检索结果-内蒙古大学图书馆

Chinese Automation Congress (CAC)

作者： Hua Gong Hui Li Ke Xu Yong Zhang College of Science Shengyang Ligong University Shenyang China School of Automation and Electrical Engineering Shengyang Ligong University Shenyang China Science and Technology on Electro-Optical Information Security Control Laboratory Tianjing China

Deep learning has gradually become the mainstream object detection algorithm because of its powerful feature extraction ability and adaptive ability. However, how to guarantee the accuracy and speed is still a huge challenge in the field of object detection. This paper proposes an improved YOLOv3-tiny for object detection based on the idea of feature fusion. YOLOv3-tiny is chosen as the basic network framework to ensure the identification speed. In order to improve the poor detection accuracy of YOLOv3-tiny network, feature fusion is carried out based on Feature Pyramid Network. The last DBL (Darknetconv2d_ Batch Normalization_Leaky) output of the second scale output layer is fused with the fourth DBL output of the network. And a 52$\times$ 52 scale output is added on the basis of the original network. Experiments were carried out on our dataset. The experimental results show that compared with YOLOv3-tiny, the accuracy of the improved network structure is increased by 6.3%, and the detection speed is 31.8fps, ensuring the requirements of real-time detection.

关键词： Feature extraction Object detection Real-time systems Training Convolution Microprocessors Computer architecture

来源：评论

学校读者我要写书评

暂无评论

A Multitask Cascaded Convolutional Neural Network based on Full Frame Histogram Equalization for Vehicle Detection

A Multitask Cascaded Convolutional Neural Network based on F...

引用

Chinese Automation Congress

作者： Hua Gong Yong Zhang Ke Xu Fang Liu College of Science Shengyang Ligong University Shenyang China Science and Technology on Electro-Optical Information Security Control Laboratory Tianjing China

In order to improve the accuracy of vehicle detection, a Multitask Cascaded Convolutional Neural Network (MC-CNN) based on Full Frame Histogram Equalization (FFHE) algorithm is proposed. FFHE is used to enhance the image, which can solve the problem of unclear image, low contrast, low overall gray value and uneven illumination when the imaging condition is not ideal. MC-CNN locates and classifies the object in the image. The algorithm first performs FFHE operation on the test datasets. Then the region image containing a single vehicle is obtained by the object location algorithm based on Faster R-CNN. Next, the six classification (bus, microbus, minivan, sedan, SUV, and truck) results of each vehicle are obtained through the object classification algorithm based on CNN. To verify the performance of the algorithm, MC-CNN was trained on BIT-Vehicle and tested on SYIT2018-Vehicle. Compared with Faster R-CNN, the average classification accuracy of our algorithm was improved by 11.5% on SYIT2018-Vehicle dataset.

关键词： Vehicle detection MC-CNN Full frame histogram equalization

来源：评论

学校读者我要写书评

暂无评论

optical Image Encryption Based on Mixed Chaotic Maps and Single-Shot Digital Holography

引用

Transactions of Tianjin University 2017年第2期23卷 184-191页

作者： Yonggang Su Chen Tang Xia Chen Biyuan Li Wenjun Xu Zhenkun Lei School of Electrical and Information Engineering Tianjin University Tianjin 300072 China Science and Technology on Electro-Optical Information Security Control Laboratory Tianjin 300308 China State Key Laboratory of Structural Analysis for Industrial Equipment Dalian University of Technology Dalian 116024China

Random phase masks play a key role in optical image encryption schemes based on double random phase technique. In this paper, a mixed chaotic method is proposed, which can efficiently solve some weaknesses that one-dimensional (1-D) single chaotic maps encounter to generate random phase masks. Based on the chaotic random phase masks, optical image encryption and decryption are realized with a single-shot digital holographic technique. In the proposed encryption scheme, the initial value and parameters of mixed chaotic maps serve as secret keys, which is convenient for the key management and transmission. Moreover, it also possesses high resistance against statistical attack, brute-force attack, noise attack and shear attack. Simulation results and security analysis verify the validity and security of the proposed encryption scheme. © 2017, Tianjin University and Springer-Verlag Berlin Heidelberg.

关键词： Chaotic systems Geometrical optics Holography Image processing Lyapunov methods optical data processing

来源：评论

学校读者我要写书评

暂无评论

Wide-range-tunable Dirac-cone band structure in a chiral-time-symmetric non-Hermitian system

引用

Physical Review A 2017年第5期96卷 052121-052121页

作者： S. Lin Z. Song School of Physics Nankai University Tianjin 300071 China Science and Technology on Electro-Optical Information Security Control Laboratory Tianjin 300308 China

We establish a connection between an arbitrary Hermitian tight-binding model with chiral (C) symmetry and its non-Hermitian counterpart with chiral-time (CT) symmetry. We show that such a non-Hermitian Hamiltonian is pseudo-Hermitian. The eigenvalues and eigenvectors of the non-Hermitian Hamiltonian can be easily obtained from those of its parent Hermitian Hamiltonian. It provides a way to generate a class of non-Hermitian models with a tunable full real band structure by means of additional imaginary potentials. We also present an illustrative example that could achieve a cone structure from the energy band of a two-layer Hermitian square lattice model.

关键词： Fermi gases Discrete symmetries Schroedinger equation

来源：评论

学校读者我要写书评

暂无评论

构建新型网络空间安全生态体系实现从网络大国走向网络强国

引用

Engineering 2018年第1期 105-116页

作者：杨小牛王巍许小丰庞国荣张春磊 Science and Technology on Communication Information Security Control Laboratory Science and Technology on Electro-Optical Information Security Control Laboratory

首先面向我国网络空间实际情况,提出网络空间"三分论",针对公共互联网(C空间)、党政军保密网(S空间)和关键基础设施网(K空间)三大网络空间的不同需求,给出了相应的安全应对策略和研究框架。然后重点针对公共互联网,探讨了基于... 详细信息

首先面向我国网络空间实际情况,提出网络空间"三分论",针对公共互联网(C空间)、党政军保密网(S空间)和关键基础设施网(K空间)三大网络空间的不同需求,给出了相应的安全应对策略和研究框架。然后重点针对公共互联网,探讨了基于SMCRC环(situation awareness,monitoring,cooperative defense,recovery,countermeasure)的网络空间安全生态体系的功能特点和内在信息交互方式,并分别对"SMCRC环"中态势感知、持续监控、协同防御、快速恢复、溯源反制等五大环节的核心关键技术进行了讨论,以期对国家网络空间安全研究起到抛砖引玉之作用。

关键词：网络空间安全生态体系 SMCRC环网络空间三分论

来源：评论

学校读者我要写书评

暂无评论

High visibility temporal ghost imaging with classical light

arXiv

引用

arXiv 2017年

作者： Liu, Jianbin Wang, Jingjing Chen, Hui Zheng, Huaibin Liu, Yanyan Zhou, Yu Li, Fu-Li Xu, Zhuo Electronic Materials Research Laboratory Key Laboratory of the Ministry of Education International Center for Dielectric Research Xi'an Jiaotong University Xi'an710049 China Science and Technology on Electro-Optical Information Security Control Laboratory MOE Key Laboratory for Nonequilibrium Synthesis and Modulation of Condensed Matter Department of Applied Physics Xi'an Jiaotong University Xi'an710049 China

High visibility temporal ghost imaging with classical light is possible when superbunching pseudothermal light is employed. In the numerical simulation, the visibility of temporal ghost imaging with pseudothermal light equaling (4.7 ± 0.2)% can be increased to (75 ± 8)% in the same scheme with superbunching pseudothermal light. The reasons for the difference in visibility and quality of the retrieved images in different situations are discussed in detail. It is concluded that high visibility and high quality temporal ghost image can be obtained by collecting large enough number of data points. The results are helpful to understand the difference between ghost imaging with classical light and entangled photon pairs. The superbunching pseudothermal light can be employed to improve the image quality in ghost imaging applications. Copyright © 2017, The Authors. All rights reserved.

关键词： Visibility

来源：评论

学校读者我要写书评

暂无评论

Agile laser beam deflection with high steering precision and angular resolution using liquid crystal optical phased array

Agile laser beam deflection with high steering precision and...

引用

IEEE International Nanoelectronics Conference (INEC)

作者： Xiangru Wang Shuanghong Wu Man Li School of Optoelectronic Information University of Electronic Science and Tech. of China Chengdu China Science and Technology on Electro-Optical Information Security Control Laboratory Sanhe China

To improve the steering precision and angular resolution of liquid crystal optical phased array, in this paper, we proposed a modified periodic phase controlled method to realize a continuous scanning with a constant angular resolution of less than 20 μrad, meanwhile, its precision is less than 5 μrad (RMS). Also, another method we proposed is called sub-aperture coherence can realize even better angular resolution less than 5 μrad which is the limitation of our measurement instrument.

关键词： Measurement by laser beam Laser beams Phased arrays optical arrays Coherence Laser theory Laser radar

来源：评论

学校读者我要写书评

暂无评论

GB-RC4: Effective brute force attacks on RC4 algorithm using GPU 7

GB-RC4: Effective brute force attacks on RC4 algorithm using...

引用

7th International Green and Sustainable Computing Conference, IGSC 2016

作者： Xue, Pei Li, Tao Dong, Han Liu, Chunbo Ma, Wenjing Pei, Songwen College of Computer and Control Engineering Nankai University Tianjin300071 China Information Security Evaluation Center Civil Aviation University of China Tianjin300300 China Laboratory of Parallel Software and Computational Science State Key Laboratory of Computing Science Institute of Software CAS Beijing China Shanghai Key Lab of Modern Optical Systems Department of Computer Science and Engineering University of Shanghai for Science and Technology Shanghai200093 China

ISBN: (纸本)9781509051175

Encryption algorithms are applied to a variety of fields and the security of encryption algorithms depends heavily on the computational infeasibility of exhaustive key-space search. RC4 algorithm has an extensive application for stream encryption, however, the disadvantages of traditional RC4 serial algorithm are large computational quantity and slow computation speed, which means a great challenge for throughput of computing system. Graphics processors recently have been widely used as accelerators for high throughput computing. In this paper, GPU is used to improve the efficiency and performance of attacking RC4 algorithm. First, we revised RC4 algorithm to make it easier to parallelize on a GPU. Then we take advantage of constant memory and shared memory in GPU to search the key space of RC4 algorithm efficiently. Besides, we solved the bank conflict problem in shared memory to further improve the performance. Our approach was evaluated in the whole key space of 40-bit, and the results show it can complete searching of the whole key space within 12.8 hours. © 2016 IEEE.

关键词： Graphics processing unit

来源：评论

学校读者我要写书评

暂无评论

Mid-infrared ZnGeP2 optical parametric oscillator pumped by Tm:YAlO3 laser

Mid-infrared ZnGeP2 optical parametric oscillator pumped by ...

引用

作者： Wang, Lixin Zhang, Xiaojuan Xu, Hong School of Electro-Optical Beijing Institute of Technology Beijing100081 China Science and Technology on Electro-Optical Information Security Control Laboratory Sanhe065201 China

We report experimental results on a mid-infrared (mid-IR) ZnGeP2 optical parametric oscillator (OPO) pumped by a thulium laser. A high efficient Q-switched a-cut Tm:YAlO3 (Tm:YAP) laser was used to directly drive a ZnGeP2 OPO to generate 3-5 μm wavelength range. The diode pumped Tm:YAP laser operated on 1.99 μm at 10 kHz repetition rate, and its optical-tooptical conversion efficiency was 22.2%. The laser was run with room temperature components and achieved nearly 2 W average output power at 3-5 μm region with an efficiency of 5.5% starting from the pump diode. The influences of curvature and reflectivity of OPO output mirror on laser output are analysed and validated by experiments, and the results have benefit for the optimisation of such lasers. Copyright © 2015 Inderscience Enterprises Ltd.

关键词： optical parametric oscillators

来源：评论

学校读者我要写书评

暂无评论

Fast multi-level connected component labeling for large-scale images

Fast multi-level connected component labeling for large-scal...

引用

International Conference on Optoelectronics and Microelectronics (ICOM)

作者： Yuhai Li Science and Technology on Electro-Optical Information Security Control Laboratory Sanhe Hebei P.R.China

Connected component (CC) labeling is time consuming during image segmentation and object identification, especially when input images are large-scale and binary converting uses multi-threshold. So in this paper, we present a fast multi-level CC labeling by combining online threshold segmentation and a one-pass labeling algorithm, also a detailed architecture for hardware implementation is proposed. During labeling, firstly a gray image is converted to several binary data flows with different thresholds. Then they are marked by run-length code respectively with 2× 2 scanning windows, in the meantime, the equal labels between two adjacent rows are obtained by these windows for CC merging. By executing this process until the end of input image, the CC labels are obtained after one-pass image scanning. Our method is tested on more than 1000 gray images which resolution are 1920×1080. Experimental results show that our algorithm can extract the positions and areas of CC in multi-threshold binary images at least 50 frames per second (fps) on a Stratix IV FPGA platform running at 109.7 MHz.

关键词： Labeling Memory management Random access memory Signal processing algorithms Algorithm design and analysis Hardware Image processing

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：