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: Download Citations EndNote Ris BibTeX Toggle Thumbnails Select Survey of Few-Shot Image Classification Based on Deep Meta-Learning ZHOU Bojun, CHEN Zhiyu Computer Engineering and Applications    2024, 60 (8): 1-15.   DOI: 10.3778/j.issn.1002-8331.2308-0271 Abstract （43）      PDF（pc） （1091KB）（75）       Knowledge map    Save Deep meta-learning has emerged as a popular paradigm for addressing few-shot classification problems. A comprehensive review of recent advancements in few-shot image classification algorithms based on deep meta-learning is provided. Starting from the problem description, the categorizes of the algorithms based on deep meta-learning for few-shot image classification are summarized, and commonly used few-shot image classification datasets and evaluation criteria are introduced. Subsequently, typical models and the latest research progress are elaborated in three aspects: model-based deep meta-learning methods, optimization-based deep meta-learning methods, and metric-based deep meta-learning methods. Finally, the performance analysis of existing algorithms on popular public datasets is presented, the research hotspots in this topic are summarized, and its future research directions are discussed. Reference | Related Articles | Metrics Select Research Progress on Deep Learning in Field of Diabetic Retinopathy Classification SUN Shilei, LI Ming, LIU Jing, MA Jingang, CHEN Tianzhen Computer Engineering and Applications    2024, 60 (8): 16-30.   DOI: 10.3778/j.issn.1002-8331.2307-0330 Abstract （30）      PDF（pc） （655KB）（24）       Knowledge map    Save Diabetic retinopathy is one of the primary causes of visual impairment in diabetic patients, and early classification and diagnosis are of significant importance for disease management and control. Deep learning methods have the capability to automatically extract features of retinal lesions and perform classification, making them essential tools for diabetic retinopathy classification. This paper begins by introducing commonly used datasets and evaluation metrics for diabetic retinopathy, summarizing the applications of deep learning in binary classification of diabetic retinopathy. It then provides an overview of various classical deep learning models used for severity classification of diabetic retinopathy, focuses on the classification and diagnosis methods of convolutional neural networks, and makes a comprehensive comparative analysis of different approaches. Finally, the paper discusses the challenges in this field and provides an outlook on future directions for research and development. Reference | Related Articles | Metrics Select Review of Construction and Applications of Spatio-Temporal Graph Neural Network in Traffic Flow Prediction WANG Weitai, WANG Xiaoqiang, LI Leixiao, TAO Yihao, LIN Hao Computer Engineering and Applications    2024, 60 (8): 31-45.   DOI: 10.3778/j.issn.1002-8331.2307-0133 Abstract （27）      PDF（pc） （827KB）（29）       Knowledge map    Save The prediction of traffic flow is a pivotal concern within urban traffic management and planning, yet conventional forecasting techniques prove inadequate in addressing challenges like data sparsity, nonlinear associations, and intricate dynamics. Graph neural network is a deep learning approach based on non-Euclidean structural data, which has been widely used in various complex network modeling and predictive tasks in recent times. To address traffic flow prediction, a spatiotemporal graph neural network is proposed, which can capture spatial and temporal correlations, making significant progress compared to earlier predictive models. An analysis is conducted on models utilizing spatiotemporal graph neural network for the prediction of traffic flow in recent times Firstly, various construction methods of adjacency matrices are summarized and compared. Then, the common components of traffic flow prediction models are listed from the perspective of spatial correlation and temporal correlation, and different spatio-temporal fusion modes are classified and compared. On the application front, spatiotemporal graph neural network models are categorized into three classes based on temporal scales: long-term prediction, short-term prediction, and combined long-short-term prediction. Analysis of respective objectives and requisites is conducted, accompanied by enumeration and comparison of prominent recent models. Finally, limitations of existing research are deliberated upon, and prospects for future studies pertaining to relevant models are outlined. Reference | Related Articles | Metrics Select Review on Detection of Drones and Birds in Photoelectric Images Based on Deep Learning Convolutional Neural Network XIE Weiyu, ZHANG Qiang Computer Engineering and Applications    2024, 60 (8): 46-55.   DOI: 10.3778/j.issn.1002-8331.2305-0372 Abstract （32）      PDF（pc） （613KB）（25）       Knowledge map    Save With the development of the civilian drone industry, drones have become a critical issue affecting public safety. At present, the surveillance method for low-altitude drones mainly adopts the method of radar detection combined with visible image identification. However, visible image recognition is susceptible to interference from flying birds, which belongs to the same “low, slow, and small” targets as UAVs. To eliminate the interference of flying bird targets in the detection of UAVs based on visible images, the deep neural network is used to accurately identify and classify UAVs and flying birds in visible images, and effectively eliminate the interference of birds in the detection of UAVs. This paper first systematically explains the development process of target detection technology, discusses the differences of various target detection algorithms based on deep learning network, and compares the advantages and disadvantages of various algorithms. The image data sets that can be used for drone and bird detection are sorted out and introduced, and the existing results of related research are analyzed. Then, starting from the practical application, the problems that may exist in the detection of drones and birds are sorted out, and the research on neural networks that can solve the corresponding detection problems is elaborated and analyzed. In the end, the probable future directions of this research are prospected. Reference | Related Articles | Metrics Select Review of Supervised Topic Models and Applications WANG Zhenbiao, XU Zhenshun, LIU Na, ZHANG Wenhao, TANG Zengjin, WANG Zheng’an Computer Engineering and Applications    2024, 60 (8): 56-68.   DOI: 10.3778/j.issn.1002-8331.2309-0030 Abstract （12）      PDF（pc） （856KB）（16）       Knowledge map    Save Topic model is a data mining method that can automatically extract potential patterns or topics from a large number of files or data, and assign the corresponding data to the corresponding patterns or topics. Topic models have been widely used in the fields of text clustering or classification, topic extraction, topic evolution, sentiment analysis and summary. The difference between a supervised topic model and an unsupervised topic model is whether it relies on annotation information. In recent years, supervised topic model has gradually emerged in data mining tasks, which makes more and more tasks tend to adopt supervised method for optimization. Firstly, the content of supervised topic model is presented, and the commonly used data sets and evaluation indicators are introduced. Secondly, from the perspective of model and application, different types of supervised topic models are analyzed in depth. Finally, the challenges facing the current research of thematic models are described, and the future research direction of supervised thematic models is prospected. Reference | Related Articles | Metrics Select Review of Development of Deep Learning Optimizer CHANG Xilong, LIANG Kun, LI Wentao Computer Engineering and Applications    2024, 60 (7): 1-12.   DOI: 10.3778/j.issn.1002-8331.2307-0370 Abstract （165）      PDF（pc） （1327KB）（217）       Knowledge map    Save Optimization algorithms are the most critical  factor in improving the performance of deep learning models, achieved by minimizing the loss function. Large language models (LLMs), such as GPT, have become the research focus in the field of natural language processing, the optimization effect of traditional gradient descent algorithm has been limited. Therefore, adaptive moment estimation algorithms have emerged, which are significantly superior to traditional optimization algorithms in generalization ability. Based on gradient descent, adaptive gradient, and adaptive moment estimation algorithms, and the pros  and cons of optimization algorithms are analyzed. This paper applies optimization algorithms to the Transformer architecture and selects the French-English translation task as the evaluation benchmark. Experiments have shown that adaptive moment estimation algorithms can effectively improve the performance of the model in machine translation tasks. Meanwhile, it discusses the development direction and applications of optimization algorithms. Reference | Related Articles | Metrics Select Survey of Deep Learning-Based on Emotion Generation in Conversation ZHOU Yutong, MA Zhiqiang, XU Biqi, JIA Wenchao, LYU Kai, LIU Jia Computer Engineering and Applications    2024, 60 (7): 13-25.   DOI: 10.3778/j.issn.1002-8331.2304-0131 Abstract （84）      PDF（pc） （573KB）（137）       Knowledge map    Save Emotion generation is a subtask in the study of artificial affective computing, where the emotion generation task in a dialog system aims to generate emotion categories in the discourse to be replied to. Conversational emotion generation aims to generate emotion categories in the discourse to be replied, which can promote the research of conversational emotion understanding and conversational expression, and also has important theoretical significance and practical application value in many intelligent fields such as intelligent gossip bots, emotional comfort, recommendation systems and human-computer emotional interaction. Thanks to the excellent performance of deep neural networks in the field of natural language processing, deep learning-based emotion generation for conversational systems has received more and more attention from researchers. To summarize the current work related to deep learning-based conversational emotion generation, the current research on emotion generation for conversational systems using deep learning mainly contains three aspects：emotion perception, emotion prediction and emotion decision making. Some commonly used emotion dialogue datasets are briefly introduced, and finally, an overview of current issues in this task is summarized and future trends are prospected. Reference | Related Articles | Metrics Select Deep Learning in Aided Diagnosis of Osteoporosis JIANG Liang, ZHANG Cheng, WEI Dejian, CAO Hui, DU Yuzheng Computer Engineering and Applications    2024, 60 (7): 26-40.   DOI: 10.3778/j.issn.1002-8331.2305-0030 Abstract （73）      PDF（pc） （844KB）（102）       Knowledge map    Save Osteoporosis, a systemic disease with an increased risk of fracture, is caused by a decrease in bone density. In clinical practice, the diagnosis is based on the imaging examination. In recent years, deep learning methods have made breakthroughs in the field of skeletal medical image processing. This paper compares and summarizes the deep learning methods that have been used in osteoporosis assisted diagnosis. Firstly, it introduces the commonly used datasets. Secondly, it systematically expounds the application of convolutional neural network, recurrent neural network, deep belief network and generative adversarial network in the classification of osteoporosis. And then, the application of full convolutional networks and U-Net in the segmentation of the lesion area of osteoporosis is described. And it introduces the model of the latest AI ChatGPT potential applications. Then it compares the performance of the different models. Finally, the paper points out the existing difficulties and puts forward the corresponding prospects in this field. Reference | Related Articles | Metrics Select Survey of Clustering Ensemble Research SHAO Chao, RUN Qingchen Computer Engineering and Applications    2024, 60 (7): 41-57.   DOI: 10.3778/j.issn.1002-8331.2307-0050 Abstract （49）      PDF（pc） （851KB）（60）       Knowledge map    Save As a basic technology in the field of data research, cluster analysis aims to discover meaningful cluster structure from unlabeled datasets. According to Kleinberg’s theorem, there is no basic clustering algorithm that can learn any dataset, which means that no clustering method can correctly find the cluster structure of all datasets. Clustering ensemble addresses this inherent challenges by combining multiple clustering results to explore the final clustering with high stability and robustness. In recent years, many clustering ensemble techniques have been proposed, resulting in new ways to solve practical problem together with new application areas of these techniques. Clustering ensemble techniques are summarized from the two dimensions of basic clustering generation mechanism and consensus function design, the advantages and disadvantages of various methods are analyzed, and experimental comparisons are made. Finally, the future research directions are discussed based on the current research status. Reference | Related Articles | Metrics Select Progress of Instantiated Reality Augmentation Method for Smart Phone Indoor Scene Elements LIU Jianhua, WANG Nan, BAI Mingchen Computer Engineering and Applications    2024, 60 (7): 58-69.   DOI: 10.3778/j.issn.1002-8331.2309-0376 Abstract （48）      PDF（pc） （623KB）（51）       Knowledge map    Save Indoor mobile phone navigation and location services are current research hotspots, of which scene element instantiation reality augmentation methods are an important part. Instantiated segmentation is a challenging and fundamental task in scene element perception, and augmented reality is an effective way to apply digital twin building maps, both of which are of great importance in the field of indoor location navigation. At present, augmented reality technology is mainly applied to semantic enhancements in scenes, and AR enhancements for smartphone indoor navigation only stay in the visual visualization effect, and have not yet really penetrated to the level of enhancement of elemental instances in scenes. To address this problem, this paper proposes an AR research idea of mobile phone scene element instantiation, by identifying objects in indoor scenes and matching them with building maps, the corresponding stored element information in the building maps will be enhanced and displayed using AR technology, thus assisting pedestrians in indoor navigation and location services and other related applications, and improving the information level of location services such as indoor positioning and navigation for users. This paper provides a systematic overview of instance segmentation and augmented reality methods for smartphone-side video, and analyses the characteristics and applicable scenarios of the relevant methods, summaries the research progress of instance segmentation and augmented reality in mobile, and finally discusses the application prospects of instantiated reality augmentation methods for indoor scene elements in the field of navigation and location services. Reference | Related Articles | Metrics Select Survey on Distributed Assembly Permutation Flowshop Scheduling Problem ZHANG Jing, SONG Hongbo, LIN Jian Computer Engineering and Applications    2024, 60 (6): 1-9.   DOI: 10.3778/j.issn.1002-8331.2307-0276 Abstract （148）      PDF（pc） （619KB）（150）       Knowledge map    Save As the rapid development of modern manufacturing, the past decades have witnessed a trend in which jobs are firstly processed in distributed production factories and then assembled into the final products in an assembly factory after completion. Such manufacturing mode brings many advantages as well as some new challenges on resource scheduling. This paper surveys literature on the distributed assembly permutation flowshop scheduling problem (DAPFSP). Firstly, the background and main issues in DAPFSP are introduced. Then, mathematical models, encoding and decoding schemes, and global and local search algorithms are thoroughly discussed for DAPFSP with the objective of minimizing the maximal completion time. Additionally, recent advances on DAPFSP with various objectives, such as total flow time, DAPFSP with other constraints like no-wait, and DAPFSP by taking issues including setup time into consideration are also surveyed. Finally, several future research directions worthy further investigation are pointed out. Reference | Related Articles | Metrics Select Review of Unsupervised Domain Adaptation in Medical Image Segmentation HU Wei, XU Qiaozhi, GE Xiangwei, YU Lei Computer Engineering and Applications    2024, 60 (6): 10-26.   DOI: 10.3778/j.issn.1002-8331.2307-0421 Abstract （112）      PDF（pc） （756KB）（116）       Knowledge map    Save Medical image segmentation has broad application prospects in the field of medical image processing, providing auxiliary information for diagnosis and treatment by locating and segmenting interested organs, tissues, or lesion areas. However, there is a domain offset problem between different modalities of medical images, which can lead to a significant decrease in the performance of the segmentation model during actual deployment. Domain adaptation technology is an effective way to solve this problem, especially unsupervised domain adaptation, which has become a research hotspot in the field of medical image processing because it does not require target domain label information. At present, there are relatively few review reports on unsupervised domain adaptation research in medical image segmentation. Therefore, this paper summarizes, analyzes, and prospects the future of unsupervised domain adaptation research in medical image segmentation in recent years, hoping to help relevant researchers quickly understand and familiarize themselves with the current research status and trends in this field. Reference | Related Articles | Metrics Select Survey on Test Input Selection and Metrics for Deep Neural Networks YAN Hong, YANG Fengyu, ZHONG Yihui, XIONG Yu, CHEN Yu’an Computer Engineering and Applications    2024, 60 (6): 27-42.   DOI: 10.3778/j.issn.1002-8331.2307-0382 Abstract （63）      PDF（pc） （651KB）（62）       Knowledge map    Save As deep neural networks are widely used in various fields, it is particularly important to test and evaluate them and ensure their safety. When the test dataset is large and the labelled cost is expensive, the test input selection method can select and sort the test samples to improve the test efficiency and test coverage. In order to further understand the research progress in the field of test input selection for deep neural networks, 91 academic papers in related fields over the past five years are systematically sorted out. Firstly, the basic concepts and processes of deep neural networks testing are introduced, including the construction of deep learning systems, test input selecting and the test metrics. Secondly, the paper outlines and analyzes the applicable scenarios and shortcomings of various metrics and test input selection methods, as well as the interconnections among them. Finally, current challenges and opportunities for deep neural networks test input selection and metrics are pointed out. Reference | Related Articles | Metrics Select Survey of Temporal Knowledge Graph Completion Methods XIAO Lei, LI Qi Computer Engineering and Applications    2024, 60 (6): 43-54.   DOI: 10.3778/j.issn.1002-8331.2307-0083 Abstract （67）      PDF（pc） （676KB）（84）       Knowledge map    Save Knowledge graph completion is a research hotspot in recent years, and it has broad application prospects in downstream applications, such as knowledge question answering, recommended system and intelligent search, etc. However, most of the completion methods ignore the dynamic characteristics of the knowledge graphs, many of which the facts will change over time. The new temporal knowledge graph completion methods take into account the limitations of the previous by incorporating time information, enabling the dynamic changes of the knowledge graph over time to be well captured. In response to the big potential of temporal knowledge graph completion methods in research fields such as social networks, transportation, finance and trade with complex time dependent characteristics, this paper summarizes temporal knowledge graph completion techniques. Based on the main different principle of model usage, completion methods based on logical rules, tensor decomposition, translation model, neural networks, deep reinforcement learning, and language model are summarized. The commonly used evaluation indicators, datasets, core ideas, advantages and disadvantages, applicable scenarios, and improvements on corresponding static models of existing methods are summarized. Finally, it looks forward to the future research directions. Reference | Related Articles | Metrics Select Review of Deep Learning Methods for Palm Vein Recognition TAN Zhenlin, LIU Ziliang, HUANG Aiquan, CHEN Huihui, ZHONG Yong Computer Engineering and Applications    2024, 60 (6): 55-67.   DOI: 10.3778/j.issn.1002-8331.2306-0168 Abstract （86）      PDF（pc） （664KB）（65）       Knowledge map    Save Palm vein recognition, as a new infrared biometrics technology, has become one of the research hotspots in the field of biometric recognition because of its advantages of high security and liveness detection. In recent years, a great deal of research in this field has promoted the development of palm vein recognition technology by introducing deep learning methods. In order to grasp the latest research status and development direction in the field of palm vein recognition, data acquisition and the mainstream algorithms of data pre-processing are classified and summarized, and the latest progress of palm vein recognition based on deep learning is classified and elaborated in terms of palm vein feature representation, network design and optimization, and lightweight network. In view of the bottleneck of single-modal recognition, the correlation algorithms of multimodal and multi-feature fusion recognition are analyzed and compared. The difficulties and challenges of current research on palm vein recognition are discussed, and the future development trends are prospected and summarized. Reference | Related Articles | Metrics Select Survey on Attack Methods and Defense Mechanisms in Federated Learning ZHANG Shiwen, CHEN Shuang, LIANG Wei, LI Renfa Computer Engineering and Applications    2024, 60 (5): 1-16.   DOI: 10.3778/j.issn.1002-8331.2306-0243 Abstract （176）      PDF（pc） （792KB）（204）       Knowledge map    Save The attack and defense techniques of federated learning are the core issue of federated learning system security. The attack and defense techniques of federated learning can significantly reduce the risk of being attacked and greatly enhance the security of federated learning systems. Deeply understanding the attack and defense techniques of federated learning can advance research in the field and achieve its widespread application of federated learning. Therefore, it is of great significance to study the attack and defense techniques of federated learning. Firstly, this paper briefly introduces the concept, basic workflow, types, and potential existing security issues of federated learning. Subsequently, the paper introduces the attacks that the federated learning system may encounter, and relevant research is summarized during the introduction. Then, starting from whether the federated learning system has targeted defense measures, the defense measures are divided into two categories：universal defense measures and targeted defense measures, and targeted summary are made. Finally, it reviews and analyzes the future research directions for the security of federated learning, providing reference for relevant researchers in their research work on the security of federated learning. Reference | Related Articles | Metrics Select Research on Blockchain P2P Network and Its Security NI Xueli, MA Zhuo, WANG Qun Computer Engineering and Applications    2024, 60 (5): 17-29.   DOI: 10.3778/j.issn.1002-8331.2307-0218 Abstract （90）      PDF（pc） （626KB）（93）       Knowledge map    Save Blockchain is proposed as a technology integrating various technological innovations such as distributed ledger, cryptography, consensus algorithm and P2P (peer-to-peer) network. It realizes decentralization, traceability and tamper resistance in an open environment with mutual distrust nodes. Blockchain has solved the problems of trust establishment, security management and privacy protection that have been plaguing the traditional centralized service architecture, and has been drawing broad attention from various fields of society. Among them, P2P network realizes the final consistency of distributed ledger through consensus algorithm, and utilizes various protocols to coordinate the services provided by each node. Its functionalities are continuously enriched and improved with each technological iteration of blockchain, thus forming a relatively independent blockchain P2P network, which has become the focus of blockchain research. Firstly, this paper systematically introduces the characteristics and key technologies of P2P network, especially the overlay network, resource query and jitter of P2P network. Secondly, the typical blockchain application scenarios are used to analyze the working mechanism and operation characteristics of blockchain P2P network from the perspective of architecture and protocol implementation. Then, the paper introduces the security risks of blockchain P2P network, and discusses the implementation process and prevention methods of typical attacks. Finally, the summary is given by exploring the security challenges and key technical responses in blockchain networks. Reference | Related Articles | Metrics Select Review of Deep Learning Approaches for Recognizing Multiple Unsafe Behaviors in Workers SU Chenyang, WU Wenhong, NIU Hengmao, SHI Bao, HAO Xu, WANG Jiamin, GAO Le, WANG Weitai Computer Engineering and Applications    2024, 60 (5): 30-46.   DOI: 10.3778/j.issn.1002-8331.2307-0168 Abstract （109）      PDF（pc） （808KB）（116）       Knowledge map    Save With the development of deep learning, target detection and behavior recognition methods have made great progress in the field of worker unsafe behavior recognition, this paper systematically summarizes the relevant research work at home and abroad in recent years, elaborates the commonly used models and effects of target detection methods and behavior recognition methods, focuses on reviewing the application of the two types of methods in the recognition of unsafe behaviors and the relevant research on the combination of the two types of methods, and provides a comprehensive analysis and comparison on the advantages, limitations, recognized behavior categories and applicable scenarios of various methods are comprehensively analyzed and compared. On this basis, the optimization measures for target detection and behavior recognition in recent years are summarized, the commonly used optimization directions and means are summarized, the improvement methods successfully applied in unsafe behavior recognition are summarized, the difficulties and problems in this research field are sorted out, and the suggestions and future development trends are given, which will provide references and lessons for the research in this field. Reference | Related Articles | Metrics Select Review of Applications of Deep Learning in Fracture Diagnosis Abudukelimu Halidanmu, FENG Ke, SHI Yaqing, Abudukelimu Nihemaiti, Abulizi Abudukelimu Computer Engineering and Applications    2024, 60 (5): 47-61.   DOI: 10.3778/j.issn.1002-8331.2304-0112 Abstract （77）      PDF（pc） （636KB）（67）       Knowledge map    Save Deep learning-assisted diagnosis is an effective method to reduce missed and misdiagnosed fractures in the clinic. Currently, there are many research results on deep learning in fracture diagnosis, but there is a lack of review articles that summarizes and analyzes the current state of research in this field. Therefore, a summary of the existing literature in the field is presented in this paper. Firstly, fracture images and related datasets are introduced. Next, three deep learning-based fracture-assisted diagnosis methods are systematically described, and the deep learning models included in each method are compared. Then it classifies the methods according to different fracture types, and shows the deep learning methods in each type of fracture diagnosis. The analysis finds that the application and research of deep learning in the field of fracture diagnosis has made significant progress, and the model performance can be comparable to that of clinicians. However, models are highly influenced by the data set during training, and new models and techniques are more difficult to implement. Deep learning-assisted fracture diagnosis still has more room for development. Reference | Related Articles | Metrics Select Review of Research on Multimodal Retrieval JIN Tao, JIN Ran, HOU Tengda, YUAN Jie, GU Xiaozhe Computer Engineering and Applications    2024, 60 (5): 62-75.   DOI: 10.3778/j.issn.1002-8331.2305-0294 Abstract （99）      PDF（pc） （657KB）（96）       Knowledge map    Save With the increasing of multimodal data, multimodal retrieval technology has received a lot of attention. With the introduction of computer and big data technology in automobile, medical and other industries, a large amount of industry data itself are presented in a multi-modal form. With the rapid development of the industry, people’s demand for information is constantly increasing, and single modal data retrieval can no longer meet people’s demand for information. In order to solve these problems and meet the needs of data retrieval from one mode and other modes, this paper studies multi-modal retrieval methods through literature review, analyzes different research methods such as common subspace, deep learning and multi-modal Hash algorithm, and sorts out the multi-modal retrieval techniques proposed by researchers in recent years to solve these problems. Finally, the multimodal retrieval methods proposed in recent years are evaluated and compared according to the accuracy, efficiency and characteristics of the retrieval. This paper analyzes the challenges encountered in multimodal retrieval and looks forward to the future application prospects of multimodal retrieval. Reference | Related Articles | Metrics Select Survey on Video-Text Cross-Modal Retrieval CHEN Lei, XI Yimeng, LIU Libo Computer Engineering and Applications    2024, 60 (4): 1-20.   DOI: 10.3778/j.issn.1002-8331.2306-0382 Abstract （214）      PDF（pc） （3662KB）（210）       Knowledge map    Save Modalities define the specific forms in which data exist. The swift expansion of various modal data types has brought multimodal learning into the limelight. As a crucial subset of this field, cross-modal retrieval has achieved noteworthy advancements, particularly in integrating images and text. However, videos, as opposed to images, encapsulate a richer array of modal data and offer a more extensive spectrum of information. This richness aligns well with the growing user demand for comprehensive and adaptable information retrieval solutions. Consequently, video-text cross-modal retrieval has emerged as a burgeoning area of research in recent times. To thoroughly comprehend video-text cross-modal retrieval and its state-of-the-art developments, a methodical review and summarization of the existing representative methods is conducted. Initially, the focus is on analyzing current deep learning-based unidirectional and bidirectional video-text cross-modal retrieval methods. This analysis includes an in-depth exploration of seminal works within each category, highlighting their strengths and weaknesses. Subsequently, the discussion shifts to an experimental viewpoint, introducing benchmark datasets and evaluation metrics specific to video-text cross-modal retrieval. The performance of several standard methods in benchmark datasets is compared. Finally, the application prospects and future research challenges of video- text cross-modal retrieval are discussed. Reference | Related Articles | Metrics Select Survey of Neural Radiance Fields for Multi-View Synthesis Technologies MA Hansheng, ZHU Yuhua, LI Zhihui, YAN Lei, SI Yiyi, LIAN Yimeng, ZHANG Yuhan Computer Engineering and Applications    2024, 60 (4): 21-38.   DOI: 10.3778/j.issn.1002-8331.2303-0218 Abstract （104）      PDF（pc） （3529KB）（85）       Knowledge map    Save Rendering realistic virtual scenes from images has been a long-standing research goal in the fields of computer graphics and computer vision. NeRF (neural radiance fields) is an emerging method based on deep neural networks, which achieves realistic rendering by learning the radiance field of each point in the scene. By using neural radiance fields, not only realistic images but also realistic three-dimensional scenes can be generated, making it have a wide range of application prospects such as virtual reality, augmented reality and computer games. However, its basic model has shortcomings such as low training efficiency, poor generalization ability, insufficient interpretability, susceptible to lighting and material changes, inability to handle dynamic scenes, and other deficiencies that may result in suboptimal rendering results in certain situations. With the continuous popularity of this field, a large amount of research has been carried out, yielding impressive results in terms of efficiency and accuracy. In order to track the latest research in this field, this paper provides a review and summary of the key algorithms in recent years. This paper first outlines the background and principles of neural radiance fields, and briefly introduces the evaluation metrics and public datasets in this field. Then, a classification discussion is conducted on the key improvements to the model, mainly including: the optimization of basic NeRF model parameters, the improvement in rendering speed and inference ability, the enhancement of spatial representation and lighting ability, the improvement in camera pose and sparse view synthesis methods for static scene, and the development in dynamic scene modeling field. Subsequently, the speed and performance of various models are classified, compared and analyzed, and the main model evaluation indicators and open datasets in this field are briefly introduced. Finally, the future development trend of neural radiance field is prospected. Reference | Related Articles | Metrics Select Survey of Vision Transformer in Low-Level Computer Vision ZHU Kai, LI Li, ZHANG Tong, JIANG Sheng, BIE Yiming Computer Engineering and Applications    2024, 60 (4): 39-56.   DOI: 10.3778/j.issn.1002-8331.2304-0139 Abstract （159）      PDF（pc） （3488KB）（125）       Knowledge map    Save Transformer is a revolutionary neural network architecture initially designed for natural language processing. However, its outstanding performance and versatility have led to widespread applications in the field of computer vision. While there is a wealth of research and literature on Transformer applications in natural language processing, there remains a relative scarcity of specialized reviews focusing on low-level visual tasks. In light of this, this paper begins by providing a brief introduction to the principles of Transformer and analyzing several variants. Subsequently, the focus shifts to the application of Transformer in low-level visual tasks, specifically in the key areas of image restoration, image enhancement, and image generation. Through a detailed analysis of the performance of different models in these tasks, this paper explores the variations in their effectiveness on commonly used datasets. This includes achievements in restoring damaged images, improving image quality, and generating realistic images. Finally, this paper summarizes and forecasts the development trends of Transformer in the field of low-level visual tasks. It suggests directions for future research to further drive innovation and advancement in Transformer applications. The rapid progress in this field promises breakthroughs for computer vision and image processing, providing more powerful and efficient solutions for practical applications. Reference | Related Articles | Metrics Select Survey of Neural Machine Translation ZHANG Junjin, TIAN Yonghong, SONG Zheyu, HAO Yufeng Computer Engineering and Applications    2024, 60 (4): 57-74.   DOI: 10.3778/j.issn.1002-8331.2305-0102 Abstract （109）      PDF（pc） （3432KB）（74）       Knowledge map    Save Machine translation (MT) mainly studies how to translate the source language into the target language, which is of great significance for promoting the communication between nationalities. At present, neural machine translation (NMT) has become the mainstream MT method by translation speed and quality. In order to better sort out the context, this paper first introduces the history and methods of MT, compares and summarizes three main methods: rule-based machine translation, statistics-based machine translation and deep learning-based machine translation. Then NMT is introduced to explain its common types. Next, six main research fields of NMT are introduced, including multimodal MT, non-autoregressive MT, document-level MT, multilingual MT, data augmentation technology and preprocessing technique. Finally, the future of NMT is prospected from four aspects: low-resource languages, context-sensitive translation, unknown words and large models. This paper provides a systematic introduction to better understand the development status of NMT. Reference | Related Articles | Metrics Select Survey of Concept Drift Detection and Adaptation Methods MENG Fanxing, HAN Meng, LI Chunpeng, ZHANG Ruihua, HE Feifei Computer Engineering and Applications    2024, 60 (4): 75-88.   DOI: 10.3778/j.issn.1002-8331.2306-0406 Abstract （80）      PDF（pc） （3069KB）（49）       Knowledge map    Save With the rapid growth of data information, the characteristics of data stream are diversified, which put forward new requirements for the study of concept drift processing methods. The previous detection ideas based on performance degradation cannot cope with the non-instant availability environment of labels, which promotes the research of un-labeled detection methods. The research of ensemble and incremental learning promotes the progress of drift adaptation method. Firstly, the drift detection methods are analyzed in detail in two cases, namely whether the label is available and immediate or not, and based on the model composition, the drift adaptation methods are analyzed and explained with emphasis on the ensemble method. Secondly, the drift processing methods of different ideas are comprehensively summarized from the core ideas, comparison models, advantages and limitations, and the drift processing methods of different ideas are compared as a whole. Finally, the future research directions in this field are given, including noise discrimination, threshold setting and criterion diversification. Reference | Related Articles | Metrics Select Survey of Machine Learning for Database Parameter Tuning Techniques JIANG Lulu, GAO Jintao Computer Engineering and Applications    2024, 60 (3): 1-16.   DOI: 10.3778/j.issn.1002-8331.2304-0101 Abstract （100）      PDF（pc） （729KB）（100）       Knowledge map    Save Database parameter tuning techniques are facing great challenge under stringent performance requirements due to the huge data scale and complex application scenarios in big data. Traditional heuristic tuning methods or manual intervention methods are lack of the ability to handle various requirements. The development of machine learning provides a brand-new opportunity for database parameter tuning with its power in learning, reasoning, and planning. Through sufficient investigation, the evolution route of database parameter tuning technology based on machine learning is given. According to the thought of content-problem-resolution, this paper describes the traditional parameter tuning techniques and the parameter tuning techniques based on Bayesian optimization (BO) model and reinforcement learning (RL), and the future research directions are proposed as well. Hopefully, this paper can provide some valuable references for researchers in this field. Reference | Related Articles | Metrics Select Review of Conversational Machine Reading Comprehension for Knowledge Graph HU Juan, XI Xuefeng, CUI Zhiming Computer Engineering and Applications    2024, 60 (3): 17-28.   DOI: 10.3778/j.issn.1002-8331.2305-0081 Abstract （59）      PDF（pc） （599KB）（55）       Knowledge map    Save Conversational machine reading comprehension develops with the development of datasets. The purpose is to allow the machine to have multiple rounds of dialogue on the basis of understanding the content of the article. However, the existing model method cannot capture the most related historical information related to the current problem from the history of the dialogue. The inference capacity of the model is poor, and it is difficult to obtain the implicit information between the entities. The application of knowledge graph to reasoning question answering is a major research hotspot. Knowledge graph technology can infer the implicit relationship between entities, and when applied to reasoning questions and answering, it can improve the model’s reasoning question answering ability and improve the accuracy of prediction. In recent years, the widespread application of knowledge map reasoning technology has greatly promoted the development of knowledge map reasoning and answers. The conversational machine reading comprehension based on knowledge graph is summarized from three aspects. Firstly, it introduces the data sets in the field of session machine reading comprehension and some current model methods, and makes a brief analysis and comparison of the performance, advantages and disadvantages of the model. Then, it introduces the definition, architecture and four core technologies of the knowledge map, and briefly introduces the model methods of the three types of knowledge map reasoning quiz. Finally, the work is summarized, and according to the characteristics and knowledge of data sets and knowledge of the reading and comprehension of the session machine, the future research focus is prospected. Reference | Related Articles | Metrics Select Research Progress of Surface Electromyography Hand Motion Recognition LI Zhenjiang, WEI Dejian, FENG Yanyan, YU Fengfan, MA Yifan Computer Engineering and Applications    2024, 60 (3): 29-43.   DOI: 10.3778/j.issn.1002-8331.2305-0269 Abstract （64）      PDF（pc） （784KB）（73）       Knowledge map    Save Surface electromyography (sEMG) is a non-invasive method of measuring muscle activity, which contains rich information related to human motion and can be used for hand motion recognition. Hand motion recognition based on sEMG refers to the classification and recognition of hand motions by analyzing the sEMG signals of the hand muscles. Driven by the development of neural networks, sEMG has made great progress in the field of hand motion recognition. However, sEMG is faced with defects such as high noise and poor stability, which cannot be efficiently utilised, bringing great difficulties in acquiring high-precision hand movement recognition models and has hindered the translation and application of research results. This paper summarizes the research progress of sEMG hand motion recognition methods in detail. Firstly, public EMG datasets commonly used in the field of action recognition are introduced, and the self-test EMG set acquisition process is described. Then the existing sEMG hand motion recognition models are classified into three categories according to the different research methods: hand motion recognition based on machine learning, hand motion recognition based on deep learning and hand motion recognition based on hybrid network structure, and the related models are summarised and analyzed respectively, and suggestions are made for the shortcomings. Finally, the problems to be solved and the future development direction of hand action recognition research are prospected. Reference | Related Articles | Metrics Select Review of Deep Learning Methods Applied to Medical CT Super-Resolution TIAN Miaomiao, ZHI Lijia, ZHANG Shaomin, CHAO Daifu Computer Engineering and Applications    2024, 60 (3): 44-60.   DOI: 10.3778/j.issn.1002-8331.2303-0224 Abstract （128）      PDF（pc） （867KB）（90）       Knowledge map    Save Image super resolution (SR) is one of the important processing methods to improve image resolution in the field of computer vision, which has important research significance and application value in the field of medical image. High quality and high-resolution medical CT images are very important in the current clinical process. In recent years, the technology of medical CT image super-resolution reconstruction based on deep learning has made remarkable progress. This paper reviews the representative methods in this field and systematically reviews the development of medical CT image super-resolution reconstruction technology. Firstly, the basic theory of SR is introduced, and the commonly used evaluation indexes are given. Then, it focuses on the innovation and progress of super-resolution reconstruction of medical CT images based on deep learning, and makes a comprehensive comparative analysis of the main characteristics and performance of each method. Finally, the difficulties and challenges in the direction of medical CT image super-resolution reconstruction are discussed, and the future development trend is summarized and prospected, hoping to provide reference for related research. Reference | Related Articles | Metrics Select Review of Research on Information Security in Federated Learning DUAN Xinru, CHEN Guirong, CHEN Aiwang, CHEN Chen, JI Weifeng Computer Engineering and Applications    2024, 60 (3): 61-77.   DOI: 10.3778/j.issn.1002-8331.2303-0332 Abstract （57）      PDF（pc） （837KB）（68）       Knowledge map    Save As a new machine learning technology, federated learning allows participants to complete collaborative training and obtain global models through parameter interaction without sharing original data. It provides a new paradigm for breaking data silos and integrating data resources and has become a research hotspot in the field of artificial intelligence. However, federated learning still faces many security risks. This paper systematically analyzes and classifies the latest research results in the field of federated learning at home and abroad. Taking the training process of the federated learning model as a clue, it analyzes the security threats that may exist in the system during each process, studies the mechanism and characteristics of different security threats, and classifies them according to the degree of threat. Based on this, the paper studies the current advanced defense strategies. Finally, it discusses the main challenges and future development directions of federated learning in order to promote the safe landing and promotion of federated learning applications. Reference | Related Articles | Metrics Select Survey of Sentiment Analysis Algorithms Based on Multimodal Fusion GUO Xu, Mairidan Wushouer, Gulanbaier Tuerhong Computer Engineering and Applications    2024, 60 (2): 1-18.   DOI: 10.3778/j.issn.1002-8331.2305-0439 Abstract （379）      PDF（pc） （954KB）（269）       Knowledge map    Save Sentiment analysis is an emerging technology that aims to explore people’s attitudes toward entities and can be applied to various domains and scenarios, such as product evaluation analysis, public opinion analysis, mental health analysis and risk assessment. Traditional sentiment analysis models focus on text content, yet some special forms of expression, such as sarcasm and hyperbole, are difficult to detect through text. As technology continues to advance, people can now express their opinions and feelings through multiple channels such as audio, images and videos, so sentiment analysis is shifting to multimodality, which brings new opportunities for sentiment analysis. Multimodal sentiment analysis contains rich visual and auditory information in addition to textual information, and the implied sentiment polarity (positive, neutral, negative) can be inferred more accurately using fusion analysis. The main challenge of multimodal sentiment analysis is the integration of cross-modal sentiment information; therefore, this paper focuses on the framework and characteristics of different fusion methods and describes the popular fusion algorithms in recent years, and discusses the current multimodal sentiment analysis in small sample scenarios, in addition to the current development status, common datasets, feature extraction algorithms, application areas and challenges. It is expected that this review will help researchers understand the current state of research in the field of multimodal sentiment analysis and be inspired to develop more effective models. Reference | Related Articles | Metrics Select Overview of 360-Degree Video and Viewport Prediction LI Zhenhuai, ZHAN Yinwei Computer Engineering and Applications    2024, 60 (2): 19-31.   DOI: 10.3778/j.issn.1002-8331.2305-0056 Abstract （75）      PDF（pc） （723KB）（59）       Knowledge map    Save 360-degree video is one of the convenient media to obtain immersive virtual reality experience, which has attracted wide attention in recent years. Viewport prediction technology is an important means to alleviate the high bandwidth requirement of 360-degree video network, focusing viewport prediction technique, the basic concept of 360-degree video, background and 360-degree video streaming framework are firstly introduced, and the common sphere to plane projection methods and video codec standards are compared. The disadvantage of 360-degree video high network resource consumption is analyzed, and the important role of viewport prediction technology for video streaming is introduced. The 360-degree attention dataset is introduced, and the mainstream public datasets are summarized. The existing viewport prediction methods are divided into user history track based method and video content-based method, and a systematic review is carried out to sort out the development of viewport prediction methods, introduce the latest work of viewport prediction methods, compare the characteristics, advantages and limitations of different methods, and briefly introduce the 360-degree image salient detection. 360-degree salient detection is the focus of viewport prediction method based on video content. Finally, the problems faced by viewport prediction methods are analyzed, and the future development trend of 360-degree video related technologies including viewport prediction methods is forecasted. Reference | Related Articles | Metrics Select Cross-Chain Technology Development and Application Research LI Guangzhu, LI Leixiao, GAO Haoyu Computer Engineering and Applications    2024, 60 (2): 32-45.   DOI: 10.3778/j.issn.1002-8331.2304-0275 Abstract （72）      PDF（pc） （666KB）（63）       Knowledge map    Save With the continuous development and innovation of blockchain technology, many underlying technology platforms with different structures have emerged. Due to the mutual independence of blockchains, these platforms have formed their own value systems, hindering asset circulation and value transfer, and the phenomenon of “value islands”gradually appears. The introduction of cross-chain technology breaks the isolation of a single blockchain and establishes a collaborative value sharing network between blockchains, becoming a bridge and link for the overall expansion of blockchain, providing technical solutions to improve the interoperability and scalability of blockchain. Firstly, based on the existing research results, the background, significance and research status of cross-chain technology are elaborated. Secondly, according to the different implementation principles, cross-chain mechanisms are summarized into three types: external verification, native verification, and local verification, and are analyzed in detail. Thirdly, based on the development trends at home and abroad, cross-chain bridge application projects that adopt different cross-chain mechanisms are introduced, and existing cross-chain technologies and cross-chain bridge projects are analyzed and compared. Finally, the challenges faced by current cross-chain technology are summarized, and the development of future cross-chain technology is prospected. Reference | Related Articles | Metrics Select Review of Research on Electric Vehicle Routing Problem Based on Bibliometrics and Knowledge Mapping WANG Wenhao, YIN Lyujiang, YAN Caozheng, MOU Guangyuan Computer Engineering and Applications    2024, 60 (2): 46-62.   DOI: 10.3778/j.issn.1002-8331.2306-0065 Abstract （58）      PDF（pc） （953KB）（48）       Knowledge map    Save To reveal the research and development status in the field of electric vehicle routing problem, knowledge mining and analysis are conducted on the papers of electric vehicle routing problem in CNKI and Web of Science database from 1994 to 2022. The external characteristics and co-citation of the literature are analyzed based on the quantitative analysis of bibliometrics and the visualization of knowledge mapping. The research hotspots and hotspots evolution trend are sorted, and the research topics is summarized in the field. The knowledge domain of the electric vehicle routing problem is summarized, including research topics and application scenarios. The research topic consists of variant research, charging scheduling, and solution methods. The future development of electric vehicle routing problem in complex practical problems and efficient algorithms is prospected. This paper will provide a certain impetus for the deepening and internationalization of the research on the electric vehicle routing problem. Reference | Related Articles | Metrics Select Survey of Agricultural Knowledge Graph TANG Wentao, HU Zelin Computer Engineering and Applications    2024, 60 (2): 63-76.   DOI: 10.3778/j.issn.1002-8331.2305-0203 Abstract （189）      PDF（pc） （629KB）（148）       Knowledge map    Save Knowledge graphs are a key technology in the era of big data, specifically for knowledge engineering. Utilizing the powerful semantic understanding and knowledge organization capabilities of knowledge graphs, issues such as scattered and disordered agricultural knowledge, and insufficient coverage of knowledge in the construction of modern agriculture can be resolved. Firstly, considering the complexity and specialty of agricultural data, the construction methods and framework of agricultural knowledge graphs are introduced. Secondly, the current domestic and international research status of the four key technologies in the construction of agricultural knowledge graphs-ontology construction, knowledge extraction, knowledge fusion, and knowledge reasoning are reviewed. Furthermore, the systematic applications of agricultural knowledge graphs in decision support, intelligent question-answering systems, and recommendation systems are sorted out. Lastly, several specific instances of agricultural knowledge graphs are presented. Based on the current status of research on agricultural knowledge graphs, prospects for its future research directions are offered. Reference | Related Articles | Metrics Select Review of SLAM Based on Lidar LIU Mingzhe, XU Guanghui, TANG Tang, QIAN Xiaojian, GENG Ming Computer Engineering and Applications    2024, 60 (1): 1-14.   DOI: 10.3778/j.issn.1002-8331.2308-0455 Abstract （409）      PDF（pc） （854KB）（258）       Knowledge map    Save Simultaneous localization and mapping (SLAM) is a crucial technology for autonomous mobile robots and autonomous driving systems, with a laser scanner (also known as lidar) playing a vital role as a supporting sensor for SLAM algorithms. This article provides a comprehensive review of lidar-based SLAM algorithms. Firstly, it introduces the overall framework of lidar-based SLAM, providing detailed explanations of the functions of the front-end odometry, back-end optimization, loop closure detection, and map building modules, along with a summary of the algorithms used. Secondly, it presents descriptions and summaries of representative open-source algorithms in a sequential order of 2D to 3D and single-sensor to multi-sensor fusion. Additionally, it discusses commonly used open-source datasets, precision evaluation metrics, and evaluation tools. Lastly, it offers an outlook on the development trends of lidar-based SLAM technology from four dimensions: deep learning, multi-sensor fusion, multi-robot collaboration, and robustness research. Reference | Related Articles | Metrics Select Survey of Chinese Named Entity Recognition Research ZHAO Jigui, QIAN Yurong, WANG Kui, HOU Shuxiang, CHEN Jiaying Computer Engineering and Applications    2024, 60 (1): 15-27.   DOI: 10.3778/j.issn.1002-8331.2304-0398 Abstract （144）      PDF（pc） （606KB）（100）       Knowledge map    Save Named entity recognition (NER) is one of the most fundamental tasks in natural language processing, and its main content is to identify the entity types and boundaries with specific meanings in natural language text. However, the data samples of Chinese named entity recognition (CNER) have problems such as blurred word boundaries, semantic diversity, blurred morphological features and small Chinese corpus content, which make it difficult to improve the performance of Chinese NER. In this paper, firstly, the dataset, annotation scheme and evaluation index of CNER are introduced. Secondly, according to the research process of CNER, CNER methods are classified into three categories: rule-based methods, statistical-based methods and deep learning-based methods, and the main models of CNER based on deep learning in the past five years are summarized. Finally, the research trends of CNER are discussed to provide some reference for the proposal of new methods and future research directions. Reference | Related Articles | Metrics Select Infrared and Visible Image Fusion Under Photoelectric Loads LIU Shuangli, HUANG Xueli, LIU Lei, XIE Yu, ZHANG Jinbao, YANG Jiangnan Computer Engineering and Applications    2024, 60 (1): 28-39.   DOI: 10.3778/j.issn.1002-8331.2305-0291 Abstract （60）      PDF（pc） （586KB）（56）       Knowledge map    Save With the application of airborne photoelectric platform in military and civil fields, the enhancement and fusion technology of infrared and visible image has gradually become a hot research direction. Infrared equipment mainly relies on the thermal radiation of the object itself for imaging, which is suitable for harsh natural environment and secret places and other special scenes, but it has some shortcomings such as poor imaging quality, low image contrast, not rich texture details and so on. The texture details and contrast of visible images are more suitable for human visual perception, but the imaging effect of visible images is poor under conditions such as smoke and night, so it is not suitable for hidden places. Therefore, the researchers proposed a complementary image fusion method with both visible image edge and detail information and infrared thermal radiation target information. Based on the characteristics of infrared and visible images, this paper summarizes the traditional infrared and visible image fusion methods applicable to airborne platforms, and gives the improvement ideas of dim small targets, texture information loss and computational efficiency in the existing infrared and visible image fusion algorithms. The future development of traditional infrared and visible image fusion technology under the background of photoelectric load is prospected. Reference | Related Articles | Metrics Select Review of Smoking Detection Methods for Computer Vision HE Jiabin, LI Leixiao, LIN Hao, XU Guoxin Computer Engineering and Applications    2024, 60 (1): 40-56.   DOI: 10.3778/j.issn.1002-8331.2305-0154 Abstract （98）      PDF（pc） （978KB）（96）       Knowledge map    Save Smoking in public places seriously harms people’s health and even life and property safety, so real-time and efficient smoking detection is of great significance. At present, smoking detection based on computer vision has gradually become the mainstream method with the advantages of high efficiency and high precision. On the basis of a brief overview of non-computer vision smoking detection methods, three kinds of detection methods based on computer vision are summarized. Firstly, the extraction methods of smoke features such as color, appearance and movement are discussed. Secondly, two methods of extracting cigarette target based on single step and multi-step target detection are introduced. Finally, different types of smoking action feature extraction methods are discussed from the perspectives of artificial feature construction and deep learning feature extraction. The above methods are analyzed and summarized, and the future research direction is prospected. Reference | Related Articles | Metrics Select Survey on Identity Management in Blockchain Systems LI Fujuan, MA Zhuo, WANG Qun Computer Engineering and Applications    2024, 60 (1): 57-73.   DOI: 10.3778/j.issn.1002-8331.2302-0189 Abstract （87）      PDF（pc） （832KB）（55）       Knowledge map    Save Blockchain technology integrates various technologies and theories such as P2P networks, consensus mechanisms, cryptography, game theory and economics as an innovative application. It challenges the constraints imposed by centralized mechanisms in traditional systems with its characteristic of decentralization. It achieves trustworthiness, traceability and tamper resistance in transactions within an open network environment without the need for a trusted third party through consensus mechanisms. The new computing paradigm and trust mechanism formed by blockchain technology contribute to the transformation of management models. The transparency of the ledger and the multi-party consensus mechanism in blockchain technology pose challenges to identity management in blockchain systems at the same time, where transactions serve as the smallest data unit. This paper provides insights into understanding blockchain identity identification and authentication methods and strengthens information discovery and value extraction in different application scenarios by reviewing the important research achievements in blockchain system identity management technology. This paper explains the characteristics of the UTXO model and the account model by discussing the characteristics of identity management in traditional systems and blockchain systems and clarifying the main contents of blockchain identity management. It further analyzes three types of blockchain system identity identification mechanisms: public key transformation, digital certificates and decentralized digital identity. It also examines three types of blockchain authentication methods: anonymous authentication, real-name authentication and controllable anonymous authentication. It finally offers prospects for the future development of blockchain identity management technology. Reference | Related Articles | Metrics page Page 1 of 8 Total 297 records First page Prev page Next page Last page