检索结果-南通市图书馆

Higher-order Sampled-data iterative learning control for nonlinear systems

在线全文

学校读者我要写书评

暂无评论

Higher-order Sampled-data iterative learning control for non...

Fusion network for small target detection based on YOLO and attention mechanism

2010 International Colloquium on Computing,Communication, Control, and Management (CCCM2010)

作者： xu hongwei college of Mechanical engineering China Jiliang university hangzhou,China Huang yanyan,Gao yanfeng college of Mechanical engineering China Jiliang university hangzhou,China

In this paper,a higher-order sampled-data iterative learning control method is presented for nonlinear system with relative degree and output time *** higher-order learning algorithm does not include numerical differentiations of any error order.A sufficient condition is proposed that the iterative learning control converges to the desired trajectory at each sampling instant if some Lipschiz conditions are satisfied and the sampling period is small enough.A simulation example shows that higher-order algorithm has larger convergence rate.

关键词： Sampled-data iterative learning control,Relative degree,Higher-order learning algorithm,Bellman lemma,convergence rate

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Optoelectronics Letters 2024年第6期20卷 372-378页

作者： xu Caie DONG Zhe ZHONG Shengyun CHEN Yijiang PAN Sishun WU Mingyang School of information and Electronic engineering Zhejiang University of Science and TechnologyHangzhou 310012China Research Development Department Hangzhou Xinhe Data Technology Co.Ltd.Hangzhou 311202China college of Mechanical engineering Zhejiang UniversityHangzhou 310013China

Target detection is an important task in computer vision research, and such an anomaly detection and the topic of small target detection task is more concerned. However, there are still some problems in this kind of researches, such as small target detection in complex environments is susceptible to background interference and poor detection results. To solve these issues, this study proposes a method which introduces the attention mechanism into the you only look once(YOLO) network. In addition, the amateur-produced mask dataset was created and experiments were conducted. The results showed that the detection effect of the proposed mothed is much better.

关键词： Fusion network for small target detection based on YOLO and attention mechanism

RGB oralscan video-based orthodontic treatment monitoring

在线全文

学校读者我要写书评

暂无评论

science China(information sciences) 2024年第1期67卷 124-139页

作者： Yan TIAN Hanshi FU Hao WANG Yuqi LIU Zhaocheng xu Hong CHEN Jianyuan LI Ruili WANG School of computer science and Technology Zhejiang Gongshang University Shining3D Tech Co. Ltd. School of information and Technology Monash University School of Mathematical and Computational sciences Massey University Department of Stomatology Zhejiang Provincial People's HospitalHangzhou Medical College School of computer and Computing science Zhejiang University City College

Orthodontic treatment monitoring involves using current images and previous 3D models to estimate the relative position of individual teeth before and after orthodontic *** process differs from image-based object 6D pose estimation due to the gingiva deformation and varying pose offsets for each tooth during *** by the fact that the poses of molars remain relatively fixed in implicit orthodontics,we design an approach that employs multiview pose evaluation and bidirectional temporal propagation for jaw pose estimation and then employs an iteration-based method for tooth *** handle changes in tooth appearance or location with weak texture across frames,we also introduce an instance propagation module that leverages positional and semantic information to explore instance relations in the temporal *** evaluated the performance of our approach using both the Shining3D tooth pose dataset and the Aoralscan3 tooth registration *** experimental results demonstrate remarkable accuracy improvements compared with existing methods.

关键词： digital dentistry object 6D pose estimation deep learning computer vision

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

图像去运动物体模糊算法

计算机应用与软件 2024年第9期41卷 273-278,328页

作者：刘旭王维东浙江省信息处理与通信网络重点实验室浙江杭州310027 浙江大学信息与电子工程学院浙江杭州310027

该文提出一种去运动物体模糊算法,将模糊图像分成运动模糊层和非运动模糊层,降低非运动物体区域对去模糊的影响,对运动模糊层使用局部线性运动进行建模并估计各图像块的模糊核,最后利用非盲解卷积方法求解清晰图像。实验结果表明在合成... 详细信息

该文提出一种去运动物体模糊算法,将模糊图像分成运动模糊层和非运动模糊层,降低非运动物体区域对去模糊的影响,对运动模糊层使用局部线性运动进行建模并估计各图像块的模糊核,最后利用非盲解卷积方法求解清晰图像。实验结果表明在合成图像上相对于其他传统算法,平均峰值信噪比提高12.11 dB,在自然图像上同样获得更好的效果。

关键词：图像处理解卷积去模糊运动物体

Single Image Enhancement in Sandstorm Weather via Tensor Least Square

在线全文

学校读者我要写书评

暂无评论

IEEE/CAA Journal of Automatica Sinica 2020年第6期7卷 1649-1661页

作者： Guanlei xu Xiaotong Wang Xiaogang xu college of computer and information engineering Zhejiang Gongshang UniversityHangzhou 310018China

In this paper,we present a tensor least square based model for sand/sandstorm removal in *** main contributions of this paper are as ***,an important intrinsic natural feature of outdoor scenes free of sand/sandstorm is found that the outlines in RGB channels are somewise similar,which discloses the physical validation using the tensor instead of the ***,a tensor least square optimization model is presented for the decomposition of edge-preserving base layers and *** model not only decomposes the color image(taken as an inseparable indivisibility)in X,Y directions,but also in Z direction,which meets the statistical feature of natural scenes and can physically disclose the intrinsic color *** model’s advantages are twofold:one is the decomposition of edgepreserving base layers and details that can be employed for contrast enhancement without artificial halos,and the other one is the color driving ability that makes the enhanced images as close to natural images as possible via the inherent color ***,the tensor least square optimization model based image enhancement scheme is discussed for the sandstorm weather ***,the experiments and comparisons with the stateof-the-art methods on real degraded images under sandstorm weather are shown to verify our method’s efficiency.

关键词： Image enhancement least square sandstorm weather tensor

Generalized Uncertainty Inequalities on Fisher information Associated with LCT

在线全文

学校读者我要写书评

暂无评论

Journal of Beijing Institute of Technology 2021年第3期30卷 217-227页

作者： Guanlei xu Xiaogang xu Xiaotong Wang School of computer and information engineering Zhejiang Gongshang UniversityHangzhou 310018China

Uncertainty principle plays an important role in multiple fields such as physics,mathem-atics,signal processing,*** linear canonical transform(LCT)has been used widely in optics and information processing and so *** this paper,a few novel uncertainty inequalities on Fisher information associated with linear canonical transform are *** newly deduced uncer-tainty relations not only introduce new physical interpretation in signal processing,but also build the relations between the uncertainty lower bounds and the LCT transform parameters a,b,c and d for the first time,which give us the new ideas for the analysis and potential *** addi-tion,these new uncertainty inequalities have sharper and tighter bounds which are the generalized versions of the traditional ***,some numeric examples are given to demon-strate the efficiency of these newly deduced uncertainty inequalities.

关键词： linear canonical transform(LCT) Fisher information uncertainty principle

Design of adiabatic two's complement multiplier-accumulator based on CTGAL

在线全文

学校读者我要写书评

暂无评论

Journal of zhejiang university-science A(Applied Physics & engineering) 2009年第2期10卷 172-178页

作者： Peng-jun WANG Jian xu Shi-yan YING Institute of Circuits and Systems Ningbo University Ningbo 315211 China college of information engineering Zhej iang University of Technology Hangzhou 310014 China

We propose a new design scheme for a Booth encoder based on clocked transmission gate adiabatic logic(CTGAL). In the new design the structural complexity of the Booth encoder is reduced while the speed of the multiplier is improved. The adiabatic two's complement multiplier-accumulator(MAC) is furthermore a design based on the CTGAL. The computer simulation results indicate that the designed circuit has the correct logic function and remarkably less energy consumption compared to that of the MAC based on complementary metal oxide semiconductor(CMOS) logic.

关键词： CTGAL circuit Adiabatic circuit Booth arithmetic Multiplier Two's complement MAC

Temporally Consistent Depth Map Prediction Using Deep Convolutional Neural Network and Spatial-Temporal Conditional Random Field

在线全文

学校读者我要写书评

暂无评论

Journal of computer science & Technology 2017年第3期32卷 443-456页

作者： xu-Ran Zhao xun Wang Qi-Chao Chen School of computer and information engineering Zhejiang Gongshang University Hangzhou 310018 China

Deep convolutional neural networks （DCNNs） based methods recently keep setting new records on the tasks of predicting depth maps from monocular images. When dealing with video-based applications such as 2D （2-dimensional） to 3D （3-dimensional） video conversion, however, these approaches tend to produce temporally inconsistent depth maps, since their CNN models are optimized over single frames. In this paper, we address this problem by introducing a novel spatial-temporal conditional random fields （CRF） model into the DCNN architecture, which is able to enforce temporal consistency between depth map estimations over consecutive video frames. In our approach, temporally consistent superpixel （TSP） is first applied to an image sequence to establish the correspondence of targets in consecutive frames. A DCNN is then used to regress the depth value of each temporal superpixel, followed by a spatial-temporal CRF layer to model the relationship of the estimated depths in both spatial and temporal domains. The parameters in both DCNN and CRF models are jointly optimized with back propagation. Experimental results show that our approach not only is able to significantly enhance the temporal consistency of estimated depth maps over existing single-frame-based approaches, but also improves the depth estimation accuracy in terms of various evaluation metrics.

关键词： depth estimation temporal consistency convolutional neural network conditional random fields

Design System of the Two-step Gear Reducer on Case-based Reasoning

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Mechanical engineering 2009年第5期22卷 671-679页

作者： JI Aimin HUANG Quansheng xu Huanmin CHEN Zhengming college of Mechanical and Electrical engineering Hohai University Changzhou 213022 China State Key Lab ofCAD＆CG Zhejiang University Hangzhou 310058 China college of computer and information engineering Hohai University Changzhou 213022 China

The design of the two-step gear reducer is a tedious and time-consuming process. For the purpose of improving the efficiency and intelligence of design process, case-based reasoning（CBR） technology was applied to the design of the two-step gear reducer. Firstly, the current design method for the two-step gear reducer was analyzed and the principle of CBR was described. Secondly, according to the characteristics of the reducer, three key technologies of CBR were studied and the corresponding methods were provided, which are as follows：（a） an object-oriented knowledge representation method, （b） a retrieval method combining the nearest neighbor with the induction indexing, and （c） a case adaptation algorithm combining the revision based on rule with artificial revision. Also, for the purpose of improving the credibility of case retrieval, a new method for determining the weights of characteristics and a similarity formula were presented, which is a combinatorial weighting method with the analytic hierarchy process（AHP） and roughness set theory. Lastly, according to the above analytic results, a design system of the two-step gear reducer on CBR was developed by VC＋＋, UG and Access 2003. A new method for the design of the two-step gear reducer is provided in this study. If the foregoing developed system is applied to design the two-step gear reducer, design efficiency is improved, which enables the designer to release from the tedious design process of the gear reducer so as to put more efforts on innovative design. The study result fully reflects the feasibility and validity of CBR technology in the process of the design of the mechanical parts.

关键词： two-step gear reducer case-based reasoning(CBR) weights of characteristics similarity