Most Download articles

    Published in last 1 year| In last 2 years| In last 3 years| All| Most Downloaded in Recent Month | Most Downloaded in Recent Year|

    Published in last 1 year
    Please wait a minute...
    For Selected: Toggle Thumbnails
    Review of Research on Generative Adversarial Networks and Its Application
    WEI Fuqiang, Gulanbaier Tuerhong, Mairidan Wushouer
    Computer Engineering and Applications    2021, 57 (19): 18-31.   DOI: 10.3778/j.issn.1002-8331.2104-0248
    Abstract395)      PDF(pc) (1078KB)(998)       Save

    The theoretical research and applications of generative adversarial networks have been continuously successful and have become one of the current hot spots of research in the field of deep learning. This paper provides a systematic review of the theory of generative adversarial networks and their applications in terms of types of models, evaluation criteria and theoretical research progress; analyzing the strengths and weaknesses of generative models with explicit and implicit density-based, respectively; summarizing the evaluation criteria of generative adversarial networks, interpreting the relationship between the criteria, and introduces the research progress of the generative adversarial network in image generation from the application level, that is, through the image conversion, image generation, image restoration, video generation, text generation and image super-resolution applications; analyzing the theoretical research progress of generative adversarial networks from the perspectives of interpretability, controllability, stability and model evaluation methods. Finally, the paper discusses the challenges of studying generative adversarial networks and looks forward to the possible future directions of development.

    Reference | Related Articles | Metrics
    Review of Text Sentiment Analysis Methods
    WANG Ting, YANG Wenzhong
    Computer Engineering and Applications    2021, 57 (12): 11-24.   DOI: 10.3778/j.issn.1002-8331.2101-0022
    Abstract445)      PDF(pc) (906KB)(511)       Save

    Text sentiment analysis is an important branch of natural language processing, which is widely used in public opinion analysis and content recommendation. It is also a hot topic in recent years. According to different methods used, it is divided into sentiment analysis based on emotional dictionary, sentiment analysis based on traditional machine learning, and sentiment analysis based on deep learning. Through comparing these three methods, the research results are analyzed, and the paper summarizes the advantages and disadvantages of different methods, introduces the related data sets and evaluation index, and application scenario, analysis of emotional subtasks is simple summarized. The future research trend and application field of sentiment analysis problem are found. Certain help and guidance are provided for the researchers in the related areas.

    Related Articles | Metrics
    Research Progress of Medical Image Registration Technology Based on Deep Learning
    GUO Yanfen, CUI Zhe, YANG Zhipeng, PENG Jing, HU Jinrong
    Computer Engineering and Applications    2021, 57 (15): 1-8.   DOI: 10.3778/j.issn.1002-8331.2101-0281
    Abstract501)      PDF(pc) (681KB)(465)       Save

    Medical image registration technology has a wide range of application values for lesion detection, clinical diagnosis, surgical planning, and efficacy evaluation. This paper systematically summarizes the registration algorithm based on deep learning, and analyzes the advantages and limitations of various methods from deep iteration, full supervision, weak supervision to unsupervised learning. In general, unsupervised learning has become the mainstream direction of medical image registration research, because it does not rely on golden standards and uses an end-to-end network to save time. Meanwhile, compared with other methods, unsupervised learning can achieve higher accuracy and spends shorter time. However, medical image registration methods based on unsupervised learning also face some research difficulties and challenges in terms of interpretability, cross-modal diversity, and repeatable scalability in the field of medical images, which points out the research direction for achieving more accurate medical image registration methods in the future.

    Related Articles | Metrics
    Review of Application of Transfer Learning in Medical Image Field
    GAO Shuang, XU Qiaozhi
    Computer Engineering and Applications    2021, 57 (24): 39-50.   DOI: 10.3778/j.issn.1002-8331.2107-0300
    Abstract292)      PDF(pc) (896KB)(435)       Save

    Deep learning technology has developed rapidly and achieved significant results in the field of medical image treatment. However, due to the small number of medical image samples and difficult annotation, the effect of deep learning is far from reaching the expectation. In recent years, using transfer learning method to alleviate the problem of insufficient medical image samples and improve the effect of deep learning technology in the field of medical image has become one of the research hotspots. This paper first introduces the basic concepts, types, common strategies and models of transfer learning methods, then combs and summarizes the representative related research in the field of medical images according to the types of transfer learning methods, and finally summarizes and prospects the future development of this field.

    Reference | Related Articles | Metrics
    Review of Attention Mechanism in Convolutional Neural Networks
    ZHANG Chenjia, ZHU Lei, YU Lu
    Computer Engineering and Applications    2021, 57 (20): 64-72.   DOI: 10.3778/j.issn.1002-8331.2105-0135
    Abstract357)      PDF(pc) (973KB)(420)       Save

    Attention mechanism is widely used in deep learning tasks because of its excellent effect and plug and play convenience. This paper mainly focuses on convolution neural network, introduces various mainstream methods in the development process of convolution network attention mechanism, extracts and summarizes its core idea and implementation process, realizes each attention mechanism method, and makes comparative experiments and results analysis on the measured data of the same type of emitter equipment. According to the main ideas and experimental results, the research status and future development direction of attention mechanism in convolutional networks are summarized.

    Reference | Related Articles | Metrics
    Overview of Chinese Domain Named Entity Recognition
    JIAO Kainan, LI Xin, ZHU Rongchen
    Computer Engineering and Applications    2021, 57 (16): 1-15.   DOI: 10.3778/j.issn.1002-8331.2103-0127
    Abstract457)      PDF(pc) (928KB)(417)       Save

    Named Entity Recognition(NER), as a classic research topic in the field of natural language processing, is the basic technology of intelligent question answering, knowledge graph and other tasks. Domain Named Entity Recognition(DNER) is the domain-specific NER scheme. Drived by deep learning technology, Chinese DNER has made a breakthrough. Firstly, this paper summarizes the research framework of Chinese DNER, and reviews the existing research results from four aspects:the determination of domain data sources, the establishment of domain entity types and specifications, the annotation of domain data sets, and the evaluation metrics of Chinese DNER. Then, this paper summarizes the current common technology framework of Chinese DNER, introduces the pattern matching method based on dictionaries and rules, statistical machine learning method, deep learning method, multi-party fusion deep learning method, and focuses on the analysis of Chinese DNER method based on word vector representation and deep learning. Finally, the typical application scenarios of Chinese DNER are discussed, and the future development direction is prospected.

    Related Articles | Metrics
    Robot Dynamic Path Planning Based on Improved A* and DWA Algorithm
    LIU Jianjuan, XUE Liqi, ZHANG Huijuan, LIU Zhongpu
    Computer Engineering and Applications    2021, 57 (15): 73-81.   DOI: 10.3778/j.issn.1002-8331.2103-0525
    Abstract244)      PDF(pc) (1452KB)(401)       Save

    Traditional A* algorithm is one of the commonly used algorithms for global path planning of mobile robot, but the algorithm has low search efficiency, many turning points in planning path, and can’t achieve dynamic path planning in the face of random dynamic obstacles in complex environment. To solve these problems, the improved A* algorithm and DWA algorithm are integrated on the basis of global optimization. The obstacle information in the environment is quantified, and the weight of heuristic function of A* algorithm is adjusted according to the information to improve the efficiency and flexibility of the algorithm. Based on the Floyd algorithm, the optimization algorithm of path nodes is designed, which can delete redundant nodes, reduce turning points and improve the path smoothness. The dynamic window evaluation function of DWA algorithm is designed based on the global optimal, which is used to distinguish known obstacles from unknown dynamic and static obstacles, and the key points of the improved A* algorithm planning path are extracted as the temporary target points of DWA algorithm. On the basis of the global optimal, the fusion of the improved A* algorithm and DWA algorithm is realized. The experimental results show that, in the complex environment, the fusion algorithm can not only ensure the global optimal path planning, but also effectively avoid the dynamic and static obstacles in the environment, and realize the dynamic path planning in the complex environment.

    Related Articles | Metrics
    Overview of Visual Multi-object Tracking Algorithms with Deep Learning
    ZHANG Yao, LU Huanzhang, ZHANG Luping, HU Moufa
    Computer Engineering and Applications    2021, 57 (13): 55-66.   DOI: 10.3778/j.issn.1002-8331.2102-0260
    Abstract296)      PDF(pc) (931KB)(396)       Save

    Visual multi-object tracking is a hot issue in the field of computer vision. However, the uncertainty of the number of targets in the scene, the mutual occlusion between targets, and the difficulties of discrimination between target features has led to slow progress in the real-world application of visual multi-target tracking. In recent years, with the continuous in-depth research of visual intelligent processing, a variety of deep learning visual multi-object tracking algorithms have emerged. Based on the analysis of the challenges and difficulties faced by visual multi-object tracking, the algorithm is divided into Detection-Based Tracking(DBT) and Joint Detection Tracking(JDT) two categories and six sub-categories class, and studied about its advantages and disadvantages. The analysis shows that the DBT algorithm has a simple structure, but the correlation of each sub-step of the algorithm is not high. The JDT algorithm integrates multi-module joint learning and is dominant in multiple tracking evaluation indicators. The feature extraction module is the key to solve the target occlusion in the DBT algorithm with the expense of the speed of the algorithm, and the JDT algorithm is more dependent on the detection module. At present, multi-object tracking is generally developed from DBT-type algorithms to JDT, achieving a balance between algorithm accuracy and speed in stages. The future development direction of the multi-object tracking algorithm in terms of datasets, sub-modules, and specific scenarios is proposed.

    Related Articles | Metrics
    Progress on Deep Reinforcement Learning in Path Planning
    ZHANG Rongxia, WU Changxu, SUN Tongchao, ZHAO Zengshun
    Computer Engineering and Applications    2021, 57 (19): 44-56.   DOI: 10.3778/j.issn.1002-8331.2104-0369
    Abstract551)      PDF(pc) (1134KB)(388)       Save

    The purpose of path planning is to allow the robot to avoid obstacles and quickly plan the shortest path during the movement. Having analyzed the advantages and disadvantages of the reinforcement learning based path planning algorithm, the paper derives a typical deep reinforcement learning, Deep Q-learning Network(DQN) algorithm that can perform excellent path planning in a complex dynamic environment. Firstly, the basic principles and limitations of the DQN algorithm are analyzed in depth, and the advantages and disadvantages of various DQN variant algorithms are compared from four aspects:the training algorithm, the neural network structure, the learning mechanism and AC(Actor-Critic) framework. The paper puts forward the current challenges and problems to be solved in the path planning method based on deep reinforcement learning. The future development directions are proposed, which can provide reference for the development of intelligent path planning and autonomous driving.

    Reference | Related Articles | Metrics
    Review of Neural Style Transfer Models
    TANG Renwei, LIU Qihe, TAN Hao
    Computer Engineering and Applications    2021, 57 (19): 32-43.   DOI: 10.3778/j.issn.1002-8331.2105-0296
    Abstract362)      PDF(pc) (1078KB)(380)       Save

    Neural Style Transfer(NST) technique is used to simulate different art styles of images and videos, which is a popular topic in computer vision. This paper aims to provide a comprehensive overview of the current progress towards NST. Firstly, the paper reviews the Non-Photorealistic Rendering(NPR) technique and traditional texture transfer. Then, the paper categorizes current major NST methods and gives a detailed description of these methods along with their subsequent improvements. After that, it discusses various applications of NST and presents several evaluation methods which compares different style transfer models both qualitatively and quantitatively. In the end, it summarizes the existing problems and provides some future research directions for NST.

    Reference | Related Articles | Metrics
    Research Progress of Transformer Based on Computer Vision
    LIU Wenting, LU Xinming
    Computer Engineering and Applications    2022, 58 (6): 1-16.   DOI: 10.3778/j.issn.1002-8331.2106-0442
    Abstract360)      PDF(pc) (1089KB)(342)       Save
    Transformer is a deep neural network based on the self-attention mechanism and parallel processing data. In recent years, Transformer-based models have emerged as an important area of research for computer vision tasks. Aiming at the current blanks in domestic review articles based on Transformer, this paper covers its application in computer vision. This paper reviews the basic principles of the Transformer model, mainly focuses on the application of seven visual tasks such as image classification, object detection and segmentation, and analyzes Transformer-based models with significant effects. Finally, this paper summarizes the challenges and future development trends of the Transformer model in computer vision.
    Reference | Related Articles | Metrics
    Semantic Similarity Calculation Based on Transformer Encoder
    QIAO Weitao, HUANG Haiyan, WANG Shan
    Computer Engineering and Applications    2021, 57 (14): 158-163.   DOI: 10.3778/j.issn.1002-8331.2004-0096
    Abstract104)      PDF(pc) (1087KB)(326)       Save

    The calculation of semantic similarity aims to calculate the similarity between texts at the semantic level, which is an important task in natural language processing. Aiming at the problem that the existing calculation methods cannot fully represent the semantic features of sentences, the model TEAM based on the Transformer encoder is proposed. It can extract the semantic information in sentences by using the contextual semantic encoding ability of the Transformer model. In addition, the interactive attention mechanism is introduced. When encoding two sentences the interactive attention mechanism is used to extract similar features between the two sentences, making the model better at capturing important semantic information within the sentence and improving the model’s understanding of semantics and generalization capabilities. The experimental results show that the model can improve the accuracy of the results on the semantic similarity calculation task of English and Chinese, and exhibit better results than existing methods.

    Related Articles | Metrics
    Research Progress of Object Detection Based on Weakly Supervised Learning
    YANG Hui, QUAN Jichuan, LIANG Xinyu, WANG Zhongwei
    Computer Engineering and Applications    2021, 57 (16): 40-49.   DOI: 10.3778/j.issn.1002-8331.2103-0306
    Abstract250)      PDF(pc) (633KB)(306)       Save

    With the continuous development of Convolutional Neural Network(CNN), as the most basic technology in computer vision, object detection has made remarkable progress. Firstly, the current situation that the strong supervised object detection algorithm requires high precision for labeling datasets is introduced. Secondly, the object detection algorithm based on weakly supervised learning is studied. The algorithm is classified into four categories according to different feature processing methods, and the advantages and disadvantages of each algorithm are analyzed and compared. Thirdly, the detection accuracy of all kinds of object detection algorithms based on weakly supervised learning is compared through experiments. At the same time, it is compared with the mainstream strong supervised object detection algorithms. Finally, the future research hotspots of object detection algorithms based on weakly supervised learning are prospected.

    Related Articles | Metrics
    Remote Sensing Military Target Detection Algorithm Based on Lightweight YOLOv3
    QIN Weiwei, SONG Tainian, LIU Jieyu, WANG Hongwei, LIANG Zhuo
    Computer Engineering and Applications    2021, 57 (21): 263-269.   DOI: 10.3778/j.issn.1002-8331.2106-0026
    Abstract133)      PDF(pc) (14418KB)(298)       Save

    In the process of intelligent missile penetration, detecting enemy anti-missile positions from massive remote sensing image data has great application value. Due to the limited computing power of the missile-borne deployment environment, this paper designs a remote sensing target detection algorithm that takes into account lightweight, detection accuracy and detection speed. A typical remote sensing military target data set is produced, and the data set is clustered and analyzed by the K-means algorithm. The MobileNetV2 network is used to replace the backbone network of the YOLOv3 algorithm to ensure the lightweight and detection speed of the network. A lightweight and efficient channel coordinated attention module and a target rotation invariance detection module suitable for remote sensing target characteristics are proposed, and they are embedded in the detection algorithm to improve the detection accuracy on the basis of network lightweight. Experimental results show that the accuracy rate of the algorithm in this paper reaches 97.8%, an increase of 6.7 percentage points, the recall rate reaches 95.7%, an increase of 3.9 percentage points, the average detection accuracy reaches 95.2%, an increase of 4.4 percentage points, and the detection speed reached 34.19 images per, and the network size is only 17.5?MB. The results show that the algorithm in this paper can meet the comprehensive requirements of intelligent missile penetration.

    Reference | Related Articles | Metrics
    Highway Vehicle Object Detection Based on Improved YOLOv4 Method
    WANG Yingxuan, SONG Huansheng, LIANG Haoxiang, YU Xiaoyu, YUN Xu
    Computer Engineering and Applications    2021, 57 (13): 218-226.   DOI: 10.3778/j.issn.1002-8331.2012-0581
    Abstract186)      PDF(pc) (1746KB)(292)       Save

    Aiming at the problem of vehicle object detection in highway scenes, an improved YOLOv4 network is proposed to detect vehicles in traffic scenes. Firstly, a multi-weather, multi-period, multi-scene vehicle dataset is proposed, and vehicle detection models are obtained based on the datasets. Secondly, a multi-label detection method is proposed, and a constraint relationship between multiple labels is established to obtain more complete vehicle information. Finally, an image stitching detection method is proposed, which connects multiple images through the stitching layer for vehicle detection, so as to improve the running speed of the network. Experimental results show that the diversified dataset improves the accuracy of vehicle detection, reduces the false detection and missed detection of vehicle objects, and the improved network structure greatly improves the detection speed. The above methods can provide references for vehicle detection and practical applications in highway scenes.

    Related Articles | Metrics
    Overview on Reinforcement Learning of Multi-agent Game
    WANG Jun, CAO Lei, CHEN Xiliang, LAI Jun, ZHANG Legui
    Computer Engineering and Applications    2021, 57 (21): 1-13.   DOI: 10.3778/j.issn.1002-8331.2104-0432
    Abstract224)      PDF(pc) (779KB)(290)       Save

    The use of deep reinforcement learning to solve single-agent tasks has made breakthrough progress. Since the complexity of multi-agent systems, common algorithms cannot solve the main difficulties. At the same time, due to the increase in the number of agents, taking the expected value of maximizing the cumulative return of a single agent as the learning goal often fails to converge and some special convergence points do not satisfy the rationality of the strategy. For practical problems that there is no optimal solution, the reinforcement learning algorithm is even more helpless. The introduction of game theory into reinforcement learning can solve the interrelationship of agents very well and explain the rationality of the strategy corresponding to the convergence point. More importantly, it can use the equilibrium solution to replace the optimal solution in order to obtain a relatively effective strategy. Therefore, this article investigates the reinforcement learning algorithms that have emerged in recent years from the perspective of game theory, summarizes the important and difficult points of current game reinforcement learning algorithms and gives several breakthrough directions that may solve the above-mentioned difficulties.

    Reference | Related Articles | Metrics
    Overview of Image Super-Resolution Algorithms
    SUN Jingyang, CHEN Fengdong, HAN Yueyue, WU Yuwen, GAN Yu, LIU Guodong
    Computer Engineering and Applications    2021, 57 (17): 1-9.   DOI: 10.3778/j.issn.1002-8331.2103-0556
    Abstract300)      PDF(pc) (1343KB)(289)       Save

    Image super-resolution reconstruction aims to recover high-resolution and clear images from low-resolution images. This article first explains the idea of typical image super-resolution reconstruction methods, and then reviews typical and latest image super-resolution reconstruction algorithms based on deep learning from the dimensions of up-sampling position and up-sampling method, learning strategy, loss function, etc. It analyzes the latest development status, and looks forward to the future development trend.

    Related Articles | Metrics
    Computer Engineering and Applications    2021, 57 (24): 0-0.  
    Abstract203)      PDF(pc) (1168KB)(280)       Save
    Related Articles | Metrics
    Small Sample DGA Malicious Domain Names Detection Method Based on Transfer Learning
    GU Zhaojun, YANG Wenjin, ZHOU Jingxian
    Computer Engineering and Applications    2021, 57 (14): 103-109.   DOI: 10.3778/j.issn.1002-8331.2004-0209
    Abstract98)      PDF(pc) (1277KB)(275)       Save

    The Domain name Generation Algorithm(DGA) is easy to evolve, and some category of samples are difficult to obtain, which makes the detection of malicious domain names using traditional machine learning models inaccurate. A small sample DGA malicious domain name detection model based on transfer learning and multi-core CNN is proposed. The model maps the domain name into the vector space, and then uses the DGA with sufficient samples for pre-training, and migrates the pre-trained parameters to the small sample detection model. Finally, the multi-core CNN classification model of small data DGA extracts the characters of domain according to pronunciation habits, and determines whether the domain is a DGA domain. By comparison, the small sample classification model without knowledge transfer has only 11 types of domain names with an accuracy of more than 92%. The classification results of the multi-core CNN model after transfer learning has 20 types of DGA with an accuracy more than 92% and the 11 types more than 97%. Through knowledge transfer, the classification effect of the model trained by insufficient DGA data can be close to the model trained by sufficient data.

    Related Articles | Metrics
    Adaptive Salp Swarm Algorithm with Golden Sine Algorithm and Hybrid Mutation
    ZHOU Xin, ZOU Hai
    Computer Engineering and Applications    2021, 57 (12): 75-85.   DOI: 10.3778/j.issn.1002-8331.2006-0011
    Abstract142)      PDF(pc) (905KB)(274)       Save

    In order to solve the problems of slow convergence speed, low accuracy and easy to fall into local optimum solution of the standard salp swarm algorithm, an adaptive salp swarm algorithm based on golden sine algorithm and hybrid mutation is proposed. The adaptive weight factor is introduced to strengthen the leading role of elite individuals and improve the convergence speed and accuracy of the basic salp swarm algorithm. The golden sine algorithm is used to optimize the position update mode of the leader and improve the global exploration and local exploitation capacity of the algorithm. The hybrid neighborhood centroid opposition-based learning with cauchy mutation strategy is introduced to disturb the best individual’s position and improve the ability of the algorithm to jump out of local optimum. The optimization performance of the proposed algorithm is evaluated by a sets of simulation experiments on 12 benchmark functions, the experimental results show that the proposed algorithm can significantly improve the optimum speed and accuracy, besides, it has a strong ability to jump out of local optimum.

    Related Articles | Metrics
    Research on Edge Computing Security of Railway 5G Mobile Communication System
    LIU Jiajia, WU Hao, LI Panpan
    Computer Engineering and Applications    2021, 57 (12): 1-10.   DOI: 10.3778/j.issn.1002-8331.2102-0052
    Abstract276)      PDF(pc) (688KB)(271)       Save

    Edge computing, as a key technology of the intelligent railway 5G network, it sinks data caching capabilities, traffic forwarding capabilities and application service capabilities to the edge of the network, effectively meets the low latency, large bandwidth, and massive connection requirements of intelligent railways to support intelligent rail transit application. However, due to it changes in physical location, business types and other aspects, and the complex external environment of the railway scene, highly dynamic, and low credibility, the edge nodes of the intelligent railway business are faced with new security challenges. Combined with the current research status of 5G edge computing security, the security threats faced by railway 5G edge computing are analyzed based on the analysis of the four aspects of terminal, edge network, edge node and edge application. On the basis of detailed security requirements and challenges, and standard progress, the research methods and evaluation indicators are summarized that can be applied to railway MEC safety. Combined with the characteristics of railway 5G edge computing, this paper proposes railway MEC end-to-end safety service solutions and the development direction of future intelligent railway MEC security research.

    Related Articles | Metrics
    Survey on Zero-Shot Learning
    WANG Zeshen,YANG Yun,XIANG Hongxin, LIU Qing
    Computer Engineering and Applications    2021, 57 (19): 1-17.   DOI: 10.3778/j.issn.1002-8331.2106-0133
    Abstract270)      PDF(pc) (1267KB)(270)       Save

    Although there have been well developed in zero-shot learning since the development of deep learning, in the aspect of the application, zero-shot learning did not have a good system to order it. This paper overviews theoretical systems of zero-shot learning, typical models, application systems, present challenges and future research directions. Firstly, it introduces the theoretical systems from definition of zero-shot learning, essential problems, and commonly used data sets. Secondly, some typical models of zero-shot learning are described in chronological order. Thirdly, it presents the application systems about of zero-shot learning from the three dimensions, such as words, images and videos. Finally, the paper analyzes the challenges and future research directions in zero-shot learning.

    Reference | Related Articles | Metrics
    Computer Engineering and Applications    2022, 58 (9): 0-0.  
    Abstract57)      PDF(pc) (38025KB)(266)       Save
    Related Articles | Metrics
    Text Classification Method Based on LSTM-Attention and CNN Hybrid Model
    TENG Jinbao, KONG Weiwei, TIAN Qiaoxin, WANG Zhaoqian
    Computer Engineering and Applications    2021, 57 (14): 126-133.   DOI: 10.3778/j.issn.1002-8331.2011-0037
    Abstract205)      PDF(pc) (780KB)(262)       Save

    For the problem that traditional Long Short-Term Memory(LSTM) and Convolution Neural Network(CNN) cannot reflect the importance of each word in the text when extracting features, a text classification method based on the hybrid model of LSTM-Attention and CNN is proposed. Firstly, CNN is used to extract the local information of the text and then integrate the semantics of the whole text. Secondly, LSTM is used to extract text context features. After LSTM, Attention mechanism is added to extract the Attention score of output information. Finally, the output of LSTM-Attention is fused with the output of CNN, so as to realize the effective extraction of text features and focus Attention on important words. The experimental results on three open data sets show that the proposed model is more effective than LSTM, CNN and their improved models, and can effectively improve the effect of text classification.

    Related Articles | Metrics
    SDN Routing Optimization Algorithm Based on Reinforcement Learning
    CHE Xiangbei, KANG Wenqian, OUYANG Yuhong, YANG Kehan, LI Jian
    Computer Engineering and Applications    2021, 57 (12): 93-98.   DOI: 10.3778/j.issn.1002-8331.2003-0423
    Abstract224)      PDF(pc) (869KB)(257)       Save

    Aiming at the network routing optimization in SDN controller, a routing optimization algorithm is designed based on the PPO model in reinforcement learning. The algorithm can adjust the reward function for different optimization goals to dynamically update the routing strategy, and this algorithm does not depend on any specific network state and has very good generalization performance. Because of adopting the strategy method in reinforcement learning, the control of routing strategy is more elaborate than various Q-learning-based algorithms. Based on Omnet++ simulation software, the performance of the algorithm is evaluated through experiments. Compared with the traditional shortest path routing algorithm, the average delay and end-to-end maximum delay of this routing optimization algorithm on the Sprint structure network are reduced by 29.3% and 17.4%, respectively and throughput rate is increased by 31.77%. The experimental result shows that this PPO-based SDN routing control algorithm not only has good convergence, but also has better performance and stability than the shortest path routing algorithm and the Q-learning based QAR routing algorithm.

    Related Articles | Metrics
    Survey of Multimodal Data Fusion
    REN Zeyu, WANG Zhenchao, KE Zunwang, LI Zhe, Wushour·Silamu
    Computer Engineering and Applications    2021, 57 (18): 49-64.   DOI: 10.3778/j.issn.1002-8331.2104-0237
    Abstract221)      PDF(pc) (1214KB)(257)       Save

    With the rapid development of information technology, information exists in various forms and sources. Different forms of existence or information sources can be referred to as one modal, and data composed of two or more modalities is called multi-modal data. Multi-modal data fusion is responsible for effectively integrating the information of multiple modalities, absorbing the advantages of different modalities, and completing the integration of information. Natural phenomena have very rich characteristics, and it is difficult for a single mode to provide complete information about a certain phenomenon. Faced with the fusion requirements of maintaining the diversity and completeness of the modal information after fusion, maximizing the advantages of each modal, and reducing the information loss caused by the fusion process, how to integrate the information of each modal has become a new challenge that exists in many fields. This paper briefly describes common multimodal fusion methods and fusion architectures, summarizes three common fusion models, and briefly analyzes the advantages and disadvantages of the three architectures of collaboration, joint, and codec, as well as specific fusion methods such as multi-core learning and image models. In the application of multi-modality, it analyzes and summarizes multi-modal video clip retrieval, comprehensive multi-modal information generation content summary, multi-modal sentiment analysis, and multi-modal man-machine dialogue system. The paper also proposes the current problems of multi-modal fusion and the future research directions.

    Related Articles | Metrics
    Application Research of Improved YOLOv4 in Remote Sensing Aircraft Target Detection
    HOU Tao, JIANG Yu
    Computer Engineering and Applications    2021, 57 (12): 224-230.   DOI: 10.3778/j.issn.1002-8331.2011-0248
    Abstract230)      PDF(pc) (2986KB)(255)       Save

    Aiming at the problems of low accuracy, slow detection speed and complex background of aircraft targets in remote sensing images, an improved YOLOv4 target detection algorithm based on deep learning is proposed. The backbone feature extraction network of YOLOv4 is improved to retain the high-resolution feature layer, remove the feature layer used to detect large targets, and to reduce semantic loss. DenseNet (Densely connected Network) is adopted to enhance feature extraction and reduce the vanishing gradient problem. The [K]-means algorithm on the data set is used to get the best prior frame number and size. Experimental results on RSOD(Remote Sensing Object Detection) data set and DIOR(Detection in Optical Remote sensing images) data set show that the accuracy of the proposed algorithm reaches 95.4%, which is 0.3 percentage points higher than original algorithm, and the recall rate reaches 86.04%, an increase of 4.68 percentage points, and then mAP value reaches 85.52%, an increase of 5.27 percentage points.

    Related Articles | Metrics
    Research on Algorithm of Inspection Path Planning for Substation Robot
    WANG Xiuli, ZHOU Peng, HOU Jingnan, WANG Shijun, LIN Xia
    Computer Engineering and Applications    2021, 57 (14): 245-250.   DOI: 10.3778/j.issn.1002-8331.2004-0378
    Abstract113)      PDF(pc) (1355KB)(254)       Save

    In view of the inspection path problem of the inspection robot which is widely used in the substation at present, this paper discusses the inspection path problem, and according to the Hamilton loop method in the discrete mathematics theory and the routine path planning algorithm, this paper studies a kind of robot inspection path problem when the inspection paths of substations are rectangular. According to the different inspection methods, the following suggestions are put forward:global inspection mode, using the new Hamilton algorithm, when the starting point is set, any other point being inspected can be found out the Hamiltonian loop of the inspection path; the inspection mode of key equipment, using the method of Dijkstra algorithm combing with genetic algorithm; fixed-point inspection mode, using Dijkstra algorithm. The simulation results of three inspection methods are given respectively for sufficient power, low power return and continuing inspection after charging, and they are tested in a substation. The results show that the three algorithms are effective and feasible, and the new Hamilton loop algorithm is relatively shorter than the conventional algorithm, and the algorithm is relatively faster.

    Related Articles | Metrics
    Research on Dimensional Emotion Recognition Model Based on ConvLSTM Network
    MI Zhenmei, ZHAO Hengbin, GAO Pan
    Computer Engineering and Applications    2021, 57 (18): 289-296.   DOI: 10.3778/j.issn.1002-8331.2005-0200
    Abstract86)      PDF(pc) (1680KB)(252)       Save

    Academic emotions can affect and regulate learners’ attention, memory, thinking and other cognitive activities. Automatic emotion recognition is the basis of emotion interaction and instructional decision in intelligent learning environment. At present, the research of emotion recognition mainly focuses on the recognition of discrete emotions, which is discontinuous in the timeline, and cannot accurately depict the evolution process of students’ academic emotions. In order to solve this problem, this paper establishes the dimensional emotional database of middle school students based on the crowd-sourcing method in the real online learning situation. And a deep learning analysis model based on continuous dimensional affective prediction is designed. In the experiment, learning materials that stimulate students’ academic emotions are identified according to students’ learning styles firstly. And then 32 experimenter are recruited for independent online learning and collecting real-time facial images. Next, dimensional database with 157 students’ academic emotion videos and 2 178 students’ facial expressions is obtained by the two-denationalization for each video emotion. Finally, a ConvLSTM net-based dimensional emotion model is established and tested on the dimensional emotion database for middle school students. The mean value of the Concordance Correlation Coefficient(CCC) is 0.581. Meanwhile, the mean value of the uniform correlation coefficient is 0.222 after the experiment on Aff-Wild database. The experiment shows that the dimension-based emotion model proposed in this paper improves the CCC correlation coefficient index by 7.6% to 43.0% in the dimension-based emotion recognition of Aff-Wild database.

    Related Articles | Metrics
    Modified Ant Colony Optimization Algorithm for Discounted {0-1} Knapsack Problem
    ZHANG Ming, DENG Wenhan, LIN Juan, ZHONG Yiwen
    Computer Engineering and Applications    2021, 57 (13): 85-95.   DOI: 10.3778/j.issn.1002-8331.2006-0278
    Abstract101)      PDF(pc) (932KB)(251)       Save

    Discounted {0-1} Knapsack Problem (DKP) is a NP-hard combinatorial optimization problem. Although several intelligent optimization algorithms have been proposed for the DKP, no study has been found to use Ant Colony Optimization (ACO) algorithm to solve this problem. This paper presents a Modified ACO(MACO) algorithm for the DKP. The MACO algorithm uses integer coding to ensure that at most one item in each group is selected. In each step of the solution construction, MACO adopts an intra-group competitive selection to decrease time complexity. In the selection probability calculation equation, the original heuristic information is discarded to decrease the number of parameters and simplify the implementation of parameters tuning. Furthermore, a greedy hybrid optimization strategy based on both value density and value of items is designed to enhance the repaired solutions constructed by ants. The overall performance of MACO is tested and compared with other methods on four kinds of test cases. The experimental results show that the MACO algorithm is significantly superior to other algorithms.

    Related Articles | Metrics
    Application of Deep Reinforcement Learning Algorithm on Intelligent Military Decision System
    KUANG Liqun, LI Siyuan, FENG Li, HAN Xie, XU Qingyu
    Computer Engineering and Applications    2021, 57 (20): 271-278.   DOI: 10.3778/j.issn.1002-8331.2104-0114
    Abstract146)      PDF(pc) (1223KB)(249)       Save

    Deep reinforcement learning algorithm can well achieve discrete decision-making behavior, but it is difficult to apply to the highly complex and continuous modern battlefield situations, and the algorithm is difficult to converge in multi-agent environment. To solve these problems, an improved Deep Deterministic Policy Gradient(DDPG) algorithm is proposed, which introduces the experience replay technology based on priority and single training mode to improve the convergence speed of the algorithm; at the same time, an exploration strategy of mixed double noise is designed in the algorithm to realize complex and continuous military decision-making and control behavior. The intelligent military decision simulation platform based on the improved DDPG algorithm is developed by unity3D. The simulation environment of Blue Army Infantry attacking Red Army military base is built to simulate multi-agent combat training. The experimental results show that the algorithm can drive multiple combat agents to complete tactical maneuvers and achieve tactical behaviors, such as bypassing obstacles to reach the dominant area for shooting. The algorithm has faster convergence speed and better stability. It can get higher round rewards, and achieves the purpose of improving the efficiency of intelligent military decision-making.

    Reference | Related Articles | Metrics
    Overview on Video Abnormal Behavior Detection of GAN via Human Density
    SHEN Xulin, LI Chaobo, LI Hongjun
    Computer Engineering and Applications    2022, 58 (7): 21-30.   DOI: 10.3778/j.issn.1002-8331.2110-0364
    Abstract98)      PDF(pc) (1720KB)(246)       Save
    As an important branch of computer vision, video anomaly detection is a challenging task for intelligent video surveillance systems. It is generally referred to as automatic recognition of videos that contain abnormal targets, events or behaviors, which plays a vital role in ensuring public safety. Generative adversarial network(GAN) is anemerging unsupervised method, which can not only be used to generate images, its unique adversarial learning idea also shows good development potential in the field of anomaly detection. Firstly, the framework of the GAN is introduced. Secondly, according to the density of the scene and the object on which the action is taking place, the research status of video anomaly detection based on GAN is discussed from two aspects of individual behavior anomalies, group anomalies. These two types of abnormalities are further elaborated on the basic of reconstruction and prediction methods respectively. Thirdly, the common datasets for video anomaly detection are briefly introduced, finally, the future development is prospected.
    Reference | Related Articles | Metrics
    Algorithm for Portrait Segmentation Combined with MobileNetv2 and Attention Mechanism
    WANG Xin, WANG Meili, BIAN Dangwei
    Computer Engineering and Applications    2022, 58 (7): 220-228.   DOI: 10.3778/j.issn.1002-8331.2106-0334
    Abstract95)      PDF(pc) (1064KB)(237)       Save
    As for low precision and efficiency in portrait segmentation, an algorithm for portrait segmentation combined with MobileNetv2 and attention mechanism is proposed to achieve the portrait segmentation. With keeping the encoder-decoder of U-typed network , MobileNetv2 is used as the backbone of the network and streamline the upsampling process, it can reduce the parameters of the network. It is helpful for transfer and network training. The network with attention mechanism can learn portrait features more effectively, and the mixed loss is beneficial to the classification of difficult pixels of portrait edges. A portrait bust can be selected as the input of the model, and the corresponding image mask can be produced by the network. The proposed algorithm is tested on Human_Matting dataset and EG1800 dataset. The results show that the accuracy of the proposed algorithm is 98.3%(Matting) and 97.8%(EG1800), which is higher than PortraitNet(96.3%(Matting) and 95.8%(EG1800)) and DeepLabv3+(96.8%(Matting) and 96.4%(EG1800)). The algorithm can clearly separate the target person from the background. The proposed algorithm’s IOU can reach to 98.6%(Matting) and 98.2%(EG1800), which can be used in lightweight applications and provides a new research idea for portrait segmentation.
    Reference | Related Articles | Metrics
    Review of Cognitive and Joint Anti-Interference Communication in Unmanned System
    WANG Guisheng, DONG Shufu, HUANG Guoce
    Computer Engineering and Applications    2022, 58 (8): 1-11.   DOI: 10.3778/j.issn.1002-8331.2109-0334
    Abstract227)      PDF(pc) (913KB)(232)       Save
    As the electromagnetic environment becomes more and more complex as well as the confrontation becomes more and more intense, it puts forward higher requirements for the reliability of information transmission of unmanned systems whereas the traditional cognitive communication mode is difficult to adapt to the independent and distributed development trend of broadband joint anti-interference in future. For the need of low anti-interference intercepted communications surrounded in unmanned systems, this paper analyzes the cognitive anti-interference technologies about interference detection and identification, transformation analysis and suppression in multiple domains and so on. The research status of common detection and estimation, classification and recognition are summarized. Then, typical interference types are modeled correspondingly, and transformation methods and processing problems are concluded. Furthermore, traditional interference suppression methods and new interference suppression methods are systematically summarized. Finally, the key problems of restricting the joint interference of broadband are addressed, such as the classification and recognition of unknown interference, the temporal elimination of multiple interference, the joint separation of distributed interference and the optimal control of collaborative interference, which highlight the important role of cognitive interference suppression technology in unmanned system communication.
    Reference | Related Articles | Metrics
    Detection Method of Illegal Building Based on YOLOv5
    YU Juan,LUO Shun
    Computer Engineering and Applications    2021, 57 (20): 236-244.   DOI: 10.3778/j.issn.1002-8331.2106-0178
    Abstract164)      PDF(pc) (1653KB)(231)       Save

    Aiming at solving the problem of slow detection rate and high false detection rate caused by the illegal buildings in the UAV images, which are mostly small targets and partially occluded targets, a detection method of illegal buildings based on YOLOv5 network is proposed. Firstly, at the beginning and end of original batch normalization module, centering and scaling calibration are added to enhance the effective features and form a more stable feature distribution, which strengthens the feature extraction ability of network model. Then, the smoothed KL(Kullback-Leibler) divergence loss function is adopted to replace the cross entropy in the confidence of original loss function, which further improves the generalization performance of model. Finally, the backbone feature extraction network of YOLOv5 is improved, and the residual module is replaced with the LSandGlass module to reduce information loss and eliminate low-resolution feature layers to reduce semantic loss. Experimental results show that the training of the proposed improved model is easier to make network converge in comparison with original YOLOv5, and the speed of detecting illegal buildings has been greatly improved, and then detection accuracy has been improved.

    Reference | Related Articles | Metrics
    Summary of Dynamic Gesture Recognition Based on Vision
    XIE Yinggang, WANG Quan
    Computer Engineering and Applications    2021, 57 (22): 68-77.   DOI: 10.3778/j.issn.1002-8331.2105-0314
    Abstract157)      PDF(pc) (598KB)(229)       Save

    Gestures have played a very important role in human communication since ancient times, and the visual dynamic gesture identification technology is to use new technologies such as computer vision and IOT(Internet of Things) perception, and 3D visual sensors, allowing the machine to understand human gestures, thus making humanity and machine more good communication, because of far-reaching research significance for human-machine interaction. The sensor techniques used in dynamic gesture identification are introduced, and the technical parameters of the related sensors are compared. By tracking the dynamic gesture recognition technology of vision at home and abroad, the processing process of dynamic gesture recognition is first stated:gesture detection and segmentation, gesture tracking, gesture classification. By comparing the methods involved in each process, it can be seen that deep learning has strong fault tolerance, robustness, high parallelism, anti-interference, etc., which has achieved great achievements above the traditional learning algorithm in the field of gesture identification. Finally, the challenges currently encountering and the future possible development of dynamic gesture identification are analyzed.

    Reference | Related Articles | Metrics
    Application of Generative Adversarial Networks in Medical Image Processing
    LI Xiangxia, XIE Xian, LI Bin, YIN Hua, XU Bo, ZHENG Xinwei
    Computer Engineering and Applications    2021, 57 (18): 24-37.   DOI: 10.3778/j.issn.1002-8331.2104-0176
    Abstract211)      PDF(pc) (726KB)(229)       Save

    Generative Adversarial Nets(GANs) models can learn more abundant data information in unsupervised learning. GANs consist of a generator and a discriminator, and these two are alternately optimized through mutual games in the training of the confrontation to improve performance. In view of the problems of traditional generative confrontation network, such as gradient disappearance, mode collapse and inability to generate discrete data distribution, the researchers have proposed a number variations of GANs model. The paper describes the theory and structure of the GANs model. Then, the paper introduces several typical variant models, and elaborates the current research progress and status of the GANs model in image generation, image segmentation, image classification, target detection applications and super resolution image reconstruction. The in-depth analysis is carried out based on the research status and existing problems in the paper, and the future development trend and challenges of deep learning in the field of medical image processing are further summarized and discussed.

    Related Articles | Metrics
    Research on Optimal of Time-of-Day Control Data Input Source at Intersection
    XU Chen, DONG Decun, OU Dongxiu
    Computer Engineering and Applications    2021, 57 (17): 230-236.   DOI: 10.3778/j.issn.1002-8331.2103-0142
    Abstract69)      PDF(pc) (1127KB)(228)       Save

    In order to overcome the arbitrariness and empirical nature of the continuous data discretization selection of time-of-day control model data input source, this paper proposes an optimization method for selecting the input source of an deep attention recursive network based on the combination of sensor network and artificial intelligence theory. Firstly, it uses the Synchron and Sumo simulation evaluation function module to standardize the typical samples of the input source. Secondly, the key attributes such as the starting time are used as the model input. At the same time, the optimal data input source is selected as the data output to build a model. Finally, it is realized by simulating the input layer, middle layer, output layer and optimization method of the entire model, and the actual traffic flow data of a certain city is used as the test data for evaluation and comparison analysis. The results show that, compared with other traditional methods that select 50% and 80% of the traffic volume as the fixed input source of the model, the innovative model in this paper is more accurate and efficient. The total delay time of the whole day is effectively reduced.

    Related Articles | Metrics
    Design of Digital Twin System for Quadrotor
    WU Dongyang, DOU Jianping, LI Jun
    Computer Engineering and Applications    2021, 57 (16): 237-244.   DOI: 10.3778/j.issn.1002-8331.1912-0038
    Abstract98)      PDF(pc) (3249KB)(226)       Save

    With the widespread use of quadrotors in engineering practice, the requirements for their control systems’ performance such as accuracy, rapidity, adaptability are also enhanced. Digital twin as research focus in recent years can simulate the behavior of physical entities in the real environment by means of data and models. The operational efficiency of physical entities or systems can be improved by the precise digitization of physical objects. The digital twin technology is introduced into the control system of quadrotor, and the architecture of digital twin system of quadrotor is established. The related technologies involved in the implementation of the system are described in detail:How to model the digital twin system of quadrotor? How to use sensor technology and wireless communication technology to solve the data acquisition and transmission? And how to use Hadoop technology for big data storage and mining? Subsequently, model simulations and field tests are carried out on the actual quadrotor to verify the effectiveness of the system. The system can optimize the flight control and fuselage state by using the twin data and make the operation process of the quadrotors transparent.

    Related Articles | Metrics
    Adversarial Attack Algorithm Based on Erosion Batch Normalization
    ZHANG Wu, ZHOU Xingyu, ZOU Junhua, PAN Zhisong, DUAN Yexin, CHEN Jun
    Computer Engineering and Applications    2021, 57 (16): 116-124.   DOI: 10.3778/j.issn.1002-8331.2101-0301
    Abstract89)      PDF(pc) (1230KB)(223)       Save

    For adversarial examples generation research, gradient-based attack methods are widely used due to fast generation speed and low resource consumption. However, the adversarial examples generated by most existing gradient-based attack methods still exhibit low efficiency in black-box attacks. The state-of-the-art gradient-based attack method only reaches an average success rate of 78.2% when attacking six advanced defense black-box models. To this end, a generation algorithm based on erosion batch normalization layer in deep neural network architecture is proposed to improve existing gradient-based attack methods, so as to generate adversarial examples with higher black-box attack success rates. Extensive experiments on an ImageNet-compatible dataset are conducted under single-model setting and multi-model setting, and the results show that the proposed algorithm can be effectively combined with existing gradient-based attack methods and obtain higher attack success rates with similar computational cost. In addition, the proposed algorithm makes the state-of-the-art gradient-based attack method achieve an increase of 9.0 percentage points in the average success attack rate against six advanced black-box defense models.

    Related Articles | Metrics