检索结果-内蒙古大学图书馆

5th IEEE Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2021

作者： Zhao, Luan Zhang, Xiaofeng Nanchang Hangkong University Key Laboratory of Jiangxi Province for Image Processing and Pattern Recognition School of Information Engineering Nanchang China

Constructing the pyramidal architecture for the feature is currently a very effective way to obtain feature information of objects at different scales. Although the feature pyramid can realize the recognition and detection of multi-scale objects in the object detection task well, it still has some limitations. Since the feature information of different levels is often not from the same layer of the network, it is difficult to obtain the feature of different objects information at a certain scale from a certain level feature map of the pyramid network. To solve this problem, we present a novel object detection architecture, named Enhanced Multi-scale Feature Fusion Pyramid Network (EMFFPNet). Our network consists of Enhanced Multi-scale Feature Fusion Module (EMFFM) and Predictor Optimization Module (POM). In EMFFM, Features at different levels can be fused into the Enhanced features as outputs, which are more representative and deterministic. In order to enable the enhanced features to play their respective roles in the pyramid network, we assign different weights to fusion features of different levels in POM. We perform the experiments on the COCO detection benchmark. The experimental results indicate that the performance of our model is much better than the state-of-the-art model. © 2021 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Dense Connected Residual Generative Adversarial Network for Single Image Deblurring 5

Dense Connected Residual Generative Adversarial Network for ...

引用

5th IEEE Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2021

作者： Qian, Pan Wu, Yeyun Zhang, Xiaofeng Nanchang Hangkong University School of Information Engineering Key Laboratory of Jiangxi Province for Image Processing and Pattern Recognition Nanchang China

Recently, deep convolutional learning has been applied to image deblurring, which greatly improves the performance of single-image blind deblurring algorithms. However, most deep image deblurring models based on convolutional neural networks do not make full use of the information of low-level spatial convolutions, so they achieve relatively poor performance. In this paper, we improved the generator of the Generative Adversarial Network (GAN) and proposed dense networks and residual networks to solve this problem in image deblurring. The network makes full use of the hierarchical features of all convolutional layers. Specifically, we extract the local features of the convolutional layer through the proposed densely connected convolutional layer. Moreover, the dense connections of the residual dense network can connect all layers from the previous state to the current state. Then using local jump connection and feature fusion, it can adaptively learn more effective features from the previous and current local features, making the trained network more extensive. After fully obtaining the dense local features, the global residual connection is used to learn the global feature information in an overall manner. Experimental comparison and verification have proved the effectiveness of our method and achieved excellent results both quantitatively and qualitatively. © 2021 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Object classification via PCANet and color constancy model 4

Object classification via PCANet and color constancy model

引用

4th International Conference on Advanced Design and Manufacturing Engineering, ADME 2014

作者： Hu, De Kun Zhang, L. Zhao, Wei Dong Yan, Tao Key Laboratory of Pattern Recognition and Intelligent Information Processing Institutions of Higher Education of Sichuan Province Chengdu University Chengdu China

ISBN: (纸本)9783038352570

In order to classify the objects in nature images, a model with color constancy and principle component analysis network (PCANet) is proposed. The new color constancy model imitates the functional properties of the HVS from the retina to the double-opponent cells in V1. PCANet can be designed and learned extremely, which comprises only the very basic data processing components: cascaded principal component analysis (PCA), binary hashing, and block-wise histograms. At last, a SVM is trained to classify the object in the image. The results of experiments demonstrate the potential of the model for object classification in wild color images. © (2014) Trans Tech Publications, Switzerland.

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

recognition of facial expression via kernel PCA network 3

Recognition of facial expression via kernel PCA network

引用

3rd International Conference on Information Technology and Management Innovation, ICITMI 2014

作者： Hu, De Kun Ye, An Sheng Li, Li Zhang, Li Key Laboratory of Pattern Recognition and Intelligent Information Processing Institutions of Higher Education of Sichuan Province Chengdu University Chengdu China

ISBN: (纸本)9783038352396

In this work, a kernel principle component analysis network (KPCANet) is proposed for classification of the facial expression in unconstrained images, which comprises only the very basic data processing components: cascaded kernel principal component analysis (KPCA), binary hashing, and block-wise histograms. In the proposed model, KPCA is employed to learn multistage filter banks. It is followed by simple binary hashing and block histograms for indexing and pooling. For comparison and better understanding, We have tested these basic networks extensively on many benchmark visual datasets(such as the JAFFE [13] database, the CMU AMP face expression database, a part of the Extended Cohn-Kanade (CK+) database), The results demonstrate the potential of the KPCANet serving as a simple but highly competitive baseline for facial expression recognition. © 2014 Trans Tech Publications, Switzerland.

关键词： Database systems

来源：评论

学校读者我要写书评

暂无评论

An efficient multiparty quantum secret sharing scheme using a single qudit

引用

Chinese Physics B 2023年第8期32卷 161-170页

作者：胡文文熊邦书周日贵 School of Information Engineering Nanchang Hangkong UniversityNanchang 330063China Key Laboratory of Image Processing and Pattern Recognition of Jiangxi Province Nanchang Hangkong UniversityNanchang 330063China College of Information Engineering Shanghai Maritime UniversityShanghai 201306China

The aim of quantum secret sharing,as one of most promising components of quantum cryptograph,is one-tomultiparty secret communication based on the principles of quantum *** this paper,an efficient multiparty quantum secret sharing protocol in a high-dimensional quantum system using a single qudit is *** participant's shadow is encoded on a single qudit via a measuring basis encryption method,which avoids the waste of qudits caused by basis *** analysis indicates that the proposed protocol is immune to general attacks,such as the measure-resend attack,entangle-and-measure attack and Trojan horse *** to former protocols,the proposed protocol only needs to perform the single-qudit measurement operation,and can share the predetermined dits instead of random bits or dits.

关键词： quantum secret sharing high-dimensional measurement basis encrypted security analysis

来源：评论

学校读者我要写书评

暂无评论

A Robust CPD Approach Based on Shape Context 33

A Robust CPD Approach Based on Shape Context

引用

第三十三届中国控制会议

作者： YANG Yawen ZHANG Peng Peng QIAO Yu YANG Jie WANG Sheng Zheng The Key Laboratory of Ministry of Education for System Control and Information Processing Institute of Image Processing and Pattern RecognitionShanghai Jiao Tong University Merchant Marine College Shanghai Maritime University

ISBN: (纸本)9781479947249

Point matching is an important component of image *** years,Coherent Point Drift(CPD) method becomes a very popular point matching *** treats point matching as a probability estimation problem and speeds up the process of matching a *** this method,one set of points are thought to be sampled from a Gaussian Mixture Model(GMM),which is centered by the other set of ***,CPD is sensitive to outliers and noises,especially when the noise ratio increased or the number of outliers gets much *** deal with this problem,we introduce shape context into the step of searching for matching points and then improve the form of prior probabilities of GMM in this *** main idea of our method is that if the most points in a data set are likely to be matched to a particular centroid,this Gaussian component should be have a more influence to ***,we set prior probability of GMM with the similarity between GMM components and the data *** the computation of similarity is based on shape *** experiments on 2D and 3D images show that when noise ratio is low,our method performs as well as CPD does,but as the ratio increased,our method is more robust and satisfactory than CPD.

关键词： CPD,Shape context,Image registration,Point matching

来源：评论

学校读者我要写书评

暂无评论

3D multipath planning for UAV based on network graph

引用

Journal of Systems Engineering and Electronics 2011年第4期22卷 640-646页

作者： Xin Liu Chengping Zhou Mingyue Ding Institute for Pattern Recognition and Artificial Intelligence Key Laboratory of National Defense Science and Technology forMulti-spectral Information Processing Technologies Huazhong University of Science and Technology Wuhan 430074 R R. China School of Life Science and Technology Image Processing and Intelligence Control Key Laboratory of Education Ministry of China Huazhong University of Science and Technology Wuhan 430072 P. R. China

According to the characteristic and the requirement of multipath planning, a new multipath planning method is proposed based on network. This method includes two steps： the construction of network and multipath searching. The construction of network proceeds in three phases： the skeleton extraction of the configuration space, the judgment of the cross points in the skeleton and how to link the cross points to form a network. Multipath searching makes use of the network and iterative penalty method （IPM） to plan multi-paths, and adjusts the planar paths to satisfy the requirement of maneuverability of unmanned aerial vehicle （UAV）. In addition, a new height planning method is proposed to deal with the height planning of 3D route. The proposed algorithm can find multiple paths automatically according to distribution of terrain and threat areas with high efficiency. The height planning can make 3D route following the terrain. The simulation experiment illustrates the feasibility of the proposed method.

关键词： path planning skeleton graph iterative penaltymethod （IPM） network graph.

来源：评论

学校读者我要写书评

暂无评论

Generalized quantum genetic algorithm of multidimensional space

Generalized quantum genetic algorithm of multidimensional sp...

引用

2014 International Conference on Logistics Engineering, Management and Computer Science, LEMCS 2014

作者： Zhang, Xiaofeng Miao, Wei Sui, Guifang Yang, Guowei Key Laboratory of JiangXi Procince for Image Processing and Pattern Recognition Information Engineering of NanChang HangKong University NanChang JiangXi330063 China

ISBN: (纸本)9789462520103

Quantum coding plays an important role in quantum genetic algorithm and affects the optimizing efficiency of algorithm, However, there are some defects in existing quantum genetic algorithm: the quantum coding scheme reduces the stability of the algorithm, making the algorithm easily occurs premature convergence and falls into local extremums. Therefore, coding scheme of multi gene chain is extended in a four dimensional space and multidimensional space based on an improved algorithm. Through simulation experiments of function extremum, combine double and triplestranded coding scheme, in accordance with the characteristics of qubits and the normalization condition. Through the experiment, study the effect of multiple genes encoding scheme chain on the performance of the algorithm. The results show that the optimal results of the algorithm can be better and the precision of solution can be higher by increasing the genetic chain. It is an effective strategy to improve the performance of algorithm. © 2014. The authors - Published by Atlantis Press.

关键词： Chains

来源：评论

学校读者我要写书评

暂无评论

Fast KNN search for big data with set compression tree and best bin first 2

Fast KNN search for big data with set compression tree and b...

引用

2nd International Conference on Cloud Computing and Internet of Things, CCIOT 2016

作者： Chen, Zhenjie Yan, Jingqi Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Key Laboratory of Ministry of Education for System Control and Information Processing China Shanghai China

ISBN: (纸本)9781467398213

This paper proposes k nearest neighbors (kNN) search based on set compression tree (SCT) and best bin first (BBF) to deal with the problem for big data. The large compression rate by set compression tree is achieved by compressing the set of descriptors jointly instead of compressing on a per-descript or basis. So set compression tree has a good performance in kNN search at a low bit rate. At the same time, the best bin first (BBF) is a very efficient algorithm to find the approximately kNN from a large number of high dimensional feature descriptors. SCT-BBF is a novel exploration and it improves search performance in three aspects: First, SCT-BBF requires less memory footprint, which is important in big data age. Second, it increases accuracy compared traditional method like KD-Tree and original SCT. SCT-BBF can be used with other data processing methods like PCA and SIFT to perform better. Third, this paper adopts finer search to increase accuracy at a slight loss of speed. And it can extend to big data easily. © 2016 IEEE.

关键词： Big data

来源：评论

学校读者我要写书评

暂无评论

A new circle targets extraction method from high resolution remote sensing imagery

A new circle targets extraction method from high resolution ...

引用

4th International Workshop on Advanced Computational Intelligence, IWACI 2011

作者： Chen, Z. Liu, J.G. Wang, G.Y. Institute for Pattern Recognition and Artificial Intelligence Huazhong University of Science and Technology Multi-Spectral Information Processing State Key Laboratory China

ISBN: (纸本)9781612843735

According to the drawback of the traditional circle target extraction algorithm from high resolution remote sensing imagery used by Hough Transform, such as computation complexity, low efficiency and etc, a new circle target extraction method is proposed in this paper which can extract multiple circle targets with different radius at one time. First, the Average Absolute Difference is implemented to enhance the edge of the circle targets and suppress the noise of the background. Secondly, the locally self-adaptive segmentation algorithm is implemented to obtain the binary image. Thirdly, the thinning algorithm based on model computation is implanted to obtain the single pixel edge of the circle targets and in order to reduce the computation times in the following process. Furthermore, a pruning algorithm is necessary;finally, a modified Hough transform algorithm is proposed to obtain the circle targets. The experimental results demonstrate that the new circle targets algorithm can extract the multiple circle targets quickly and accurately, which has three advantages: low time consuming, high detection rate, robust to noise and fragmentary boundaries. © 2011 IEEE.

关键词： Binary images

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：