Most Read articles

    Published in last 1 year |  In last 2 years |  In last 3 years |  All

    Please wait a minute...
    For Selected: Toggle Thumbnails
    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
    Abstract672)      PDF(pc) (1134KB)(422)       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 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
    Abstract452)      PDF(pc) (1078KB)(1045)       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 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
    Abstract433)      PDF(pc) (1078KB)(419)       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
    Abstract412)      PDF(pc) (1089KB)(373)       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
    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
    Abstract407)      PDF(pc) (973KB)(475)       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
    Research Progress of Natural Language Processing Based on Deep Learning
    JIANG Yangyang, JIN Bo, ZHANG Baochang
    Computer Engineering and Applications    2021, 57 (22): 1-14.   DOI: 10.3778/j.issn.1002-8331.2106-0166
    Abstract407)      PDF(pc) (1781KB)(109)       Save

    This paper comprehensively analyzes the research of deep learning in the field of natural language processing through a combination of quantitative and qualitative methods. It uses CiteSpace and VOSviewer to draw a knowledge graph of countries, institutions, journal distribution, keywords co-occurrence, co-citation network clustering, and timeline view of deep learning in the field of natural language processing to clarify the research. Through mining important researches in the field, this paper summarizes the research trend, the main problems, development bottlenecks, and gives corresponding solutions and ideas. Finally, suggestions are given on how to track the research of deep learning in the field of natural language processing, and provides references for subsequent research and development in the field.

    Reference | Related Articles | Metrics
    Improved U-Net Network for COVID-19 Image Segmentation
    SONG Yao, LIU Jun
    Computer Engineering and Applications    2021, 57 (19): 243-251.   DOI: 10.3778/j.issn.1002-8331.2010-0207
    Abstract376)      PDF(pc) (915KB)(204)       Save

    The novel corona virus pneumonia(COVID-19) pandemic is spreading globally. Computerized Tomography(CT) imaging technology plays a vital role in the fight against global COVID-19. When diagnosing new coronary pneumonia, it will be helpful if the new coronary pneumonia focus area can be automatically and accurately segmented from the CT image, the doctor makes a more accurate and quick diagnosis. Aiming at the segmentation problem of new coronary pneumonia lesions, an automatic segmentation method based on the improved U-Net model is proposed. The EfficientNet-B0 network pre-trained on ImageNet is used in the encoder to extract features of effective information. In the decoder, the traditional up-sampling operation is replaced with a DUpsampling structure, in order to fully obtain the detailed feature information of the lesion edge, and finally the accuracy of the segmentation is improved through the integration of model snapshots. The experimental results on the public data set show that the accuracy, recall and Dice coefficients of the proposed algorithm are 84.24%, 80.43% and 85.12%, respectively. Compared with other segmentation networks, this method can effectively segment the neo-coronary pneumonia lesion area and has good segmentation performance.

    Reference | 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
    Abstract329)      PDF(pc) (896KB)(491)       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
    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
    Abstract315)      PDF(pc) (1343KB)(309)       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
    Multi-channel Attention Mechanism Text Classification Model Based on CNN and LSTM
    TENG Jinbao, KONG Weiwei, TIAN Qiaoxin, WANG Zhaoqian, LI Long
    Computer Engineering and Applications    2021, 57 (23): 154-162.   DOI: 10.3778/j.issn.1002-8331.2104-0212
    Abstract304)      PDF(pc) (844KB)(216)       Save

    Aiming at the problem that traditional Convolutional Neural Network(CNN) and Long Short-Term Memory (LSTM) can not reflect the importance of each word in the text when extracting features, this paper proposes a multi-channel text classification model based on CNN and LSTM. Firstly, CNN and LSTM are used to extract the local information and context features of the text; secondly, multi-channel attention mechanism is used to extract the attention score of the output information of CNN and LSTM; finally, the output information of multi-channel attention mechanism is fused to achieve the effective extraction of text features and focus attention on important words. Experimental results on three public datasets show that the proposed model is better than CNN, LSTM and their improved models, and can effectively improve the effect of text classification.

    Reference | Related Articles | Metrics
    COVID-19 Medical Imaging Dataset and Research Progress
    LIU Rui, DING Hui, SHANG Yuanyuan, SHAO Zhuhong, LIU Tie
    Computer Engineering and Applications    2021, 57 (22): 15-27.   DOI: 10.3778/j.issn.1002-8331.2106-0118
    Abstract283)      PDF(pc) (1013KB)(227)       Save

    As imaging technology has been playing an important role in the diagnosis and evaluation of the new coronavirus(COVID-19), COVID-19 related datasets have been successively published. But few review articles discuss COVID-19 image processing, especially in datasets. To this end, the new coronary pneumonia datasets and deep learning models are sorted and analyzed, through COVID-19-related journal papers, reports, and related open-source dataset websites, which include Computer Tomography(CT) image and X-rays(CXR)image datasets. At the same time, the characteristics of the medical images presented by these datasets are analyzed. This paper focuses on collating and describing open-source datasets related to COVID-19 medical imaging. In addition, some important segmentation and classification models that perform well on the related datasets are analyzed and compared. Finally, this paper discusses the future development trend on lung imaging technology.

    Reference | Related Articles | Metrics
    Intelligent Analysis of Text Information Disclosure of Listed Companies
    LYU Pin, WU Qinjuan, XU Jia
    Computer Engineering and Applications    2021, 57 (24): 1-13.   DOI: 10.3778/j.issn.1002-8331.2106-0270
    Abstract281)      PDF(pc) (724KB)(225)       Save

    The analysis of the text disclosure issued by listed companies is an important way for investors to understand the companies’ operating conditions and to make investment decisions. However, the method based on manual reading and analysis has low efficiency and high cost. The development of artificial intelligence technology provides an opportunity for intelligent analysis of companies’ text information, which can mine valuable information from massive enterprise text data, fulfill the advantages of data-driven, and greatly improve the analysis efficiency. Hence, it has become a research hotspot in recent years. The research work in recent ten years about the announcement of listed companies is summarized from three aspects:the event types of the text information disclosure, intelligent analysis method and application scenario. The current challenges in this field are also discussed, and possible future research directions according to the existing shortcomings are finally pointed out.

    Reference | 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
    Abstract277)      PDF(pc) (1267KB)(273)       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
    Improved Lightweight Attention Model Based on CBAM
    FU Guodong, HUANG Jin, YANG Tao, ZHENG Siyu
    Computer Engineering and Applications    2021, 57 (20): 150-156.   DOI: 10.3778/j.issn.1002-8331.2101-0369
    Abstract253)      PDF(pc) (808KB)(186)       Save

    In recent years, the attention model has been widely used in the field of computer vision. By adding the attention module to the convolutional neural network, the performance of the network can be significantly improved. However, most of the existing methods focus on the development of more complex attention modules to enable the convolutional neural network to obtain stronger feature expression capabilities, but this also inevitably increases the complexity of the model. In order to achieve a balance between performance and complexity, a lightweight EAM(Efficient Attention Module) model is proposed to optimize the CBAM model. For the channel attention module of CBAM, one-dimensional convolution is introduced to replace the fully connected layer to aggregate the channels. For the spatial attention module of CBAM, the large convolution kernel is replaced with a dilated convolution to increase the receptive field for aggregation Broader spatial context information. After integrating the model into YOLOv4 and testing it on the VOC2012 data set, mAP is increased by 3.48 percentage points. Experimental results show that the attention model only introduces a small amount of parameters, and the network performance can be greatly improved.

    Reference | 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
    Abstract248)      PDF(pc) (779KB)(387)       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
    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
    Abstract240)      PDF(pc) (913KB)(236)       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
    YOLOv5 Helmet Wear Detection Method with Introduction of Attention Mechanism
    WANG Lingmin, DUAN Jun, XIN Liwei
    Computer Engineering and Applications    2022, 58 (9): 303-312.   DOI: 10.3778/j.issn.1002-8331.2112-0242
    Abstract240)      PDF(pc) (1381KB)(212)       Save
    For high-risk industries such as steel manufacturing, coal mining and construction industries, wearing helmets during construction is one of effective ways to avoid injuries. For the current helmet wearing detection model in a complex environment for small and dense targets, there are problems such as false detection and missed detection, an improved YOLOv5 target detection method is proposed to detect the helmet wearing. A coordinate attention mechanism(coordinate attention) is added to the backbone network of YOLOv5, which embeds location information into channel attention so that the network can pay attention on a larger area. The original feature pyramid module in the feature fusion module is replaced with a weighted bi-directional feature pyramid(BiFPN)network structure to achieve efficient bi-directional cross-scale connectivity and weighted feature fusion. The experimental results on the homemade helmet dataset show that the improved YOLOv5 model achieves an average accuracy of 95.9%, which is 5.1 percentage points higher than the YOLOv5 model, and meets the requirements for small and dense target detection in complex environments.
    Reference | Related Articles | Metrics
    Review of Sign Language Recognition Methods and Techniques
    Minawaer·Abula, Alifu·Kuerban, XIE Qina, GENG Liting
    Computer Engineering and Applications    2021, 57 (18): 1-12.   DOI: 10.3778/j.issn.1002-8331.2104-0220
    Abstract238)      PDF(pc) (719KB)(235)       Save

    Sign language, as the main communication channel for deaf and hearing people, plays a crucial role in daily life. With the rapid development of the field of computer vision and deep learning, the field of sign language recognition has also ushered in new opportunities. The advanced methods and technologies used in the research of sign language recognition based on computer vision in recent years are reviewed. Starting from the three branches of static sign language, isolated words and continuous sentence sign language recognition, the common methods and technical difficulties of sign language recognition are systematically explained. The steps of sign language recognition such as image preprocessing, detection and segmentation, tracking, feature extraction, and classification are introduced in detail. It summarizes and analyzes the commonly used algorithms and neural network models for sign language recognition, summarizes and organizes commonly used sign language datasets, analyzes the status quo of different sign language recognition, and finally discusses the challenges and limitations of sign language recognition.

    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
    Abstract238)      PDF(pc) (726KB)(238)       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
    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
    Abstract235)      PDF(pc) (1214KB)(314)       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
    Computer Engineering and Applications    2021, 57 (19): 0-0.  
    Abstract215)      PDF(pc) (635KB)(179)       Save
    Related Articles | Metrics
    Survey of Task Assignment for Crowd-Based Cooperative Computing
    CHEN Baotong, WANG Liqing, JIANG Xiaomin, YAO Hanbing
    Computer Engineering and Applications    2021, 57 (20): 1-12.   DOI: 10.3778/j.issn.1002-8331.2105-0396
    Abstract213)      PDF(pc) (689KB)(194)       Save

    Task allocation is one of the core issues in crowd-based cooperative computing and crowdsourcing, that is, by designing a reasonable task allocation strategy, the tasks are assigned to the appropriate workers under the task constraints, so as to improve the result quality and completion efficiency of tasks. The problems of the current task allocation method are analyzed first, and then, a general task allocation framework is proposed and the relevant research work at home and abroad is analyzed in three aspects:worker model, task model, and task allocation algorithm. Finally, the key issues and future research trends in the research of task allocation for crowd-based cooperative computing are put forward.

    Reference | Related Articles | Metrics
    Review on Integration Analysis and Application of Multi-omics Data
    ZHONG Yating, LIN Yanmei, CHEN Dingjia, PENG Yuzhong, ZENG Yuanpeng
    Computer Engineering and Applications    2021, 57 (23): 1-17.   DOI: 10.3778/j.issn.1002-8331.2106-0341
    Abstract212)      PDF(pc) (806KB)(293)       Save

    With the continuous emergence and popularization of new omics sequencing technology, a large number of omics data have been produced, which is of great significance for people to further study and reveal the mysteries of life. Using multi-omics data to integrate and analyze life science problems can obtain more abundant and more comprehensive information related to life system, which has become a new direction for scientists to explore the mechanism of life. This paper introduces the research background and significance of multi-omics data integration analysis, summarizes the methods of data integration analysis of multiomics in recent years and the applied research in related fields, and finally discusses the current existing problems and future prospects of multi-omics data integration analysis methods.

    Reference | Related Articles | Metrics
    Survey of Single Image Super-Resolution Based on Deep Learning
    HUANG Jian, ZHAO Yuanyuan, GUO Ping, WANG Jing
    Computer Engineering and Applications    2021, 57 (18): 13-23.   DOI: 10.3778/j.issn.1002-8331.2102-0257
    Abstract211)      PDF(pc) (996KB)(234)       Save

    Image super-resolution reconstruction refers to the use of a specific algorithm to restore a low-resolution blurry image in the same scene to a high-resolution image. In recent years, with the active development of deep learning, this technology has been widely used in many fields, and methods based on deep learning are being increasingly studied in the field of image super-resolution reconstruction. In order to understand the current status and research trends of image super-resolution reconstruction algorithms based on deep learning, popular image super-resolution algorithms are summarized. Mainly, the network model structure, scaling method, loss function of existing single image super-resolution algorithm are explained in detail. The drawbacks and advantages of various methods are analyzed. The reconstruction effects of various network models and various loss functions are compared and analyzed throughout the experiment. Finally, the future development direction of the single-image super-resolution reconstruction algorithm based on deep learning is forecasted.

    Related Articles | Metrics
    Quick Semantic Segmentation Network Based on U-Net and MobileNet-V2
    LAN Tianxiang, XIANG Ziyu, LIU Mingguo, CHEN Kai
    Computer Engineering and Applications    2021, 57 (17): 175-180.   DOI: 10.3778/j.issn.1002-8331.2005-0278
    Abstract211)      PDF(pc) (1156KB)(144)       Save

    The U-Net model is large and accordingly has relatively slow speed on image processing. This drawback makes it difficult to satisfy the requirement of industrial real-time applications. On the consideration of the question above, this paper designs a light-weight full convolution neural network named LU-Net. In the proposed network, it integrates the thought of MobileNet-V2 into the U-Net framework. The depth separable convolution method in MobileNet-V2 is efficient to reduce the parameters and the computation complexity of the proposed network. The proposed network also reserves the advantages of normal convolution and bottleneck model. Accordingly, it is efficient to utilize the high-level features to keep the accuracy and reduce the processing time simultaneously. The experiments on hollow symbol dataset and DRIVE dataset indicate that, on the comparison with U-Net, the parameters of the proposed LU-Net is 0.59×106,which is 1.9% of the original model, and the processing speed is 5 times faster. Under the experimental environment, LU-Net takes only 25?ms to process a picture under the resolution of 360×270 size. LU-Net is a promising method for the industrial real-time applications on picture processing.

    Related Articles | Metrics
    Relation Network Based on Attention Mechanism and Graph Convolution for Few-Shot Learning
    WANG Xiaoru, ZHANG Heng
    Computer Engineering and Applications    2021, 57 (19): 164-170.   DOI: 10.3778/j.issn.1002-8331.2104-0275
    Abstract208)      PDF(pc) (949KB)(174)       Save

    Deep neural networks have dominated image recognition task with large amounts of labeled data. But training a well-performing network on a smaller dataset is still a very challenging task. How to learn from limited labeled data is a key research with excellent scenarios and potential applications. There are many ways to solve few-shot recognition problem, but there is still a problem of low recognition accuracy. The fundamental reason is that in few-shot learning, the traditional neural network can only accept a small amount of labeled data, which makes the network unable to obtain enough information for identification. Therefore, the paper proposes a few-shot classification model based on attention mechanism and graph convolutional neural network, which can not only extract features better, but also make full use of the features to classify the target image. Through the attention mechanism, it can guide the neural network to pay attention to more useful information, and graph convolution enables the network to make more accurate judgments by using the information from other classes of support set. Through many experiments, it is proved that the classification accuracy of the model on the Omniglot dataset and the miniImageNet dataset surpasses the original relational network which based on traditional neural network.

    Reference | Related Articles | Metrics
    Computer Engineering and Applications    2021, 57 (24): 0-0.  
    Abstract206)      PDF(pc) (1168KB)(284)       Save
    Related Articles | Metrics
    Review on Semantic Segmentation of UAV Aerial Images
    CHENG Qing, FAN Man, LI Yandong, ZHAO Yuan, LI Chenglong
    Computer Engineering and Applications    2021, 57 (19): 57-69.   DOI: 10.3778/j.issn.1002-8331.2105-0423
    Abstract203)      PDF(pc) (926KB)(197)       Save

    With the rapid development of Unmanned Aerial Vehicle(UAV) technology, research institutions and industries have attached importance of UAV’s application. Optical images and videos are vital for the UAV to sense the environment, occupying an important position in UAV vision. As a hot spot of the current research of computer vision, semantic segmentation is widely investigated in the fields of unmanned driving and intelligent robot. Semantic segmentation of UAV aerial images is based on the UAV aerial image semantic segmentation technology to enable the UAV to work in complex scenes. First of all, a brief introduction to the semantic segmentation technology and the application development of UAV is given. Meanwhile, the relevant UAV aerial data sets, characteristics of aerial images and commonly used evaluation metrics for semantic segmentation are introduced. Secondly, according to the characteristics of UAV aerial images, it introduces the relevant semantic segmentation methods. In this section, analysis and comparison are made in three aspects including the small object detection, the real-time performance of the models and the multi-scale information integration. Finally, the related applications of semantic segmentation for UAV are reviewed, including line detection, the application of agriculture and building extraction, and analysis of the development trend and challenges in the future is made.

    Reference | Related Articles | Metrics
    Summary of Intrusion Detection Models Based on Deep Learning
    ZHANG Hao, ZHANG Xiaoyu, ZHANG Zhenyou, LI Wei
    Computer Engineering and Applications    2022, 58 (6): 17-28.   DOI: 10.3778/j.issn.1002-8331.2107-0084
    Abstract199)      PDF(pc) (997KB)(178)       Save
    With the continuous in-depth development of deep learning technology, intrusion detection model based on deep learning has become a research hotspot in the field of network security. This paper summarizes the commonly used data preprocessing operations in network intrusion detection. The popular intrusion detection models based on deep learning, such as convolutional neural network, long short-term memory network, auto-encode and generative adversarial networks, are analyzed and compared. The data sets commonly used in the research of intrusion detection model based on deep learning are introduced. It points out the problems of the existing intrusion detection models based on deep learning in data set timeliness, real-time, universality, model training time and other aspects, and the possible research focus in the future.
    Reference | Related Articles | Metrics
    Review of Occlusion Face Recognition Methods
    XU Xialing, LIU Tao, TIAN Guohui, YU Wenjuan, XIAO Dajun, LIANG Shanpeng
    Computer Engineering and Applications    2021, 57 (17): 46-60.   DOI: 10.3778/j.issn.1002-8331.2101-0389
    Abstract196)      PDF(pc) (1793KB)(154)       Save

    With the gradual expansion of the application field of face recognition, face detection in occlusion environment is facing certain technical challenges. Because of its strong learning ability, deep learning method has become a better solution to the problem of face occlusion detection, but it still faces many problems to be solved. Reducing the influence of occlusion on the performance of detection algorithm is one of the key and difficult problems in this field. This paper analyzes the related research progress from the perspective of model, algorithm and data set, compares the basic principles of different algorithms, model performance, advantages and disadvantages and existing problems, and discusses the possible research direction in the future.

    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
    Abstract188)      PDF(pc) (1653KB)(262)       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
    Chinese Character Generation Method Based on Deep Learning
    HUANG Zijun, CHEN Qi, LUO Wenbing
    Computer Engineering and Applications    2021, 57 (17): 29-36.   DOI: 10.3778/j.issn.1002-8331.2103-0297
    Abstract187)      PDF(pc) (775KB)(183)       Save

    Handwritten Chinese Character Generation(HCCG) is an important research direction in machine learning. In the past two decades the studies of handwritten Chinese character generation can be roughly divided into two stages:the research in the early stage mainly used the explicit characteristics of Chinese characters, such as the structure and strokes, to realize the decomposition of Chinese characters, and then realized the generation of Chinese characters through algorithms. This type of method has relatively high requirements for the decomposition accuracy of Chinese characters dataset, which limits the wide application of this type of method. The current research on Chinese character generation mainly uses deep neural networks to extract the implicit features of Chinese characters, to generate higher-quality Chinese characters and overcome problems such as insufficient data sets in the early research stage. The main purpose of this article is to conduct a comprehensive and systematic review of the existing research on Chinese character generation.

    Related Articles | Metrics
    Review of Path Planning Algorithms for Mobile Robots
    LIN Hanxi, XIANG Dan, OUYANG Jian, LAN Xiaodong
    Computer Engineering and Applications    2021, 57 (18): 38-48.   DOI: 10.3778/j.issn.1002-8331.2103-0519
    Abstract186)      PDF(pc) (865KB)(167)       Save

    Path planning is one of the hot research topics of mobile robot, and it is the key technology to realize autonomous navigation of robot. In this paper, the path planning algorithms of mobile robots are studied to understand the development and application of path planning algorithms under different environments, and the research status and development of path planning are systematically summarized. According to the characteristics of mobile robot path planning, it is divided into intelligent search algorithm, artificial intelligence-based algorithm, geometric model based algorithm and local obstacle avoidance algorithm. Based on the above classification, this paper introduces the representative research results in recent years, analyzes the advantages and disadvantages of various planning algorithms, and forecasts the future development trend of mobile robot path planning, which provides some ideas for the research of robot path planning.

    Related Articles | Metrics
    Overview of Blockchain Application in Supply Chain Management
    TIAN Yang, CHEN Zhigang, SONG Xinxia, LI Tianming
    Computer Engineering and Applications    2021, 57 (19): 70-83.   DOI: 10.3778/j.issn.1002-8331.2105-0136
    Abstract185)      PDF(pc) (740KB)(172)       Save

    Supply chain management faces many challenges, such as ensuring the integrity, authenticity and transparency of product information, trust management between upstream and downstream enterprises, and privacy protection in the process of enterprise interaction, etc. While the blockchain technology provides effective solutions to many tasks of supply chain, there still remain some noticeable challenges. This paper systematically reviews the applications of blockchain technology in the area of supply chain management. Firstly, it discusses the problems faced by the current supply chain management technology, analyzes the limitations of traditional solutions, and highlights the advantages of blockchain technology. Secondly, the mainstream application frameworks and principle of blockchain are systematically reviewed, and the differences between different technologies are compared. Finally, it provides a comprehensive survey of supply chain solutions in different industries, including the solutions based on blockchains. It anticipates that the paper will provide a useful guidance for the application and development of blockchain in supply chain management.

    Reference | Related Articles | Metrics
    Aspect-Level Sentiment Analysis Model Incorporating Multi-layer Attention
    YUAN Xun, LIU Rong, LIU Ming
    Computer Engineering and Applications    2021, 57 (22): 147-152.   DOI: 10.3778/j.issn.1002-8331.2104-0019
    Abstract179)      PDF(pc) (699KB)(161)       Save

    Aspect sentiment analysis aims to analyze the sentiment polarity of a specific aspect in a given text. In order to solve the problem of insufficient introduction of aspect emotional attention in current research methods, this paper proposes an aspect level emotion classification model based on the fusion of BERT and Multi-Layer Attention(BMLA). Firstly, the model extracts the multi-layer aspect emotional attention information from the inner part of BERT, and designs the multi-layer aspect attention by fusing the encoded aspect information with the representation vector of the hidden layer of BERT, then cascades the multi-layer aspect attention with the encoded output text, finally enhances the long dependency relationship between sentences and aspect words. Experiments on the SemEval2014 Task4 and the AI Challenger 2018 datasets show that the proposed model is effective to enhance the weight of the target aspect and interact in context for aspect sentiment classification.

    Reference | Related Articles | Metrics
    TLS Malicious Encrypted Traffic Identification Research
    KANG Peng, YANG Wenzhong, MA Hongqiao
    Computer Engineering and Applications    2022, 58 (12): 1-11.   DOI: 10.3778/j.issn.1002-8331.2110-0029
    Abstract176)      PDF(pc) (747KB)(144)       Save
    With the advent of the 5G era and the increasing public awareness of the Internet, the public has paid more and more attention to the protection of personal privacy. Due to malicious communication in the process of data encryption, to ensure data security and safeguard social and national interests, the research work on encrypted traffic identification is particularly important. Therefore, this paper describes the TLS traffic in detail and analyzes the improved technology of early identification method, including common traffic detection technology, DPI detection technology, proxy technology, and certificate detection technology. It also introduces machine learning models for selecting different TLS encrypted traffic characteristics, as well as many recent research results of deep learning models without feature selection. The deficiencies of the related research work are summarized, and the future research work and development trend of the technology have been prospected.
    Reference | Related Articles | Metrics
    Survey of Opponent Modeling Methods and Applications in Intelligent Game Confrontation
    WEI Tingting, YUAN Weilin, LUO Junren, ZHANG Wanpeng
    Computer Engineering and Applications    2022, 58 (9): 19-29.   DOI: 10.3778/j.issn.1002-8331.2202-0297
    Abstract175)      PDF(pc) (904KB)(71)       Save
    Intelligent game confrontation has always been the focus of artificial intelligence research. In the game confrontation environment, the actions, goals, strategies, and other related attributes of agent can be inferred by opponent modeling, which provides key information for game strategy formulation. The application of opponent modeling method in competitive games and combat simulation is promising, and the formulation of game strategy must be premised on the action strategy of all parties in the game, so it is especially important to establish an accurate model of opponent behavior to predict its intention. From three dimensions of connotation, method, and application, the necessity of opponent modeling is expounded and the existing modeling methods are classified. The prediction method based on reinforcement learning, reasoning method based on theory of mind, and optimization method based on Bayesian are summarized. Taking the sequential game(Texas Hold’em), real-time strategy game(StarCraft), and meta-game as typical application scenarios, the role of opponent modeling in intelligent game confrontation is analyzed. Finally, the development of adversary modeling technology prospects from three aspects of bounded rationality, deception strategy and interpretability.
    Reference | Related Articles | Metrics
    Research of Deep Learning-Based Semantic Segmentation for 3D Point Cloud
    WANG Tao, WANG Wenju, CAI Yu
    Computer Engineering and Applications    2021, 57 (23): 18-26.   DOI: 10.3778/j.issn.1002-8331.2107-0142
    Abstract175)      PDF(pc) (682KB)(167)       Save

    This paper summarizes the methods of deep learning-based semantic segmentation for 3D point cloud. The literature research method is used to describe deep learning-based semantic segmentation methods for 3D point cloud according to the representation of data. It discusses the current situation of domestic and foreign development in recent years, and analyzes the advantages and disadvantages of the current related methods, and prospects the future development trend. Deep learning plays an extremely important role in the research of semantic segmentation technology for point cloud, and promotes the manufacturing, packaging fields and etc to development in the direction of intelligence. According to the advantages and disadvantages of various methods, it is an important research direction to construct a framework model of semantic segmentation combined with 2D-3D for projection, voxel, multi-view and point cloud in the future.

    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
    Abstract174)      PDF(pc) (598KB)(258)       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
    GNN-Based Matrix Factorization Recommendation Algorithm
    WANG Yingbo, SUN Yongdi
    Computer Engineering and Applications    2021, 57 (19): 129-134.   DOI: 10.3778/j.issn.1002-8331.2009-0013
    Abstract171)      PDF(pc) (955KB)(168)       Save

    Compared with collaborative filtering, matrix factorization has better scalability and flexibility, but it is also troubled by data sparseness and cold start. Aiming at the above problems, a recommendation algorithm GNN_MF combining Graph Neural Network(GNN) and Probabilistic Matrix Factorization(PMF) is proposed. The algorithm uses GNN to model social network graphs and user item graphs, connects the two graphs internally, and learns the feature vector of the target user in the social space and item space. Then through Multi-Layer Perceptron(MLP), the two feature vectors are connected in series to extract the user’s potential feature vector. Finally, it is integrated on the probability matrix factorization model to generate prediction scores. A large number of experiments on real data sets Epinions and Ciao show that the root mean square error and average absolute error of the GNN_MF algorithm are reduced by 2.91%, 3.10% and 4.83%, 3.84% respectively compared with traditional PMF. The effectiveness and feasibility of the GNN_MF algorithm in the recommendation system are verified.

    Reference | Related Articles | Metrics