Most Read articles

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

    In last 3 years
    Please wait a minute...
    For Selected: Toggle Thumbnails
    Application of Knowledge Graph in Full-Service Unified Data Center of National Grid
    WANG Yuan, PENG Chenhui, WANG Zhiqiang, FAN Qiang, YAO Yiyang, HUA Zhaoyun
    Computer Engineering and Applications    2019, 55 (15): 104-109.   DOI: 10.3778/j.issn.1002-8331.1810-0002
    Abstract1793)      PDF(pc) (629KB)(1443)       Save
    To solve the problems that the business data in state grid corporation cannot be crossed professionally, and the data resources cannot be intelligently analyzed and managed, this paper proposes a knowledge graph construction method based on the full-service unified data center. On the basis of multi-source data in the full-service unified data center collected by the big data technology, the semantic annotation method is used to extract the knowledge entities, attributions and relations from the structured, semi-structured and unstructured data. The knowledge graph is constructed through the knowledge fusion technology. The accurate result and related information can be returned intelligently according to the user’s search. Experiments show that this method improves the precision and recall rate, and has better intelligent search and analysis ability.
    Related Articles | Metrics
    Research and Implementation of Efficient Connection Middleware Between Flink and MongoDB
    HU Cheng, YE Feng
    Computer Engineering and Applications    2019, 55 (23): 64-69.   DOI: 10.3778/j.issn.1002-8331.1808-0342
    Abstract1354)      PDF(pc) (758KB)(941)       Save
    In order to improve the reading and writing rate between big data processing platform Flink and MongoDB, this paper proposes and implements an efficient connection middleware of Flink and MongoDB. Based on Flink’s parallelization idea, by logically fragmenting the data, the interface in the Mongo-Java package is called to realize parallel reading and writing of data. With different scale of hydrological sensor datasets as experimental data, the reading and writing speeds of the data in Java single-threaded operation, Hadoop and MongoDB connector and the Flink and MongoDB connection middleware proposed in this paper are tested. The results show that the efficiency of using Flink to read and write data is 1.5 times higher than the single-threaded operation, which validates that the connection middleware can effectively improve the reading and writing speed of massive data.
    Related Articles | Metrics
    Comparative Study of Several New Swarm Intelligence Optimization Algorithms
    LI Yali, WANG Shuqin, CHEN Qianru, WANG Xiaogang
    Computer Engineering and Applications    2020, 56 (22): 1-12.   DOI: 10.3778/j.issn.1002-8331.2006-0291
    Abstract961)      PDF(pc) (972KB)(566)       Save

    With the development of computer technology, algorithm technology is constantly and alternately being updated. In recent years, swarm intelligence algorithm has become more and more popular and received extensive attention and research, and has made progress in such fields as machine learning, process control and engineering prediction. Swarm intelligence optimization algorithm is a biological heuristic method, which is widely used in solving optimization problems. The traditional swarm intelligence algorithm provides some new ideas for solving some practical problems, but it also exposes some shortcomings in some experiments. In recent years, many scholars have proposed many new types of intelligent optimization algorithms. This paper selects the more typical swarm intelligence algorithms at home and abroad in recent years, such as Bat Algorithm(BA), Grey Wolf Optimization Algorithm(GWO), Dragonfly Algorithm(DA), Whale Optimization Algorithm(WOA), Grasshopper Optimization Algorithm(GOA) and Sparrows Search Algorithm(SSA), and further compares the experimental performance of these algorithms and the development potential by 22 standard CEC test functions from the convergence speed and accuracy, stability and so on, and the refinement analysis is carried out to compare and analyze the relevant improvement methods. Finally, the characteristics of swarm intelligence optimization algorithm are summarized and its development potential is discussed.

    Related Articles | Metrics
    Survey of Small Object Detection Algorithms Based on Deep Learning
    LIU Yang, ZHAN Yinwei
    Computer Engineering and Applications    2021, 57 (2): 37-48.   DOI: 10.3778/j.issn.1002-8331.2009-0047
    Abstract915)      PDF(pc) (959KB)(917)       Save

    With the development of artificial intelligence technology, deep learning technology has been widely used in face recognition, pedestrian detection, unmanned driving and other fields. As one of the most basic and challenging problems in machine vision, object detection has attracted extensive attention in recent years. Aiming at the problem of object detection, especially small object detection, this paper summarizes the common data sets and performance evaluation metrics, and compares the characteristics, advantages and difficulties of various common data sets. At the same time, this paper systematically summarizes the common object detection methods and the challenges faced by small object detection. In addition, combing the latest work based on deep learning, this paper introduces the multi-scale and super-resolution small object detection methods in the highlight and presents the lightweight strategy and the performance of some lightweight models based on the object detection. Finally, this paper summarizes the characteristics, advantages and limitations of various methods, and looks at the future development direction of small object detection method based on deep learning.

    Related Articles | Metrics
    Survey of Network Traffic Forecast Based on Deep Learning
    KANG Mengxuan, SONG Junping, FAN Pengfei, GAO Bowen, ZHOU Xu, LI Zhuo
    Computer Engineering and Applications    2021, 57 (10): 1-9.   DOI: 10.3778/j.issn.1002-8331.2101-0402
    Abstract900)      PDF(pc) (711KB)(820)       Save

    Precisely predicting the trend of network traffic changes can help operators accurately predict network usage, correctly allocate and efficiently use network resources to meet the growing and diverse user needs. Taking the progress of deep learning algorithms in the field of network traffic prediction as a clue, this paper firstly elaborates the evaluation indicators of network traffic prediction and the current public network traffic data sets. Secondly, this paper specifically analyzes four deep learning methods commonly used in network traffic prediction:deep belief networks, convolutional neural network, recurrent neural network, and long short term memory network, and focuses on the integrated neural network models used in recent years for different problems. The characteristics and application scenarios of each model are summarized. Finally, the future development of network traffic forecast is prospected.

    Related Articles | Metrics
    Literature Survey on Research and Application of Bat Algorithm
    XU Degang, ZHAO Ping
    Computer Engineering and Applications    2019, 55 (15): 1-12.   DOI: 10.3778/j.issn.1002-8331.1903-0208
    Abstract862)      PDF(pc) (763KB)(473)       Save
    Bat Algorithm(BA) is a novel swarm intelligence optimization algorithm inspired by bats using ultrasonic echolocation foraging behavior. The paper introduces the basic principles of the bat algorithm, analyzes the performance influencing factors, discusses the improvement strategy of BA, and expounds the application and development of the bat algorithm in data mining, image processing, combinatorial optimization, etc. Finally, the future research direction of bat algorithm is prospected by combining the performance characteristics and application directions of the bat algorithm.
    Related Articles | Metrics
    Survey of Application of Deep Learning in Image Recognition
    ZHENG Yuanpan1,2, LI Guangyang1, LI Ye1
    Computer Engineering and Applications    2019, 55 (12): 20-36.   DOI: 10.3778/j.issn.1002-8331.1903-0031
    Abstract862)      PDF(pc) (1086KB)(1204)       Save
    As an important technical means in the field of image recognition, deep learning has broad application prospects. Carrying out image recognition technology research has important theoretical and practical significance for promoting the development of computer vision and artificial intelligence. The application of deep learning in image recognition gives a review. The origin of deep learning is introduced. Deep learning models such as deep belief network, convolutional neural network, cyclic neural network, generated confrontation network and capsule network are analyzed. The improved models of each deep learning model are compared and analyzed one by one. In this paper, the research results of deep learning in image recognition applications such as face recognition, medical image recognition and remote sensing image classification  are summarized. The existing researches are worth discussing. The development trend of deep learning in the field of image recognition is carried out. The discussion points out that the effective use of migration learning technology to identify small sample data, the use of unsupervised learning and semi-supervised learning to identify images, how to effectively identify video images and the theoretical significance of the model are further directions in this field.
    Related Articles | Metrics
    Attention Mechanism-Based CNN-LSTM Model and Its Application
    LI Mei1,2, NING Dejun1, GUO Jiacheng1,2
    Computer Engineering and Applications    2019, 55 (13): 20-27.   DOI: 10.3778/j.issn.1002-8331.1901-0246
    Abstract862)      PDF(pc) (914KB)(2219)       Save
    Time series have temporal property, and the characteristics of its short sequences are different in importance. Aiming at the characteristics of time series, a neural network prediction model based on Convolution Neural Network(CNN) and Long Short-Term Memory(LSTM) is proposed, which combines coarse and fine grain features to achieve accurate time series prediction. The model consists of two parts. CNN based on attention mechanism adds attention branch to standard CNN network to extract important fine-grained features. The back end is LSTM, which extracts the coarse-grained features of the hidden time series from fine-grained features. Experiments on real cogeneration heat load dataset demonstrate that the model is better than the autoregressive integrated moving average, support vector regression, CNN and LSTM models. Compared with the pre-determined method currently used by enterprises, the Mean Absolute Scaled Error(MASE) and Root Mean Square Error(RMSE) have been increased by 89.64% and 61.73% respectively.
    Related Articles | Metrics
    Review of Occlusion Face Recognition Method
    DONG Yanhua, ZHANG Shumei, ZHAO Junli
    Computer Engineering and Applications    2020, 56 (9): 1-12.   DOI: 10.3778/j.issn.1002-8331.2001-0029
    Abstract795)      PDF(pc) (689KB)(572)       Save

    In the process of image acquisition in real face recognition system, there are always uncertain factors such as illumination, attitude and occlusion, and the recognition effect of traditional face recognition method is not good. Therefore, effectively handling these problems and improving recognition efficiency are still difficulty in part of face recognition systems. This article reviews the traditional methods of face recognition, focusing on the processing methods of face occlusion, this paper makes a detailed review from three aspects:how to reconstruct the generation model of occlusion image, how to detect the discrimination model of occlusion position and robust feature extraction, their advantages, disadvantages and application occasions are compared, the existing problems and future research directions of occluded face recognition are summarized.

    Related Articles | Metrics
    Improved Algorithm of RetinaFace for Natural Scene Mask Wear Detection
    NIU Zuodong, QIN Tao, LI Handong, CHEN Jinjun
    Computer Engineering and Applications    2020, 56 (12): 1-7.   DOI: 10.3778/j.issn.1002-8331.2002-0402
    Abstract786)      PDF(pc) (1216KB)(816)       Save

    The 2019-nCoV can be transmitted through airborne droplets, aerosols, and other carriers. Correctly wearing a mask in public places can effectively prevent the infection of the virus. A face mask wearing detection method in a natural scene is proposed. The RetinaFace algorithm is improved to add the task of face mask wearing detection by optimizing the loss function. An improved self-attention mechanism is introduced into the feature pyramid network to enhance the expressive ability of the feature map. A data set containing 3, 000 pictures is created and manually annotated for network training. The experimental results show that the algorithm can effectively detect the wearing of masks, and has achieved good detection results in natural scene videos.

    Related Articles | Metrics
    Survey of Intelligent Question Answering Research Based on Knowledge Graph
    WANG Zhiyue, YU Qing, WANG Nan, WANG Yaoguo
    Computer Engineering and Applications    2020, 56 (23): 1-11.   DOI: 10.3778/j.issn.1002-8331.2004-0370
    Abstract785)      PDF(pc) (774KB)(1043)       Save

    The answer selection model based on knowledge graph has become one of the hottest directions at present. This paper introduces the implementation of answer selection model based on knowledge graph from four aspects of template method, semantic parsing, deep learning and knowledge graph embedding, sums up their advantages, disadvantages and unsolved problem. Combined with the development of artificial intelligence technology, this paper introduces intelligent question-answer system based on deep learning. This research is helpful for more researchers to devote themselves to the intelligent question-answer system and develops different kinds of intelligent question-answer system to improve the social intelligent information service.

    Related Articles | Metrics
    Overview of 3D Tracking Registration Technology in Augmented Reality
    HAN Yuren, LI Tiejun, YANG Dong
    Computer Engineering and Applications    2019, 55 (21): 26-35.   DOI: 10.3778/j.issn.1002-8331.1907-0283
    Abstract760)      PDF(pc) (908KB)(940)       Save
    The 3D tracking registration technology is an important key technology in the field of augmented reality. By tracking and locating images or objects in a real scene, the virtual objects are superimposed into the real scene according to the correct spatial perspective relationship. This paper comprehensively introduces the 3D tracking registration technology in augmented reality, elaborates the advantages and disadvantages of different tracking registration methods, introduces the current application status of different tracking registration methods, and then discusses the development trend and existence of tracking registration technology. The problem is further explored by the further study of 3D tracking registration technology.
    Related Articles | Metrics
    Survey of Multimodal Deep Learning
    SUN Yingying, JIA Zhentang, ZHU Haoyu
    Computer Engineering and Applications    2020, 56 (21): 1-10.   DOI: 10.3778/j.issn.1002-8331.2002-0342
    Abstract757)      PDF(pc) (769KB)(731)       Save

    Modal refers to the way people receive information, including hearing, vision, smell, touch and other ways. Multimodal learning refers to learning better feature representation by using the complementarity between multimodes and eliminating the redundancy between them. The purpose of multimodal learning is to build a model that can deal with and correlate information from multiple modes. It is a dynamic multidisciplinary field, with increasing importance and great potential. At present, the popular research direction is multimodal learning among image, video, audio and text. This paper focuses on the application of multimodality in audio-visual speech recognition, image and text emotion analysis, collaborative annotation and other practical levels, as well as the application in the core level of matching and classification, alignment representation learning, and gives an explanation for the core issues of multimodal learning:matching and classification, alignment representation learning. Finally, the common data sets in multimodal learning are introduced, and the development trend of multimodal learning in the future is prospected.

    Related Articles | Metrics
    Object Detection Based on Deep Learning and Attention Mechanism
    SUN Ping, HU Xudong, ZHANG Yongjun
    Computer Engineering and Applications    2019, 55 (17): 180-184.   DOI: 10.3778/j.issn.1002-8331.1902-0155
    Abstract740)      PDF(pc) (689KB)(374)       Save
    In the Convolution Neural Network(CNN), convolutional layers are translation-invariant, which weaken the localization performance of object detector. Actually, objects usually have distinct sub-region spatial characteristics and aspect ratio characteristics, but in prevalent two-stage object detection methods, these translation-variant feature components are rarely considered. In order to optimize the feature representations, the sub-region attention bank and aspect ratio attention bank are introduced into the two-stage object detection framework and generate the corresponding attention maps to refine the original ROI features.In addition, with the aid of the attention maps, the feature dimension can be greatly reduced.The experimental results show that object detectors equipped with attention module improve the accuracy and inference speed signi cantly.
    Related Articles | Metrics
    Survey of Data Fusion Based on Deep Learning
    ZHANG Hong, CHENG Chuanqi, XU Zhigang, LI Jianhua
    Computer Engineering and Applications    2020, 56 (24): 1-11.   DOI: 10.3778/j.issn.1002-8331.2007-0475
    Abstract724)      PDF(pc) (683KB)(1045)       Save

    As data fusion is the key to maximize the value of big data, while deep learning is a technical tool for mining deep characteristic information of data, data fusion based on deep learning can fully tap the potential value of big data, thus expanding the exploration and understanding of the world to a new depth and breadth. And this paper learns the advantages of deep learning in data fusion by reviewing the literature related to data fusion based on deep learning in recent years. The common data-fusion methods are classified, the advantages and disadvantages of which are pointed out. Analysis is conducted on data fusion method based on deep learning from three perspectives, namely the data fusion method extracted based on features of deep learning, data fusion method based on deep learning fusion and data fusion method based on the whole process of deep learning, and corresponding comparisons and summaries are conducted as well. This paper summarizes the whole document, discusses the difficulties in the application of deep learning in data fusion and the problems which require further research in the future.

    Related Articles | Metrics
    Deep Learning Approach and Its Application in Fault Diagnosis and Prognosis
    YU Ping, CAO Jie
    Computer Engineering and Applications    2020, 56 (3): 1-18.   DOI: 10.3778/j.issn.1002-8331.1910-0221
    Abstract723)      PDF(pc) (848KB)(1583)       Save
    In recent years, deep learning has been widely applied and has made remarkable progress in many fields because of its unique advantages and potential in feature extraction and pattern recognition. Its application in fault diagnosis and prognosis of complex industrial systems is an emerging field. This paper starts with an overview of deep learning, including deep learning methods-based application, platforms and useful tools. Five frequently-used deep learning models are introduced in this work, including Auto-Encoder(AE), Deep Belief Networks(DBN), Convolutional neural networks(CNN), Recurrent Neural Network(RNN) and Generative Adversarial Network(GAN). The application research based on deep learning in fault diagnosis and prognosis are systematically discussed in three aspects, research background, implementation process and research dynamics, and the current related literatures published in this field in recent years are reviewed. Problems, challenges and solutions of deep learning in the application of fault diagnosis and prognosis are discussed from the point view of research practice. The future research directions are also prospected at the end of this work.
    Related Articles | Metrics
    Research Progress of Multi-label Text Classification
    HAO Chao, QIU Hangping, SUN Yi, ZHANG Chaoran
    Computer Engineering and Applications    2021, 57 (10): 48-56.   DOI: 10.3778/j.issn.1002-8331.2101-0096
    Abstract722)      PDF(pc) (906KB)(688)       Save

    As a basic task in natural language processing, text classification has been studied in the 1950s. Now the single-label text classification algorithm has matured, but there is still a lot of improvement on multi-label text classification. Firstly, the basic concepts and basic processes of multi-label text classification are introduced, including data set acquisition, text preprocessing, model training and prediction results. Secondly, the methods of multi-label text classification are introduced. These methods are mainly divided into two categories:traditional machine learning methods and the methods based on deep learning. Traditional machine learning methods mainly include problem transformation methods and algorithm adaptation methods. The methods based on deep learning use various neural network models to handle multi-label text classification problems. According to the model structure, they are divided into multi-label text classification methods based on CNN structure, RNN structure and Transfomer structure. The data sets commonly used in multi-label text classification are summarized. Finally, the future development trend is summarized and analyzed.

    Related Articles | Metrics
    Research Status and Future Analysis of Capsule Neural Network
    HE Wenliang, ZHU Minling
    Computer Engineering and Applications    2021, 57 (3): 33-43.   DOI: 10.3778/j.issn.1002-8331.2009-0209
    Abstract697)      PDF(pc) (890KB)(549)       Save

    Nowadays, the rapid development of artificial intelligence technology has promoted the great progress of society. As an important part of artificial intelligence, deep learning has a very broad application prospect. In recent years, more and more experts and scholars have begun to study the related technologies in the field of deep learning. Two typical directions are natural language processing and computer vision. Among them, the development of computer vision strongly leads the progress in the field of deep learning. The application of convolutional neural network and a new neural network model in deep learning?capsule network and its dynamic routing algorithm are introduced, and their advantages and disadvantages are compared. Then, the application of capsule network is reviewed, and the application fields and advantages of capsule network are described in terms of image and text. Finally, the paper summarizes and looks forward to the possible improvement direction of capsule network.

    Related Articles | Metrics
    Review of Intent Detection Methods in Human-Machine Dialogue System
    LIU Jiao, LI Yanling, LIN Min
    Computer Engineering and Applications    2019, 55 (12): 1-7.   DOI: 10.3778/j.issn.1002-8331.1902-0129
    Abstract680)      PDF(pc) (700KB)(2688)       Save
    Spoken Language Understanding(SLU) is a vital part of the human-machine dialogue system, which includes an important sub-task called intent detection. The accuracy of intent detection is directly related to the performance of semantic slot filling, and it is helpful to the following research of the dialogue system. Considering the difficulty of intent detection in human-machine dialogue system, the traditional machine learning methods cannot understand the deep semantic information of user’s discourse. This paper mainly analyzes, compares and summarizes the deep learning methods applied in the research of intent detection in recent years, and further considers how to apply deep learning model to multi-intent detection task, so as to promote the research of multi-intent detection methods based on deep neural network.
    Related Articles | Metrics
    Overview of Human Pose Estimation Methods Based on Deep Learning
    DENG Yinong, LUO Jianxin, JIN Fenglin
    Computer Engineering and Applications    2019, 55 (19): 22-42.   DOI: 10.3778/j.issn.1002-8331.1906-0113
    Abstract625)      PDF(pc) (2837KB)(1590)       Save
    Human pose estimation is a research hot point in the field of computer vision. The human pose estimation methods based on deep learning get directly human pose information from two-dimensional image features through an appropriate neural network. This paper mainly follows the sequence from 2D to 3D human pose estimation, from the single-person detection to multi-person detection, from sparse node detection to dense model building, has systematically introduced the human post estimation methods in recent years based on deep learning to give a preliminary understanding of how to acquire the elements of human pose through deep learning, including the relative orientation and ratio scale of limb parts, the position coordinates and connection relations of joint points, and the information of the even more complex human skin model information. In the end, it summarizes the current research challenges and future hot point trends, which clearly present the development venation of this field for readers.
    Related Articles | Metrics
    Application of LeNet-5 Neural Network in Image Classification
    LIU Jinli, ZHANG Peiling
    Computer Engineering and Applications    2019, 55 (15): 32-37.   DOI: 10.3778/j.issn.1002-8331.1811-0286
    Abstract605)      PDF(pc) (788KB)(646)       Save
    Although the LeNet-5 Convolutional Neural Network(CNN) achieves good classification results in handwritten digit recognition, the classification accuracy is not high on datasets with complex texture features. In order to improve the accuracy of network class-ification on complex texture feature images, an improved LeNet-5 network structure is proposed. The idea of cross-connection is introduced to make full use of the low-level features of network extraction. The Inception V1 module is embedded in the LeNet-5 convolutional neural network to extract multi-scale features of the image. The output layer uses the softmax function to classify the image. Experimental results on the Cifar-10 and Fashion MNIST dataset show that the improved convolutional neural network has good classification ability on complex texture feature datasets.
    Related Articles | Metrics
    Review of Pre-training Models for Natural Language Processing
    YU Tongrui, JIN Ran, HAN Xiaozhen, LI Jiahui, YU Ting
    Computer Engineering and Applications    2020, 56 (23): 12-22.   DOI: 10.3778/j.issn.1002-8331.2006-0040
    Abstract602)      PDF(pc) (689KB)(597)       Save

    In recent years, deep learning technology has been advancing, pre-training technology for deep learning brings natural language processing into a new era. Pre-training model aims to how to make pre-trained model stay in good initial state and achieve better performances in subsequent downstream tasks. This paper firstly introduces pre-training technology and its development history. And then, this paper further classifies it into the following two types, namely probability-statistics-based traditional model and deep-learning-based new model, according to different features of pre-training models to conduct corresponding detailed introductions. Firstly, it briefly analyzes the characteristics and limitations of today’s pre-training models and highlights the existing deep-learning-based pre-training models. And based on their performances in downstream tasks, it gives necessary comparisons and assessments accordingly. Finally, it combs out a series of whole-new pre-training models with instructive significances,briefly describes corresponding feasible improvement mechanisms and the performance enhancements achieved in downstream tasks, summarizes the problems existing therein, as well as prospectes its development trend in near future.

    Related Articles | Metrics
    Review of Application Prospect of Deep Reinforcement Learning in Intelligent Manufacturing
    KONG Songtao, LIU Chichi, SHI Yong, XIE Yi, WANG Kun
    Computer Engineering and Applications    2021, 57 (2): 49-59.   DOI: 10.3778/j.issn.1002-8331.2008-0431
    Abstract601)      PDF(pc) (982KB)(1030)       Save

    As the latest development of machine learning, deep reinforcement learning has been shown in many application fields. The algorithm research and application research of deep reinforcement learning have produced many classical algorithms and typical application fields. The application of deep reinforcement learning in industrial manufacturing can realize high level control in complex environment. First of all, the research on deep reinforcement learning is summarized, and the basic principles of deep reinforcement learning are introduced, including deep learning and reinforcement learning. Then, the paper introduces the theoretical methods of the application of deep reinforcement learning algorithm. On this basis, it classifies the algorithms of deep reinforcement learning, respectively introduces the reinforcement learning algorithm based on value function and the reinforcement learning algorithm based on strategy gradient, and lists the main development results of these two kinds of algorithms, as well as other related research results. Then, the typical applications of deep reinforcement learning in industrial manufacturing are classified and analyzed. Finally, the existing problems and future development direction of deep reinforcement learning are discussed.

    Related Articles | Metrics
    Analysis on Application of Machine Learning in Stock Forecasting
    XU Haoran, XU Bo, XU Kewen
    Computer Engineering and Applications    2020, 56 (12): 19-24.   DOI: 10.3778/j.issn.1002-8331.2001-0353
    Abstract586)      PDF(pc) (892KB)(829)       Save

    It has always been regarded as the emphasis of research to reveal the operation law of stock market. In recent years, machine learning method has made good progress in stock forecasting, and it has shown unique advantages over traditional methods such as fundamental analysis and technical analysis. This paper focuses on collecting the key references in the field of stock prediction that uses machine learning methods in recent years, and analyzing as well as summarizing feature engineering, the application of machine learning algorithms and the main problems in stock prediction research. The characteristics and shortcomings of each algorithm in application are reviewed, and future development direction of this field is made a thorough analysis and forecasted from the aspects of transfer learning, feature engineering and deep learning model fusion.

    Related Articles | Metrics
    Survey on K-Means Clustering Algorithm
    YANG Junchuang, ZHAO Chao
    Computer Engineering and Applications    2019, 55 (23): 7-14.   DOI: 10.3778/j.issn.1002-8331.1908-0347
    Abstract583)      PDF(pc) (825KB)(778)       Save
    The K-Means algorithm is a partition-based algorithm in cluster analysis. With an unsupervised learning algorithm, its advantages of simple thinking, good effect and easy implementation are widely used in fields such as machine learning. But the K-Means algorithm also has certain limitations. For example, the K number of clusters in the algorithm is difficult to determine how to choose the initial cluster center, how to detect and remove outliers and the distance and similarity measure. This paper summarizes the improvement of K-Means algorithm from several aspects, and compares it with the classical K-Means algorithm. In addition, it analyzes the advantages and disadvantages of the improved algorithm, and points out the problems. Finally, the development direction and trend of K-Means algorithm are prospected.
    Related Articles | Metrics
    Application of Improved YOLOV3 Algorithm in Pedestrian Identification
    GE Wen, SHI Zhengwei
    Computer Engineering and Applications    2019, 55 (20): 128-133.   DOI: 10.3778/j.issn.1002-8331.1901-0318
    Abstract580)      PDF(pc) (851KB)(379)       Save
    In order to avoid the mutual occlusion of human and objects, the inaccurate detection of small targets, and the influence of complex illumination intensity on pedestrian detection, this paper proposes an improvement of multi-scale clustering convolutional neural network MK-YOLOV3 algorithm to realize the recognition and detection of images. The algorithm improves the YOLOV3. Firstly, the image features are extracted by simple clustering, and the corresponding feature maps are obtained. Then the [K]-means clustering algorithm is combined with the kernel function to determine the anchor position to achieve better clustering. Multi-scale fusion is performed on the shallow feature information of small targets to improve the detection effect of small targets. The simulation results verify that the algorithm has a great improvement on the accuracy and speed of small target recognition on VOC dataset, and has a higher recall rate and accuracy in video intelligence analysis.
    Related Articles | Metrics
    Research on Lightweight Convolutional Neural Network Technology
    BI Pengcheng, LUO Jianxin, CHEN Weiwei
    Computer Engineering and Applications    2019, 55 (16): 25-35.   DOI: 10.3778/j.issn.1002-8331.1903-0340
    Abstract568)      PDF(pc) (753KB)(884)       Save
    In order to better apply the convolutional neural network model to mobile and embedded devices, it is necessary to reduce the amount of model parameters and reduce computational complexity. Firstly, several popular solutions are briefly introduced. Next, six lightweight convolutional neural network models are elaborated, showing the computational complexity and parameter quantities of different network computing methods. The core building blocks of the model, the overall network structure and innovations are discussed. The classification accuracy of each network and conventional convolutional network on the ImageNet dataset is analyzed. Furthermore, comparing the techniques of lightening the weight of each network, the conclusion is drawn that the direct index is used instead of the indirect index when designing the model. At the same time, the importance of the residual structure to ensure the accuracy of the lightweight model is found. Finally, the development prospect of lightweight convolutional neural network is prospected.
    Related Articles | Metrics
    Survey of Application of Deep Convolution Neural Network in Image Aesthetic Evaluation
    WEN Kunzhe, WEI Yuke, DONG Xiaohua
    Computer Engineering and Applications    2019, 55 (15): 13-23.   DOI: 10.3778/j.issn.1002-8331.1901-0185
    Abstract563)      PDF(pc) (878KB)(675)       Save
    With the development of deep learning in recent years, image aesthetic evaluation has gradually become a new hot research topic. The application of deep convolution neural network in image aesthetic evaluation has successfully achieved considerable development results and has attracted wide attention. In order to solve the problems of incomplete literature survey and insufficient understanding of the development of this technology, this paper elaborates its development in detail from the perspectives of global perception and local perception, personalized query, combination of extracting handcrafted features and deep convolution neural network. The application of image aesthetics evaluation, image clipping and tool application are also discussed. The future work is prospected from the perspectives of fully combining multi-scene, skillfully using composition rules, and establishing aesthetic image data sets in advance.
    Related Articles | Metrics
    Overview of Image Denoising Methods Based on Deep Learning
    LIU Di, JIA Jinlu, ZHAO Yuqing, QIAN Yurong
    Computer Engineering and Applications    2021, 57 (7): 1-13.   DOI: 10.3778/j.issn.1002-8331.2011-0341
    Abstract559)      PDF(pc) (1139KB)(523)       Save

    Image denoising is a kind of technology that uses the context information of image sequence to remove noise and restore clear image. It is one of the important research contents in the field of computer vision. With the development of machine learning, deep learning has been widely used in the field of image denoising, and has become an effective solution for image denoising. Firstly, the deep learning image denoising method is analyzed. Secondly, the idea of image denoising method is analyzed in detail according to the network structure, and the advantages and disadvantages are summarized. Then, through the experimental results on DND, PolyU and other data sets, the performance of deep learning based image denoising methods is compared and analyzed. Finally, the key issues of image denoising research are summarized, and the future development trend of the research of this field is discussed.

    Related Articles | Metrics
    Survey of Vehicle Object Detection Algorithm in Computer Vision
    LI Mingxi, LIN Zhengkui, QU Yi
    Computer Engineering and Applications    2019, 55 (24): 20-28.   DOI: 10.3778/j.issn.1002-8331.1908-0408
    Abstract557)      PDF(pc) (815KB)(547)       Save
    Vehicle object detection based on computer vision is an important application field. In recent years, with the great progress of deep learning in image classification, vehicle object detection algorithm combining machine vision technology with deep learning method has gradually become the research priority and hotspot in this field. This paper introduces the tasks, difficulties and current development status of vehicle detection based on machine vision technology. Several?representative convolutional neural network in deep learning is also introduced, as well as two-stage and one-stage object detection algorithms derived from these networks. In the next place, it is followed by vehicle detection?and?the relevant data sets and evaluation criteria for the detection results. The existing problems and future development direction are finally discussed.
    Related Articles | Metrics
    Survey of Knowledge Reasoning Based on Neural Network
    ZHANG Zhongwei1,2, CAO Lei1, CHEN Xiliang1, KOU Dalei1,3, SONG Tianting2
    Computer Engineering and Applications    2019, 55 (12): 8-19.   DOI: 10.3778/j.issn.1002-8331.1901-0358
    Abstract554)      PDF(pc) (777KB)(2657)       Save
    Knowledge reasoning is an important means of knowledge graph completion and has always been one of the research hotspots in the field of knowledge graph. With the development of neural network, its applications in knowledge reasoning have been paid more and more attention in recent years. The knowledge reasoning methods based on neural network have not only stronger reasoning and generalization abilities, but also higher utilization rates of entities, attributes, relations and text information in the knowledge base. These methods are more effective in reasoning. The relevant concepts of knowledge graph and knowledge graph completion are introduced, the concepts and basic principles of knowledge reasoning are indicated, and then the latest research progresses of the technology of knowledge reasoning based on neural network are reviewed. The existing problems and development directions of knowledge reasoning in the aspect of theory, algorithm and application are summarized.
    Related Articles | Metrics
    Review of Development and Application of Artificial Neural Network Models
    ZHANG Chi, GUO Yuan, LI Ming
    Computer Engineering and Applications    2021, 57 (11): 57-69.   DOI: 10.3778/j.issn.1002-8331.2102-0256
    Abstract537)      PDF(pc) (781KB)(665)       Save

    Artificial neural networks are increasingly closely related to other subject areas. People solve problems in various fields by exploring and improving the layer structure of artificial neural networks. Based on the analysis of artificial neural networks related literature, this paper summarizes the history of artificial neural network growth and presents relevant principles of artificial neural networks based on the development of neural networks, including multilayer perceptron, back-propagation algorithm, convolutional neural network and recurrent neural network, explains the classic convolutional neural network model in the development of the convolutional neural network and the widely used variant network structure in the recurrent neural network, reviews the application of each artificial neural network algorithm in related fields, summarizes the possible direction of development of the artificial neural network.

    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
    Abstract537)      PDF(pc) (1134KB)(386)       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
    Survey of Breast Cancer Histopathology Image Classification Based on Deep Learning
    LI Hua, YANG Jianeng, LIU Feng, NAN Fangzhe, QIAN Yurong
    Computer Engineering and Applications    2020, 56 (13): 1-11.   DOI: 10.3778/j.issn.1002-8331.2001-0220
    Abstract532)      PDF(pc) (919KB)(879)       Save

    Accurate and efficient histopathological image classification of breast cancer is one of the important contents of computer-aided diagnosis. With the development of machine learning technology, deep learning has gradually become an effective method to classify breast cancer histopathological images. Firstly, the classification methods of breast cancer histopathological image and the existing problems are analyzed. Secondly, four relevant deep learning models are introduced, and the classification methods of breast cancer histopathological image based on deep learning are combed, and the performance of the existing models is compared and analyzed through experiments. Finally, the key issues of histopathological image classification of breast cancer are summarized and the future research trends are discussed.

    Related Articles | Metrics
    Research on Progress of Image Semantic Segmentation Based on Deep Learning
    LIANG Xinyu, LUO Chen, QUAN Jichuan, XIAO Kaihong, GAO Weijia
    Computer Engineering and Applications    2020, 56 (2): 18-28.   DOI: 10.3778/j.issn.1002-8331.1910-0300
    Abstract531)      PDF(pc) (813KB)(1175)       Save
    Since the FCN network was proposed in 2014, a series of deep learning architectures for image semantic segmentation such as SegNet and DeepLab have been proposed. Compared with the traditional methods, these architectures are better and faster, and can be applied to the segmentation of natural images. This paper focuses on the image semantic segmentation technology. The commonly used data sets and typical network architectures are analyzed. And a comprehensive study is conducted about the new progress since 2017. The current evaluation indicators are used to compare and analyze the semantic segmentation effects of the main models. Finally, the challenges faced by semantic segmentation technology and the possible development trends are prospected.
    Related Articles | Metrics
    Review of Research on Scene Text Recognition Technology
    WANG Deqing, Wushouer·Silamu, XU Miaomiao
    Computer Engineering and Applications    2020, 56 (18): 1-15.   DOI: 10.3778/j.issn.1002-8331.2004-0333
    Abstract527)      PDF(pc) (1900KB)(534)       Save

    The text detection and recognition technology is introduced in this paper. Firstly, the research background, application fields and technical difficulties of the natural scene character recognition technology are introduced. Secondly, the preprocessing technology and process of scene text recognition are introduced. Then it introduces the universal deep learning detection network based on deep learning, deep learning text detection network based on Uyghur and Chinese and English, deep learning network based on scene character recognition, end to end scene character detection and recognition deep learning network, this paper summarizes the structure features, advantages, limitations, application scenarios and implementation costs of various networks, and then makes a comprehensive analysis. Finally, it introduces the open data sets, the development trend and possible research direction of scene character recognition technology are also discussed.

    Related Articles | Metrics
    Improved YOLO v3 Algorithm and Its Application in Helmet Detection
    WANG Bing, LI Wenjing, TANG Huan
    Computer Engineering and Applications    2020, 56 (9): 33-40.   DOI: 10.3778/j.issn.1002-8331.1912-0267
    Abstract525)      PDF(pc) (1076KB)(519)       Save

    YOLO v3 object detection algorithm has been widely used in industry due to its high speed and high precision. However, there is a problem that the object function is not consistent with the evaluation index. To solve this problem, improved YOLO v3 object detection algorithm is proposed. This algorithm improves the GIoU calculation method and combines it with the YOLO v3 algorithm’s objective function to design a new objective function to achieve IoU local optimization as the local optimization of the objective function. The test results of public dataset VOC2007 and helmet wearing dataset show that the mAP-50 of the improved YOLO v3 increased by 2.07% and 2.05%, respectively, compared with the YOLO v3 algorithm.

    Related Articles | Metrics
    Survey of Data Integrity Verification Technology Based on Provable Data Possession
    YUAN Ying, ZHU Hongliang, CHEN Yuling, OUYANG Zhi, XIN Yang, YANG Yixian
    Computer Engineering and Applications    2019, 55 (18): 1-7.   DOI: 10.3778/j.issn.1002-8331.1905-0073
    Abstract510)      PDF(pc) (620KB)(883)       Save
    In the cloud storage environment, to ensure the integrity and availability of user’s data, users need to verify the integrity of data stored in the cloud server. There are two main data integrity verification mechanisms: Provable Data Possession(PDP) and Proof of Retrievability(POR). This paper focuses on the PDP-based cloud storage data integrity verification mechanism. The characteristics of the PDP verification mechanism are combined to classify the PDP scheme and the techniques used by each category are summarized. According to the classification, the research status of PDP scheme is described, and the typical schemes are compared and analyzed in terms of dynamic verification, batch auditing and computational overhead. The future development direction of cloud storage data integrity verification mechanism based on PDP is discussed.
    Related Articles | Metrics
    Network Unknown Attack Detection with Deep Learning
    DI Chong, LI Tong
    Computer Engineering and Applications    2020, 56 (22): 109-116.   DOI: 10.3778/j.issn.1002-8331.2003-0353
    Abstract510)      PDF(pc) (753KB)(331)       Save

    A deep learning-based method for network anomaly detection is proposed to discriminate unknown attacks for an intrusion detection system. A confidence-based neural network is adopted to adaptively distinguish the traffic?information of given behaviors and?that of unknown attacks.?The proposed model?is trained to?assign a higher?confidence value to a piece of?traffic?information?from a known behavior and?a lower?confidence value to that?from an?unknown attack. Moreover, an adaptive loss balance strategy and a learning automata-based dynamic regularization strategy are designed?to improve the performance of the model. The proposed model is evaluated in benchmark datasets UNSW-NB15 and CICIDS 2017. Compared with traditional models, the simulation results indicate that the proposed model can detect the unknown attack effectively while preserving an advantageous classification effect for traffic from known attacks.

    Related Articles | Metrics
    Survey of Research on Network Security Situation Awareness
    SHI Leyi, LIU Jia, LIU Yihao, ZHU Hongqiang, DUAN Pengfei
    Computer Engineering and Applications    2019, 55 (24): 1-9.   DOI: 10.3778/j.issn.1002-8331.1906-0349
    Abstract509)      PDF(pc) (995KB)(936)       Save
    Different from traditional security measures, network security situation awareness can identify the behavior of various activities in the network and conduct intent understanding and impact assessment from a macro perspective so as to provide reasonable decision support. It has great significance in improving network monitoring capabilities, emergency response capabilities, and predicting the development trend of network security. This paper first separately generalizes the definitions of situation awareness and network security situation awareness, and then sorts out the classical and newly developed system models. It introduces the key technologies of network security situation awareness, which is mainly divided into hierarchical analysis, machine learning, immune system and game theory. Then the latest application of network security situation awareness in Internet, industrial control network and Internet of Things are explained. It summarizes and forecasts the future development trends and problems that need to be solved.
    Related Articles | Metrics