检索结果-内蒙古大学图书馆

2019 ACM Turing Celebration Conference - China, ACM TURC 2019

作者： Fang, Zheng Cao, Tieyong Zhang, Xiongwei Zheng, Changyan Deng, Xiaotong Institute of Command and Control Engineering PLA Army Engineering University Nanjing Jiangsu China

ISBN: (纸本)9781450371582

Camouflage greatly reduces the probability of the target being discovered, detecting the camouflage object from background is very hard for former object detection methods because they are designed to detect the object in an ideal environment. It brings difficulties to subsequent applications such as battlefield information acquisition, precision guided weapons and object tracking. To deal with the problem, in this paper, a new complete camouflage people detection dataset is firstly constructed, then we propose strong semantic dilation network (SSDN), which is specially designed to detect camouflage people in an end-to-end architecture. SSDN makes full use of semantic information in CNN and dilated convolutions are also added to enlarge the receptive field to find camouflage people. Experiments demonstrate that SSDN perform well in camouflage people detection dataset compared with other object detection methods. © 2019 Copyright held by the owner/author(s).

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Dual Attribute Adversarial Camouflage toward camouflaged object detection

引用

Defence Technology（防务技术） 2023年第4期22卷 166-175页

作者： Yang Wang Zheng Fang Yun-fei Zheng Zhen Yang Wen Tong Tie-yong Cao Institute of Command and Control Engineering Peoples Liberation Army Engineering UniversityNanjing210007China The Army Artillery and Defense Academy of PLA Nanjing210007China PLA University of Foreign Language Luoyang471000China

The object detectors can precisely detect the camouflaged object beyond human *** investigations reveal that the CNNs-based(Convolution Neural Networks)detectors are vulnerable to adversarial *** works can fool detectors by crafting the adversarial camouflage attached to the object,leading to wrong *** is hard for military operations to utilize the existing adversarial camouflage due to its conspicuous *** by this,this paper proposes the Dual Attribute Adversarial Camouflage(DAAC)for evading the detection by both detectors and *** DAAC includes two steps:(1)Extracting features from a specific type of scene to generate individual soldier digital camouflage;(2)Attaching the adversarial patch with scene features constraint to the individual soldier digital camouflage to generate the adversarial attribute of *** visual effects of the individual soldier digital camouflage and the adversarial patch will be improved after integrating with the scene *** results show that objects camouflaged by DAAC are well integrated with background and achieve visual concealment while remaining effective in fooling object detectors,thus evading the detections by both detectors and humans in the digital *** work can serve as the reference for crafting the adversarial camouflage in the physical world.

关键词： Adversarial camouflage Digital camouflage generation Visual concealment Object detection Adversarial patch

来源：评论

学校读者我要写书评

暂无评论

AndrOpGAN: An Opcode GAN for Android Malware Obfuscations 1

引用

3rd International Conference on Machine Learning for Cyber Security, ML4CS 2020

作者： Zhang, Xuetao Wang, Jinshuang Sun, Meng Feng, Yao Institute of Command Control Engineering Army Engineering University of PLA NanjingJiangsu210007 China

ISBN: (数字)9783030622237

ISBN: (纸本)9783030622220

With the rapid development of Android platform, the number of Android malwares is growing rapidly. Due to the limitations of traditional static and runtime Android malware analysis methods, machine learning based approaches are widely adopted recently. Whereas, evading methods are also emerging, e.g. data set pollution, feature modification. Current feature modifications are mainly based on high-level features such as API calls or sensitive permissions. Our contribution is to show it is also feasible to deceive the detectors by modifying underlying features. Through this confusion, detector deceiving can be achieved. An Android malware opcode distribution feature modification system AndrOpGAN was proposed. To adjust the opcode distribution of malware, Deep Convolution Generative Adversarial Networks (DCGAN) was proposed to generate opcodes distribution features, and opcodes would be inserted through an Opcode Frequency Optimal Adjustment algorithm (OFOA). OFOA module can keep the APK running normally after insertion with a low modification cost. Test results against four detectors show that more than 99% APKs processed by AndrOpGAN could bypass detections successfully. Test results against VirusTotal shows that, the number of successful detection engines decreased 20%–44%. AndrOpGAN validates the feasibility of such attacks based on underlying feature modifications and provides a prototype system for researchers to improve detector’s performance. © 2020, Springer Nature Switzerland AG.

关键词： Android malware

来源：评论

学校读者我要写书评

暂无评论

Distortion Evaluation of EMP Sensors Using Associated-Hermite Functions

引用

Computers, Materials & Continua 2023年第1期74卷 1093-1105页

作者： Rupo Ma Siping Gao College of Command and Control Engineering Army Engineering UniversityNanjing210007China Department of Computer Information and Cyber Security Jiangsu Police InstituteNanjing210031China Department of Electrical and Computer Engineering National University of Singapore119260Singapore

Electromagnetic pulse(EMP)is a kind of transient electromagnetic phenomenon with short rise time of the leading edge and wide spectrum,which usually disrupts communications and damages electronic equipment and *** is challenging for an EMP sensor to measure a wideband electromagnetic pulse without distortion for the whole ***,analyzing the distortion of EMP measurement is crucial to evaluating the sensor distortion characteristics and correcting the measurement *** fidelity is usually employed to evaluate the distortion of an ***,this metric depends on specific signal waveforms,thus is unsuitable for evaluating and analyzing the distortion of EMP *** this paper,an associated-hermite-function based distortion analysis method including system transfer matrices and distortion rates is proposed,which is general and independent from individual *** system transfer matrix and distortion rate can be straightforwardly calculated by the signal orthogonal transformation coefficients using associated-hermite *** of a sensor *** is then visualized via the system transfer matrix,which is convenient in quantitative analysis of the *** of a current probe,a coaxial pulse voltage probe and a B-field sensor were performed,based on which the feasibility and effectiveness of the proposed distortion analysis method is successfully verified.

关键词： Electromagnetic pulse associated-hermite functions system transfer matrix distortion rate

来源：评论

学校读者我要写书评

暂无评论

A Novel Subcharacter-enhanced Hybrid Network for Chinese Text Classification

A Novel Subcharacter-enhanced Hybrid Network for Chinese Tex...

引用

2023 International Conference on Frontiers of Robotics and Software engineering, FRSE 2023

作者： Chen, Lei Wang, Yang Zheng, Yun Fei Yang, Zhen Cao, Tieyong Institute of Command-and-Control Engineering The Army Engineering University of Pla The Army Artillery and Denfense Academy of Pla Nanjing China

ISBN: (纸本)9798350301113

Chinese text classification aimed to set pre-defined labels to given documents, and it is a very important research field. To deal with this problem, a novel subcharacter-enhanced hybrid network model called SHN is proposed. The proposed SHN can hierarchically extract the text semantic features. Firstly, our model decomposes Chinese words by Cangjie-code into subcharacters. Secondly, both word embeddings and subcharacters n-grams vectors are used as the input part of SHN, which is aimed to fuse both internal structure information and semantic information. Thirdly, SHN makes use of a bidirectional GRU module that contains word-level attention, to obtains semantic information from the sentences in document. Finally, a dynamic convolution neural network module is used by SHN, which aimed to extract high level features. Experiments carried out on two datasets show that SHN can obtain better results than conventional methods and is effective. © 2023 IEEE.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Research on Indoor Positioning of Multi-source Information Fusion Based on Improved Particle Filter 22

Research on Indoor Positioning of Multi-source Information F...

引用

6th International Conference on Computer Science and Application engineering, CSAE 2022

作者： Zhou, Chulin Chen, Shiyou Chen, Jingdong Department of Command and Control Systems Wuhan Institute of Digital Engineering Wuhan China Department of Information and Electronics Wuhan Institute of Digital Engineering Wuhan China

ISBN: (纸本)9781450396004

In the information age, people are increasingly demanding location-based services. The traditional indoor positioning based on a single signal source has low accuracy and poor stability. On the one hand, the occlusion and echo interference in the indoor environment are serious, especially in the case of crowded environment and non-line-of-sight propagation, the UWB (Ultra Wide band) positioning accuracy will be greatly reduced. On the other hand, IMU (inertial Measurement Unit) can provide an accurate inertial navigation solution in a short time but its positioning error increases fast with time due to the cumulative error of accelerometer measurement. Therefore, in the process of pedestrian walking, we use the key information obtained by PDR (Pedestrian Dead Reckoning) to establish and update the real-time motion model of the target and calculate the prior information in the process of fusion filtering. Then, the initial position of the target is obtained through the UWB positioning solution, and the PDR positioning trajectory is corrected as the observation information in the fusion filtering process. Finally, the improved differential particle filter algorithm is used to fuse the above UWB positioning results and PDR positioning results, and make up for the advantages and disadvantages of the two positioning to improve the accuracy and stability of fusion positioning. © 2022 Association for Computing Machinery.

关键词： Location based services

来源：评论

学校读者我要写书评

暂无评论

Cross-modal entity resolution based on co-attentional generative adversarial network 19

Cross-modal entity resolution based on co-attentional genera...

引用

4th International Conference on Multimedia Systems and Signal Processing, ICMSSP 2019

作者： Lv, Guojun Cao, Jianjun Chang, Chen Zheng, Qibin Institute of Command and Control Engineering Army Engineering University Nanjing China Sixty-third Research Institute National University of Defense Technology Nanjing China

ISBN: (纸本)9781450371711

Cross-modal entity resolution aims to find semantically similar items from objects of different modalities(e.g. image and text). The core way to solve the problem is to construct a shared space where multi-modal examples can be represented uniformly. In this paper, we propose a novel Co-Attentional Generative Adversarial Network(CAGAN) method for solving cross-modal entity resolution, which seeks an effective space based on co-attention mechanism and adversarial learning. The generative adversarial network that we design contains two parts, Generator and Discriminator, the generator aims to generate a shared space through intra-modal loss and inter-modal loss, while discriminator is a classifier which tries to discriminate the modalities based on the generated representation. In order to eliminate the imbalance of information between modalities, generate more consistent representation and accelerate the convergence speed of the network, co-attention mechanism is introduced into the network. Experimental results performed on two cross-modal datasets demonstrated the outstanding performance of the proposed method for cross-modal entity resolution. © 2019 Association for Computing Machinery.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Monaural speech enhancement using U-net fused with multi-head self-attention

引用

Chinese Journal of Acoustics 2023年第1期42卷 98-118页

作者： FAN Junyi YANG Jibin ZHANG Xiongwei ZHENG Changyan Graduate School Army Engineering UniversityNanjing 210007 College of Command and Control Engineering Army Engineering UniversityNanjing 210007 Department of Test Control High-tech InstituteQingzhou 262500

Under low signal-to-noise ratio(SNR)and burst noise conditions,the speech enhancement effect of existing deep learning network models is not *** contrast,humans can exploit the long-term correlation of speech to form an integrated perception of different speech ***,describing the long-term dependencies of speech can help improve the enhancement performance under low SNR and burst noise *** by this feature,a time domain end-to-end monaural speech enhancement model TU-net that fuses the multi-head self-attention mechanism and U-net deep network is *** TU-net model adopts the codec layer structure of U-net to achieve multi-scale feature *** introduces the dual-path Transformer module using the multi-head self-attention mechanism to calculate the speech mask and better model long-term *** TU-net model is trained with a weighted sum loss function in the time,time-frequency,and perceptual *** experiments are carried out and the results show that with maintaining relatively fewer network model parameters,TU-net outperforms other similar monaural enhancement network models in several evaluation metrics such as perceptual evaluation of speech quality(PESQ),short-time objective intelligibility(STOI)and SNR gain under low SNR and burst noise conditions.

关键词： network speech noise

来源：评论

学校读者我要写书评

暂无评论

Modeling and Analysis of Contact Detumbling by Using Flexible Brush

Modeling and Analysis of Contact Detumbling by Using Flexibl...

引用

2019 IEEE International Conference on Unmanned Systems, ICUS 2019

作者： Cheng, Wenhua Li, Zhi He, Yingzi Space Engineering University Department of Graduate Management Beijing China Space Engineering University Department of Space Command Beijing China Beijing Institute of Control Engineering Space Intelligent Control Laboratory Beijing China

ISBN: (纸本)9781728137926

The prerequisite for capturing a tumbling object is that the angular rate is within a certain span. Within this paper, the feasibility of detumbling a target with contact force by using flexible brush is studied. A generic detumbling model is established and the detumbling process is analysed. The contact force model, including the friction, is derived based on mechanics of materials. With these models, the detumbling process is simulated and the DE (Detumbling-Effect) relationships with each factor are researched. © 2019 IEEE.

关键词： contact detumbling contact dynamic flexible brush on-orbit service

来源：评论

学校读者我要写书评

暂无评论

A speech reconstruction algorithm via iteratively reweighted 2 minimization for MFCC codec 27

A speech reconstruction algorithm via iteratively reweighted...

引用

27th European Signal Processing Conference, EUSIPCO 2019

作者： Min, Gang Zhang, Xiongwei Liu, Xiangyang Zhang, Changqing Chen, Yanpu Institute of Information and Communication National University of Defense Technology Xi'an China Institute of Command and Control Army Engineering University Nanjing China Institute of Information Engineering Xijing University Xi'an China

ISBN: (纸本)9789082797039

This paper presents an effective method to address the inverse problem of Mel-frequency cepstral analysis, and describes how to reconstruct the speech waveforms from Mel-frequency cepstral coefficients (MFCCs) directly. To exploit the sparse characteristics of speech in the frequency domain, an iteratively reweighted 2 minimization method is proposed to cope with the under-determined nature of the reconstruction problem. The lost phase information during Mel-frequency cepstral analysis procedure is recovered by the inverse short-time Fourier transform magnitude algorithm. Experiments are conducted over the TIMIT database and evaluated by several different kinds of measures. Experimental results demonstrate that the proposed method recovers speech with high articulation and intelligibility. Specifically, it sounds very close to the original speech when using the high-resolution MFCCs, the average STOI, PESQ score reaches 93% and 4.0, respectively. This method could be easily used for MFCC codec at low bit rate. © 2019 IEEE

关键词： Inverse problems

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：