IoV). A large number of blockchain-based IoV solutions(BIoV) have been produced. However, some blockchain nodes are unwilling to participate in block verification or share due to lack of incentive, which leads to security problems such as privacy leakage and lack of trust in the IoV. Firstly, this paper introduces the research status of BIoV and the three main application scenarios of incentive mechanism in the field of Internet of thing(IoT) and IoV. Then this paper divides the application of blockchain incentive mechanism in the field of IoV from the perspective of reward types. For each incentive in the node participation levels, increase efficiency, reduce energy consumption, privacy, security and node trust five aspects to do a comprehensive comparison. Finally, the challenges and future research directions of the IoV are discussed from the perspective of blockchain incentive, so as to provide more incentive and security support for strengthening the IoV based on blockchain in the future.
    Reference | Related Articles | Metrics
    Review of Cross-Domain Object Detection Algorithms Based on Depth Domain Adaptation
    LIU Hualing, PI Changpeng, ZHAO Chenyu, QIAO Liang
    Computer Engineering and Applications    2023, 59 (8): 1-12.   DOI: 10.3778/j.issn.1002-8331.2210-0063
    Abstract230)      PDF(pc) (583KB)(181)       Save
    In recent years, the object detection algorithm based on deep learning has attracted wide attention due to its high detection performance. It has been successfully applied in many fields such as automatic driving and human-computer interaction and has achieved certain achievements. However, traditional deep learning methods are based on the assumption that the training set (source domain) and the test set (target domain) follow the same distribution, but this assumption is not realistic, which severely reduces the generalization performance of the model. How to align the distribution of the source domain and the target domain so as to improve the generalization of the object detection model has become a research hotspot in the past two years. This article reviews cross-domain object detection algorithms. First, it introduces the preliminary knowledge of cross-domain object detection:depth domain adaptation and object detection. The cross-domain object detection is decomposed into two small areas for an overview, in order to understand its development from the bottom logic. In turn, this article introduces the latest developments in cross-domain object detection algorithms, from the perspectives of differences, confrontation, reconstruction, hybrid and other five categories, and sorts out the research context of each category. Finally, this article summarizes and looks forward to the development trend of cross-domain object detection algorithms.
    Reference | Related Articles | Metrics
    Review of Real-Time Semantic Segmentation Algorithms for Deep Learning
    HE Jiafeng, CHEN Hongwei, LUO Dehan
    Computer Engineering and Applications    2023, 59 (8): 13-27.   DOI: 10.3778/j.issn.1002-8331.2210-0144
    Abstract116)      PDF(pc) (1161KB)(107)       Save
    Semantic segmentation is a technique to segment different objects in a picture from the perspective of pixels and label each pixel in the original picture. However, due to UAV navigation, remote sensing images, medical diagnosis and other application fields, real-time semantic segmentation is needed. Therefore, the real-time semantic segmentation technology based on deep learning has developed rapidly. There are many technologies and models for real-time semantic segmentation. Based on this, on the basis of studying the related literature, the real-time semantic segmentation technology is introduced by semantic segmentation technology, and the advantages of real-time semantic segmentation are briefly described. Then, the important and difficult points of real-time semantic segmentation are discussed. According to the important and difficult points, the existing related technologies and models are expounded, and the advantages and disadvantages of the technologies and models are summarized. Finally, the challenges faced by real-time semantic segmentation are prospected, and the real-time semantic segmentation is summarized, which provides some theoretical references for the follow-up discussion.
    Reference | Related Articles | Metrics
    Review of Research on Real-World Single Image Super-Resolution Reconstruction
    ZHANG Yanqing, MA Jianhong, HAN Ying, CAO Yangjie, LI Jie, YANG Cong
    Computer Engineering and Applications    2023, 59 (8): 28-40.   DOI: 10.3778/j.issn.1002-8331.2208-0223
    Abstract96)      PDF(pc) (725KB)(43)       Save
    Single image super-resolution is an important research topic in the field of computer vision in recent decades. The super-resolution reconstruction algorithm based on deep learning has made breakthroughs. When the super-resolution algorithm is applied to the image in real scene, the effect will be greatly reduced, and serious blur and ringing effect will appear. In this context, more and more researchers are committed to the study of real-world single image SR(RSISR) algorithm. Taking RSISR as the research object, this paper first introduces common image data sets and evaluation indexes, and then analyzes and compares the characteristics, performance and shortcomings of various methods from two aspects:SR methods based on external data sets and SR methods based on internal data sets. Finally, the difficulties and challenges of RSISR are discussed, and the future development trend is considered and prospected.
    Reference | Related Articles | Metrics
    Review of Research on Application of Vision Transformer in Medical Image Analysis
    SHI Lei, JI Qingyu, CHEN Qingwei, ZHAO Hengyi, ZHANG Junxing
    Computer Engineering and Applications    2023, 59 (8): 41-55.   DOI: 10.3778/j.issn.1002-8331.2206-0022
    Abstract120)      PDF(pc) (869KB)(68)       Save
    Deep self-attentive network(Transformer) has a natural ability to model global features and long-range correlations of input information, which is strongly complementary to the inductive bias property of convolutional neural networks(CNN). Inspired by its great success in natural language processing, Transformer has been widely introduced into various computer vision tasks, especially medical image analysis, and has achieved remarkable performance. In this paper, it first introduces the typical work of vision Transformer on natural images, and then organizes and summarizes the related work according to different lesions or organs in the subfields of medical image segmentation, medical image classification and medical image registration, focusing on the implementation ideas of some representative work. Finally, current researches are discussed and the future direction is pointed out. The purpose of this paper is to provide a reference for further in-depth research in this field.
    Reference | Related Articles | Metrics
    Review of Symbolic Execution Technology and Applications
    WU Hao, ZHOU Shilong, SHI Donghui, LI Qiang
    Computer Engineering and Applications    2023, 59 (8): 56-72.   DOI: 10.3778/j.issn.1002-8331.2209-0359
    Abstract77)      PDF(pc) (710KB)(28)       Save
    Symbolic execution is a program analysis technique that has the advantage of finding deep program errors by collecting constraints on program paths and generating high-coverage test cases using constraint solvers. First, the concept and development history of symbolic execution are sorted out, and the intermediate language, path search and constraint solving of symbolic execution techniques are categorized and explained from the core design of symbolic execution system. Then, it investigates the progress of existing research work, selects the most prominent security vulnerabilities, and systematically analyzes the details of the application of symbolic execution technology in terms of vulnerability exploitation and vulnerability detection. Finally, some research results are selected and analyzed according to the characteristics of symbolic execution technology, and the limitations and solutions faced by symbolic execution technology are discussed, and the future trends are foreseen.
    Reference | Related Articles | Metrics
    Survey of Camera Pose Estimation Methods Based on Deep Learning
    WANG Jing, JIN Yuchu, GUO Ping, HU Shaoyi
    Computer Engineering and Applications    2023, 59 (7): 1-14.   DOI: 10.3778/j.issn.1002-8331.2209-0280
    Abstract179)      PDF(pc) (702KB)(137)       Save
    Camera pose estimation is a technology to accurately estimate the 6-DOF position and pose of camera in world coordinate system under known environment. It is a key technology in robotics and automatic driving. With the rapid development of deep learning, using deep learning to optimize camera pose estimation algorithm has become one of the current research hotspots. In order to master the current research status and trends of camera pose estimation algorithms, the mainstream algorithms based on deep learning are summarized. Firstly, the traditional camera pose estimation methods based on feature points is briefly introduced. Then, the camera pose estimation method based on deep learning is mainly introduced. According to the different core algorithms, the end-to-end camera pose estimation, scene coordinate regression, camera pose estimation based on retrieval, hierarchical structure, multi-information fusion and cross scenescamera pose estimation are elaborated and analyzed in detail. Finally, this paper summarizes the current research status, points out the challenges in the field of camera pose estimation based on in-depth performance analysis, and prospects the development trend of camera pose estimation.
    Reference | Related Articles | Metrics
    Graph Neural Network and Its Research Progress in Field of Image Processing
    JIANG Yuying, CHEN Xinyu, LI Guangming, WANG Fei, GE Hongyi
    Computer Engineering and Applications    2023, 59 (7): 15-30.   DOI: 10.3778/j.issn.1002-8331.2205-0503
    Abstract166)      PDF(pc) (659KB)(82)       Save
    Graph neural network (GNN) is a deep learning-based model for processing graph-structured data, which has received much attention from researchers for its good interpretability and powerful nonlinear fitting ability to graph-structured data. With the rise of GNN, GNN has been developed to integrate with image processing techniques and has made breakthroughs in image classification, human body analysis and visual quizzing. Firstly, image processing techniques and the theory of traditional neural networks are introduced, and the principles, characteristics and shortcomings of five major classes of GNNs are analyzed. Secondly, the applications of GNN in the image processing field from five technical levels are analyzed respectively, and the representative models of each class of methods are listed. Thirdly, the common models described in the paper are compared and summarized from the perspective of both datasets and performance evaluation metrics, and nine common public datasets in image processing are introduced in addition. Finally, areas for improvement in GNN in the field of image processingare analyzed in depth, and the prospects of its application in the field of image processing are presented.
    Reference | Related Articles | Metrics
    Survey of Research on Intelligent System for Legal Domain
    LI Jinchen, LI Yanling, GE Fengpei, LIN Min
    Computer Engineering and Applications    2023, 59 (7): 31-50.   DOI: 10.3778/j.issn.1002-8331.2208-0383
    Abstract88)      PDF(pc) (898KB)(39)       Save
    With the improvement of people’s awareness of law and the continuous advancement of judicial digital reform, judicial organs and some corresponding platforms have accumulated a large amount of legal data. On this basis, it is critical to research and develop an intelligent system for legal domain using artificial intelligence algorithm, which cannot only assist legal practitioners in processing large amounts of data, but also provide convenient and cheap legal consultation services to the general public. This paper summarizes the topic of intelligent systems for legal domain, and selects four typical tasks for the research of legal intelligent system according to different application scenarios, which are judicial examination, civil legal question answering, judicial machine reading comprehension and legal judgment prediction. The paper introduces the definition, relevant datasets and evaluation indicators of each type of task in detail. Then, it analyzes the key points and difficulties involved in each type of task, and summarizes reasonable and effective solutions proposed by different research teams for these problems. Based on the comparison and analysis of the latest research progress, it further explores and reveals the main factors restricting the development of legal intelligence system. Finally, it prospects the future development trend of legal intelligent system.
    Reference | Related Articles | Metrics
    Review of Acoustic Signal-Based Industrial Equipment Fault Diagnosis
    ZHOU Yurong, ZHANG Qiaoling, YU Guangzeng, XU Weiqiang
    Computer Engineering and Applications    2023, 59 (7): 51-63.   DOI: 10.3778/j.issn.1002-8331.2208-0079
    Abstract68)      PDF(pc) (628KB)(26)       Save
    In order to guarantee the safety and stability of the industrial production process, it is of great significance and value to adopt reasonable fault diagnosis. Thus, fault diagnosis of industrial equipment has always been a hotspot in the field of industrial control. Firstly, this paper discusses the significance of fault diagnosis, and points out the feasibility and advantages of fault diagnosis based on acoustic signal. Then, according to whether the deep learning is involved, acoustic signal-based fault diagnosis approaches are segmented into traditional-based and deep learning-based categories. Then, it combs the essential ideas and flow of two categories respectively, expounds and summarizes the principle, advantages, limitations, main methods and diagnostic results. Finally, the paper points out the research difficulties, hotspots and the future development direction in the area of industrial equipment fault diagnosis.
    Reference | Related Articles | Metrics
    Summary of Research on Construction of Knowledge Graph for Mine Construction
    QIU Yunfei, XING Haoran, LI Gang
    Computer Engineering and Applications    2023, 59 (7): 64-79.   DOI: 10.3778/j.issn.1002-8331.2205-0409
    Abstract82)      PDF(pc) (905KB)(24)       Save
    At present, massive data have been accumulated in the mine construction work. The use of knowledge graph technology can mine the complex connection between these dynamic data, and provide effective help for the management of mine data and intelligent mine construction. Firstly, this paper analyzes the construction method and data characteristics of mine construction knowledge graph, provides theoretical support for the application of knowledge graph in mine construction, then systematically summarizes the principles and improvement methods of knowledge extraction, knowledge fusion, knowledge reasoning. Finally, it analyzes the application scene and development trend of future knowledge graph in mine construction.
    Reference | Related Articles | Metrics
    Research Status of AGV and Machine Integrated Scheduling
    WU Bin, DING Yuchao, ABLA Basri
    Computer Engineering and Applications    2023, 59 (6): 1-12.   DOI: 10.3778/j.issn.1002-8331.2207-0427
    Abstract138)      PDF(pc) (727KB)(128)       Save
    With the wide application of automated guided vehicles(AGV), the cooperation between machines and AGVs in flexible manufacturing system(FMS) is paid more and more attention. The research of AGV and machine integrated scheduling mainly includes machine allocation, process sequencing, AGV allocation of transport tasks and AGV path planning. This problem is a very complex combinatorial optimization problem, which has important academic significance and application value for its research. Based on the characteristics of the problem, the latest research literatures at home and abroad are reviewed from two aspects of model and algorithm. The constraints and optimization objectives of the existing models are classified in detail, and the representative results of the existing algorithms are summarized from five aspects:genetic algorithm, hybrid optimization algorithm and simulation optimization algorithm and so on. On this basis, the shortcomings of existing research are pointed out, and the content and direction of future research are put forward.
    Reference | Related Articles | Metrics
    Review on Application of Deep Learning in Helmet Wearing Detection
    GAO Teng, ZHANG Xianwu, LI Bai
    Computer Engineering and Applications    2023, 59 (6): 13-29.   DOI: 10.3778/j.issn.1002-8331.2207-0434
    Abstract157)      PDF(pc) (832KB)(128)       Save
    Driven by deep learning, many approaches to object detection have made great progress in the field of industrial security, and the study of helmet-wearing detection has gradually become a significant topic in intelligent image recognition. In order to comprehensively analyze the research status of deep learning technology in helmet wearing detection task, and to facilitate follow-up scientific research personnel to carry out research work, this paper analyzes the state-of-the-art helmet-wearing detection algorithms under deep learning conditions proposed by domestic and foreign scholars in recent years and compares their advantages and limitations. This paper is structured in three sections:the establishment and usage of databases, the predominate algorithms for helmet-wearing detection, the current challenges in the field of helmet-wearing detection. The future research direction of helmet wearing detection field is prospected, and the future research focus in this field is put forward.
    Reference | Related Articles | Metrics
    Survey of Deep Learning Algorithms for Agricultural Pest Detection
    JIANG Xinlu, CHEN Tian’en, WANG Cong, LI Shuqin, ZHANG Hongming, ZHAO Chunjiang
    Computer Engineering and Applications    2023, 59 (6): 30-44.   DOI: 10.3778/j.issn.1002-8331.2205-0604
    Abstract93)      PDF(pc) (679KB)(57)       Save
    Pest detection is a key step in pest forecasting, which is of great significance to pest control, and is also a prerequisite for ensuring crop yield and quality. In recent years, with the rapid development of convolutional neural networks, pest detection technology has entered the era of intelligence, using deep learning related technologies to achieve accurate pest detection has become a research topic that researchers attach great attention to. To facilitate the development of pest detection techniques in deep learning, an overview of existing detection algorithms and datasets will be presented. The four difficult problems currently faced, such as data scarcity, small target detection, multi-scale detection, and dense and occlusion detection, are summarized and the main causes are analyzed. Focusing on the above difficult problems, the improvement strategies and technical details of the deep learning pest detection algorithms proposed in recent years are summarized, as well as the application algorithms for practical scenarios. The performance of various algorithms, the applicable scenarios of improvement strategies and their advantages and disadvantages are compared and analyzed. Finally, the potential development direction of pest detection is analyzed and prospected from the aspects of complex detection scenarios, lack of data, incremental update model and application landing.
    Reference | Related Articles | Metrics
    Review of Research on Bearing Fault Diagnosis with Small Samples
    SI Weiwei, CEN Jian, WU Yinbo, HU Xueliang, HE Minzan, YANG Zhuohong, CHEN Honghua
    Computer Engineering and Applications    2023, 59 (6): 45-56.   DOI: 10.3778/j.issn.1002-8331.2208-0139
    Abstract119)      PDF(pc) (701KB)(57)       Save
    With the advent of the data era, bearing fault diagnosis methods based on data-driven have shown superior performance, but such methods rely on a large number of labeled data, and it is difficult to collect a large amount of data in the actual production process, so bearing fault diagnosis with small samples has high research value. In this paper, the bearing fault diagnosis methods under the condition of small samples are reviewed, and divided into two categories:data-based methods and model-based methods. The data-based method expands the original samples from the perspective of data. The model-based methods refer to the use of models to optimize feature extraction or improve classification accuracy. Finally, the shortcomings of current fault diagnosis methods under the condition of small samples are summarized, and future research directions of bearing fault diagnosis with small samples are prospected.
    Reference | Related Articles | Metrics
    Survey of Evaluation Metrics and Methods for Semantic Segmentation
    YU Ying, WANG Chunping, FU Qiang, KOU Renke, WU Weiyi, LIU Tianyong
    Computer Engineering and Applications    2023, 59 (6): 57-69.   DOI: 10.3778/j.issn.1002-8331.2207-0139
    Abstract96)      PDF(pc) (740KB)(66)       Save
    Deep learning has made major breakthroughs in the field of semantic segmentation. Standard, well-known and comprehensive metrics should be used to evaluate the performance of these algorithms to ensure objectivity and effectiveness of the evaluation. Through summary of the existing semantic segmentation evaluation metrics, this paper elaborates from some aspects, e.g., pixel accuracy, depth estimation error metric, operation efficiency, memory demand and robustness. Especially, the widely used accuracy metrics such as F1 score, mIoU, mPA, Dice coefficient and Hausdorff distance are introduced in detail. In addition, this paper expounds the related research on the robustness and generalization. Furthermore, this paper points out the requirements in the semantic segmentation experiment and the limitations of segmentation quality evaluation.
    Reference | Related Articles | Metrics
    Review of Deep Reinforcement Learning Model Research on Vehicle Routing Problems
    YANG Xiaoxiao, KE Lin, CHEN Zhibin
    Computer Engineering and Applications    2023, 59 (5): 1-13.   DOI: 10.3778/j.issn.1002-8331.2210-0153
    Abstract195)      PDF(pc) (1036KB)(182)       Save
    Vehicle routing problem(VRP) is a classic NP-hard problem, which is widely used in transportation, logistics and other fields. With the scale of problem and dynamic factor increasing, the traditional method of solving the VRP is challenged in computational speed and intelligence. In recent years, with the rapid development of artificial intelligence technology, in particular, the successful application of reinforcement learning in AlphaGo provides a new idea for solving routing problems. In view of this, this paper mainly summarizes the recent literature using deep reinforcement learning to solve VRP and its variants. Firstly, it reviews the relevant principles of DRL to solve VRP and sort out the key steps of DRL-based to solve VRP. Then it systematically classifies and summarizes the pointer network, graph neural network, Transformer and hybrid models four types of solving methods, meanwhile this paper also compares and analyzes the current DRL-based model performance in solving VRP and its variants. Finally, this paper sums up the challenge of DRL-based to solve VRP and future research directions.
    Reference | Related Articles | Metrics
    Overview of Research on Spatial Registration Algorithms of Underwater Opti-Acoustic Images
    GUO Yinjing, MA Xinrui, XU Yuecheng, KONG Fang, LYU Wenhong
    Computer Engineering and Applications    2023, 59 (5): 14-27.   DOI: 10.3778/j.issn.1002-8331.2208-0313
    Abstract90)      PDF(pc) (851KB)(44)       Save
    Underwater opti-acoustic image alignment is a key technology for information fusion in underwater devices. Based on a brief description of the concept and examples of underwater opti-acoustic image alignment, the current relevant algorithms for underwater opti-acoustic image reconstruction and recovery are analysed, the research progress of region and feature-based alignment algorithms for underwater heterogenous images are reviewed in detail, the development status of two research directions with high accuracy based on the similarity of image domain and shape features is focused on, and according to the research hotspots of heterogenous images alignment in other fields. The development trend of underwater opti-acoustic image alignment research is foreseen in terms of increasing structural constraints on imaging models, introducing phase congruency, generative adversarial networks and other algorithms to improve the alignment accuracy.
    Reference | Related Articles | Metrics
    Survey of Artificial Intelligence in COVID-19 Pandemic
    SUN Shukui, FAN Jing, LI Zhanwen, QU Jinshuai, LU Peidong
    Computer Engineering and Applications    2023, 59 (5): 28-39.   DOI: 10.3778/j.issn.1002-8331.2208-0229
    Abstract123)      PDF(pc) (704KB)(48)       Save
    The outbreak of novel coronavirus pneumonia(COVID-19) poses a great threat to the safety of human life and property in the world. Artificial intelligence(AI) has played an irreplaceable role in helping to win the battle against the epidemic. With the help of AI, the shortage of medical resources has been greatly alleviated, the efficiency of medical diagnosis has been improved, and the risk of contact infection has been avoided. Firstly, the background knowledge of COVID-19 and AI is expounded. Then, the research progress of AI in COVID-19 is discussed from the eight aspects of epidemic prevention and control, including epidemic trend prediction, epidemic traceability, detection and diagnosis, drug development, vaccine development, drug reuse, network public opinion control and genome sequencing, and the challenges faced by AI in the epidemic are listed. Then it discusses the impact of the epidemic on China’s AI industry and the dialectical relationship between the two. Finally, it summarizes the full text.
    Reference | Related Articles | Metrics
    Overview of Image Edge Detection
    XIAO Yang, ZHOU Jun
    Computer Engineering and Applications    2023, 59 (5): 40-54.   DOI: 10.3778/j.issn.1002-8331.2209-0122
    Abstract118)      PDF(pc) (921KB)(76)       Save
    The task of edge detection is to identify pixels with significant brightness changes as target edges, which is a low-level problem in computer vision, and edge detection has important applications in object recognition and detection, object proposal generation, and image segmentation. Nowadays, edge detection has produced several types of methods, such as traditional gradient-based detection methods and deep learning-based edge detection algorithms and detection methods combined with emerging technologies. A finer classification of these methods provides researchers with a clearer understanding of the trends in edge detection. Firstly, the theoretical basis and implementation methods of traditional edge detection are introduced; then the main edge detection methods in recent years are summarized and classified according to the methods used, and the core techniques used in them are introduced, such as branching structure, feature fusion and loss function. The evaluation indicators used to assess the algorithm’s performance are single-image optimal threshold(ODS) and frame per second(FPS), which are contrasted using the fundamental data set(BSDS500). Finally, the current state of edge detection research is examined and summarized, and the possible future research directions of edge detection are prospected.
    Reference | Related Articles | Metrics
    Survey of Graph Neural Networks in Session Recommender Systems
    ZHU Zhiguo, LI Weiyue, JIANG Pan, ZHOU Peiyao
    Computer Engineering and Applications    2023, 59 (5): 55-69.   DOI: 10.3778/j.issn.1002-8331.2207-0397
    Abstract83)      PDF(pc) (5333KB)(60)       Save
    Based on the current session of the target user, session recommenders learn the dependency relationship among items according to the auxiliary information such as item category, cross-session context, and user’s multi behaviors, to capture the long-term and short-term preferences of users for personalized recommendations. In recent years, a series of popular algorithms based on deep learning have become the forefront methods for session recommendation systems. The introduction of graph neural networks further improves the session recommendation system’s performance. Given this, the review starts with question definition, session recommendation factors, and composition analysis. Then the related works are divided into session recommendation systems based on graph convolution networks, gated graph neural networks, graph attention networks, and other architectures. After that, these works are summarized and compared. Finally, the categories of loss functions, selected data sets, and model performance evaluation indexes are deeply analyzed. The paper mainly evaluates and sorts out each model framework from algorithm principle and performance analysis, aiming at reviewing, summarizing, and looking into the related work of session recommendation systems based on graph neural networks in the last five years.
    Reference | Related Articles | Metrics
    Survey on Deep-Learning-Based Long-Term Object Tracking Algorithms
    LIANG Yitao, HAN Yongbo, LI Lei
    Computer Engineering and Applications    2023, 59 (4): 1-17.   DOI: 10.3778/j.issn.1002-8331.2206-0507
    Abstract196)      PDF(pc) (918KB)(171)       Save
    In the field of visual target tracking, long-term tracking has been paid more and more attention by researchers, because it contains more realistic challenging scenarios, such as occlusion, similar object interference and target disappearance. However, traditional long-term tracking algorithms are inefficient and have been unable to meet the application requirements of tracker performance in fields, such as video surveillance and autonomous driving. Recently, a lot of work has rapidly advanced the development of long-term tracking techniques by introducing deep neural networks. In order to analyze the current situation and future development of deep-learning-based long-term tracking algorithms, firstly, by comparing the long-term and short-term tracking datasets and their evaluation indicators, the requirements and difficulties of long-term tracking tasks are summarized, and the development of long-term tracking datasets and evaluation systems is introduced. Subsequently, based on the design framework of deep-learning-based long-term tracking algorithm, the design ideas of each component of the framework are described in detail. Then, taking the long-term tracking strategy as the starting point, the existing research work is analyzed, and the advantages and disadvantages of different models and their characteristics are summarized. Finally, based on the summary of existing research work, the challenges faced in this field are discussed, and the future research trends are presented.
    Reference | Related Articles | Metrics
    Research Advances on Graph Neural Network Recommendation of Knowledge Graph Enhancement
    WU Guodong, WANG Xueni, LIU Yuliang
    Computer Engineering and Applications    2023, 59 (4): 18-29.   DOI: 10.3778/j.issn.1002-8331.2205-0268
    Abstract143)      PDF(pc) (638KB)(106)       Save
    The existing recommendation methods are mainly based on the users’ historical interaction behavior, and the user and item-related feature information are not fully utilized, resulting in the effect of the recommendation is not ideal. The graph neural network(GNN) recommendation enhanced by knowledge graph(KG) is based on the interaction graph constructed by user and item interaction behavior, and the knowledge graph with the same graph structure is introduced and processed by the graph neural network technology, so as to realize personalized recommendation. In this paper, the research progress of graph neural network recommendation enhanced by existing knowledge graph is discussed. Firstly, on the basis of the discussion of graph neural network recommendation and knowledge graph recommendation, the relevant research results of graph neural network recommendation enhanced by the current knowledge graph are deeply analyzed from the aspects of item knowledge graph and collaborative knowledge graph. Then, the main problems in the graph neural network recommendation research based on the existing knowledge graph enhancement are pointed out from the aspects of large-scale dynamic knowledge graph processing, user preference mining for item attributes, knowledge graph embedding learning problem and so on. Finally, the main research directions of GNN recommendation enhanced by knowledge graph in the future are predicted from the following aspects:GNN recommendation enhanced by knowledge graph in dynamic sequential sequence, GNN recommendation enhanced by knowledge graph in meta-learning, GNN recommendation enhanced by multi-model knowledge graph, GNN cross-domain recommendation enhanced by knowledge graph and so on.
    Reference | Related Articles | Metrics
    Review of Recommendation Systems Using Knowledge Graph
    ZHANG Mingxing, ZHANG Xiaoxiong, LIU Shanshan, TIAN Hao, YANG Qinqin
    Computer Engineering and Applications    2023, 59 (4): 30-42.   DOI: 10.3778/j.issn.1002-8331.2209-0033
    Abstract166)      PDF(pc) (702KB)(102)       Save
    With the rapid development of the Internet, how to obtain the needed information from huge amounts of data becomes more important. The recommendation system is a method of screening information, which aims to recommend personalized content for users. However, traditional recommendation algorithms still suffer from several challenges, such as data sparsity and cold start. In recent years, researchers have used the rich entity and relationship information in the knowledge graph to alleviate the above problems. The overall performance of the recommendation system is enhanced. This paper gives a review of the recommendation system based on knowledge graph from three aspects:Firstly, basic concepts of the recommendation system and knowledge graph are introduced. The shortcomings of the existing recommendation algorithms are pointed out. Then, the research of the recommendation system based on knowledge graph is analyzed in detail. The advantages and challenges of the different approaches are assessed. Finally, relevant application scenarios and future development prospects are summarized.
    Reference | Related Articles | Metrics
    Survey of Short Text Classification Methods Based on Deep Learning
    GAN Yating, AN Jianye, XU Xue
    Computer Engineering and Applications    2023, 59 (4): 43-53.   DOI: 10.3778/j.issn.1002-8331.2209-0048
    Abstract157)      PDF(pc) (609KB)(79)       Save
    From five aspects of CNN, RNN, CNN-RNN, GCN and other deep learning methods, the research status of their application in short text classification is comprehensively analyzed, their advantages and disadvantages are compared, and the commonly used labeled datasets are summarized. The results show that:At present, the application research of deep learning in short text classification mainly focuses on the improvement of efficient algorithms and the expansion of text information. At the same time, the research on constructing labeled datasets for model testing is in the initial stage, mostly for specific fields such as movie reviews, commodity reviews, news, etc., which needs continuous improvement. In the future, the research will focus on algorithm improvement, information expansion and their mutual integration, to explore some specific applications with good classification effect in practice.
    Reference | Related Articles | Metrics
    Review of Depression Detection Using Social Media Text Data
    XU Dongdong, CAI Xiaohong, LIU Jing, CAO Hui
    Computer Engineering and Applications    2023, 59 (4): 54-63.   DOI: 10.3778/j.issn.1002-8331.2209-0042
    Abstract111)      PDF(pc) (598KB)(72)       Save
    Machine learning has been gradually applied to depression detection using social media text data, and has prominently shown important application value in recent years. Firstly, this paper organizes and classifies social media text datasets, data preprocessing and machine learning methods used for depression detection. In addition, in terms of data feature representation, the basic feature representation, static and contextual word embedding are compared and analyzed. Secondly, this paper analyzes comprehensively the performance and characteristics of traditional machine learning with different basic features and different algorithm types as well as deep learning for depression detection. Finally, this paper summarizes and suggests further explorations in Chinese dataset creation, model interpretability, metaphor-based detection and lightweight pre-training model.
    Reference | Related Articles | Metrics