检索结果-南通市图书馆

Traffic Sign Recognition Using an Attentive Context Region-Based Detection Framework

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2021年第6期30卷 1080-1086页

作者： LIU Zhigang DU Juan TIAN Feng WEN Jiazheng School of Computer and Information Technology Northeast Petroleum University Postdoctoral Workstation on Applied Technology Research Institute Northeast Petroleum University Department of Computer Science for Artificial Intelligence University of Nottingham Ningbo

Accurate small traffic sign recognition is more important for the safety of intelligent transportation systems. A recognition framework named attentive context region-based detection framework(AC-RDF) is proposed in this paper. We construct the attentive context feature for the recognition of small traffic signs,which combines the target information and the contextual information by the concatenation operation following a pointwise convolutional layer. The proposed attentive context feature exploits the surrounding information for a given object proposal. Next, we propose a novel attentive loss function to replace the original crossentropy function. It distinguishes hard negative samples from easy positive ones in the total loss, allows the proposed framework to obtain enough training, and further improve the recognition accuracy. The proposed method is evaluated on the challenging Tsinghua-Tencent 100 K dataset. The experimental results indicate that the attentive context region-based detection framework is superior at detecting small traffic signs and achieves stateof-the-art performance compared with other methods.

关键词： Image recognition Computer vision and image processing techniques Traffic engineering computing neural nets Small traffic sign Attentive context Attentive loss Hard negative samples

Video Stabilization via Prediction with Time-Series Network and Image Inpainting with Pyramid Fusion

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2021年第6期30卷 1103-1110页

作者： CHENG Keyang LI Shichao RONG Lan WANG Wenshan SHI Wenxi ZHAN Yongzhao School of Computer Science and Communication Engineering Jiangsu University Jiangsu Province Big Data Ubiquitous Perception and Intelligent Agricultural Application Engineering Research Center Cyber Space Security Academy of Jiangsu University National Engineering Laboratory for Public Safety Risk Perception and Control by Big Data Xinjiang Lianhaichuangzhi Information Technology Co. Ltd.

Due to the poor filling effect of the video image defect commonly used in the video stabilization field, the video is seemed still unstable after the image stabilization process, which seriously affects the visual effect. To solve this problem, we improve a video stabilization method based on time-series network prediction and pyramid fusion restoration is proposed to optimize the visual effect after image *** flow of the proposed method is as follows: First,it is adaptive to determine whether the defect of the corresponding frame at the current time needs padding inpainting. Then, for the frame that needs to be inpainting,the frames generated before the current moment are sent to the model combining the convolutional neural networks and the gate recurrent unit to predict the part to be filled. Next the current defect image and the complete image to be filled are brought into the Laplacian pyramid reconstruction, and the improved weighted optimal suture is introduced for splicing during the fusion. Finally, the video frame is cut after reconstruction. The method is tested on a data set composed of videos commonly used in the field of video stabilization. The experimental results show that the average peak signal to noise ratio of the method is 2 to 5 d B higher than that of the comparison algorithm, and the average structural similarity index is improved by about 2% to 7% compared with the contrast algorithm.

关键词： Optical, image and video signal processing Computer vision and image processing techniques Video signal processing neural nets Video stabilization Video inpainting Time series network Multi-resolution fusion Optimal seam

Feature Fusion Based Hand Gesture Recognition Method for Automotive Interfaces

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2020年第6期29卷 1153-1164页

作者： XU Qianyi QIN Guihe SUN Minghui YAN Jie JIANG Huiming ZHANG Zhonghan College of Computer Science and Technology Jilin University Key Laboratory of Symbolic Computing and Knowledge Engineering Ministry of Education Jilin University College of Software Jilin University

Hand gesture recognition on the depth videos is a promising approach for automotive interfaces because it is less sensitive to light variation and more accurate than other traditional ***,video gestures recognition is still a challenging task since lots of interferences are induced by the uncorrelated gesture *** that if the displays are more relevant,the results will more accurate,so ResNext,a kind of compact and efficient neural network,is firstly used as feature extractor,then an improved weighted frame unification method is adopted to obtain the key frame samples,finally the Discriminant correlation analysis(DCA) is employed to fuse features for static data and dynamic data after conducting Feature embedding branch(FEB) on static *** public dataset named Depth based gesture recognition database(DGRD) is used in this paper,but the dataset is a little small and the class distribution is largely imbalance,and we find the performance of ResNext degrades badly in the condition of imbalance problem although it achieves excellent result at sufficient training *** order to conquer the disadvantages of limited dataset,a special loss function scheme combining the soft max loss and dice loss is *** of the algorithm performances in comparison with other state-of-the-art methods indicates that the proposed method is more practical for gesture recognition and may be widely adopted by automotive interfaces.

关键词： Feature fusion based hand gesture recognition method depth videos static data Other topics in statistics improved weighted frame unification method compact network Image recognition neural nets Feature embedding branch Depth based gesture recognition database automotive interfaces video gestures recognition Optical, image and video signal processing Computer vision and image processing techniques feature extraction feature extractor uncorrelated gesture factors efficient neural network key frame samples dynamic data gesture recognition learning (artificial intelligence) Knowledge engineering techniques

Non-profiled Deep-Learning-Based Power Analysis of the SM4 and DES Algorithms

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2021年第3期30卷 500-507页

作者： MA Xiangliang LI Bing WANG Hong WU Di ZHANG Lizhen HUANG Kezhen DUAN Xiaoyi School of Integrated Circuits Tsinghua University National Research Center for Information Technology Security China Cybersecurity Review Technology and Certification Center Trusted Computing and Information Assurance Laboratory Institute of Software Chinese Academy of Sciences University of Chinese Academy of Sciences Department of Electronics and Information Engineering Beijing Electronic Science and Technology Institute

Power analysis methods are commonly used for evaluating the security of cryptographic *** are characteristically low-cost and display a high success rate and the ability to obtain important device information, e.g., keys. Given the current wide application of deep-learning technology, there is a growing tendency to incorporate power-analysis technology in *** study investigates non-profiled deep-learning-based power analysis. The labels used in this attack are uncertain,and the attack conditions required are greatly reduced. We choose the Recurrent neural network(RNN), multilayer perceptron, and convolutional neural network algorithms,which use the same network structure, to recover the keys for the SM4 software and DES hardware *** propose combining the RNN algorithm with power analysis, and validate the benefits experimentally. The experimental results show that they all successfully recover the correct key for the SM4 software implementation,although the RNN algorithm by itself achieves a better effect. This conclusion also applies to attacks on the DES hardware implementation but is limited to labels based on the bit model.

关键词： Cryptography Data security neural nets Non-profiled deep learning Power analysis SM4 DES Key recovery

Quantum Contextuality for Training neural Networks

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2020年第6期29卷 1178-1184页

作者： ZHANG Junwei LI Zhao School of Computer Science and Technology Tianjin University Search Division Alibaba Group

In the training process of neural networks(NNs),the selection of hyper-parameters is crucial,which determines the final training effect of the *** them,the Learning rate decay(LRD) can improve the learning speed and accuracy;the Weight decay(WD)improves the over-fitting to varying ***,the decay methods still have problems such as hysteresis and stiffness of parameter adjustment,so that the final model will be *** on the Quantum contextuality(QC)theory,we propose a Quantum contextuality constraint(QCC) to constrain the weights of nodes in NNs to further improve the training *** the simplest classification model,we combine this constraint with different methods of LRD and WD to verify that QCC can further improve the training effect on the decay *** performance of the experiments shows that QCC can significantly improve the convergence and accuracy of the model.

关键词： decay method Other topics in statistics hysteresis quantum contextuality constraint WD neural nets weight decay stiffness Data handling techniques QCC parameter adjustment simplest classification model learning speed learning rate decay quantum contextuality theory training process LRD training neural networks final training effect neural computing techniques learning (artificial intelligence) NN pattern classification Knowledge engineering techniques hyper-parameters

Targeted Adversarial Examples Generating Method Based on cVAE in Black Box Settings

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2021年第5期30卷 866-875页

作者： YU Tingyue WANG Shen ZHANG Chunrui WANG Zhenbang LI Yetian YU Xiangzhan School of Cyberspace Science Harbin Institute of Technology Institute of Computer Application China Academy of Engineering Physics State Grid Heilongjiang Power Co. Ltd.

In recent years, adversarial examples has become one of the most important security threats in deep learning applications. For testing the security of deep learning models in adversarial environment,many researches focus on generating adversarial examples quickly and efficiently. In order to solve the problems of existing generative adversarial networks based methods which can not effectively generate the targeted adversarial examples in black box settings, and to improve the temporal performance of gradient-based generating methods,an adversarial examples generating method based on conditional Variational autoencoder(cVAE) is proposed in this paper, where a cVAE is designed elaborately to generate adversarial examples without most of the detailed information about the attacked deep learning models, of which the output can be controlled arbitrarily by these crafted inputs, used to test the robustness of deep learning models against adversarial examples. The experimental results show that the proposed method can achieve a comparable attack success rate and a better temporal performance than the existing gradient-based generating methods in black box environment.

关键词： Data security neural nets Deep learning Image recognition Adversarial examples Conditional variational autoencoder

Research and Application of Machine Learning in Automatic Program Generation

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2020年第6期29卷 1001-1015页

作者： ZHANG Xiaojiang JIANG Ying Yunnan Key Laboratory of Computer Technology Application Kunming University of Science and Technology Faculty of Information Engineering and Automation Kunming University of Science and Technology

With the development of artificial intelligence,machine learning has been applied in more and more *** order to improve the quality and efficiency of software,automatic program generation is becoming a research *** recent years,machine learning has also been gradually applied in automatic program *** trees,language models,and cyclic neural networks have been applied in code generation,code completion and code knowledge *** efficiency of software development has been improved to a certain extent using machine *** at the automatic program generation,this paper analyzes and summarizes the models of machine learning,the modifications involved in the models and the application *** research direction is discussed from the aspects of programmer behavior and automatic program generation of machine learning.

关键词： code generation Combinatorial mathematics neural nets decision trees machine learning code completion automatic programming language models Data handling techniques data mining cyclic neural networks Programming support software engineering automatic program generation Software engineering techniques programmer behavior code knowledge mining neural computing techniques learning (artificial intelligence) Knowledge engineering techniques

A Quaternion Two-Stream R-CNN Network for Pixel-Level Color Image Splicing Localization

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2021年第6期30卷 1069-1079页

作者： CHEN Beijing JU Xingwang GAO Ye WANG Jinwei Engineering Research Center of Digital Forensics Ministry of Education Nanjing University of Information Science &Technology School of Computer Nanjing University of Information Science & Technology Jiangsu Collaborative Innovation Center of Atmospheric Environment and Equipment Technology (CICAEET) Nanjing University of Information Science & Technology

Recently, Zhou et al. designed a twostream faster Region-Convolutional neural networks(RCNN) model RGB-N for color image splicing localization in CVPR2018. However, the RGB-N locates spliced regions only at block-level and ignores the entirety and inherent correlation of three channels. Therefore, an improved quaternion two-stream R-CNN model is proposed to solve these drawbacks: a mask branch combining fully convolutional network and condition random field is added for locating spliced regions at pixel-level; quaternion representation of color images is used to process color spliced images in a holistical way. In addition, feature pyramid network based on quaternion residual network is considered to extract multi-scale features for color spliced images; attention region proposal network is combined with attention mechanism and is designed to pay more attention to the spliced regions; a high-pass filter designed for image splicing detection specifically is adopted to replace steganalysis rich model filter in the RGB-N to obtain noise input for the noise stream. Experimental results on a new synthetic dataset and three standard forgery datasets demonstrate that the proposed method is superior to the existing methods in the abilities of localization, generalization, and robustness.

关键词： Optical, image and video signal processing Computer vision and image processing techniques neural nets Color image Quaternion Image splicing localization R-CNN ResNet

Recurrent neural Networks for Computing the Moore-Penrose Inverse with Momentum Learning

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2020年第6期29卷 1039-1045页

作者： ZHANG Naimin ZHANG Ting Department of Mathematics Wenzhou University

We are concerned with a kind of iterative method for computing the Moore-Penrose inverse,which can be considered as a discrete-time form of recurrent neural *** study the momentum learning scheme of the method and discuss its semi-convergence when computing the Moore-Penrose inverse of a rankdeficient *** prove the semi-convergence for our new acceleration algorithm and obtain the optimal momentum factor which makes the fastest *** tests demonstrate the effectiveness of our new acceleration algorithm.

关键词： Moore-Penrose inverse Optimisation techniques recurrent neural nets iterative method neural nets momentum learning scheme gradient methods convergence of numerical methods discrete-time form matrix algebra Numerical approximation and analysis semiconvergence convergence recurrent neural networks Interpolation and function approximation (numerical analysis) Discrete control systems iterative methods optimal momentum factor neural computing techniques backpropagation learning (artificial intelligence) discrete time systems Knowledge engineering techniques