检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Li, Chenglong Zhu, Qiwen Liu, Tubiao Tang, Jin Su, Yu Information Materials and Intelligent Sensing Laboratory of Anhui Province Anhui Provincial Key Laboratory of Multimodal Cognitive Computation School of Artificial Intelligence Anhui University Hefei230601 China Information Materials and Intelligent Sensing Laboratory of Anhui Province Key Lab of Intelligent Computing and Signal Processing of Ministry of Education Anhui Provincial Key Laboratory of Multimodal Cognitive Computation School of Computer Science and Technology Anhui University Hefei230601 China Hefei Normal University Institute of Artificial Intelligence Hefei Comprehensive National Science Center Hefei230061 China

Hand hygiene is a standard six-step hand-washing action proposed by the World Health Organization (WHO). However, there is no good way to supervise medical staff to do hand hygiene, which brings the potential risk of disease spread. Existing action assessment works usually make an overall quality prediction on an entire video. However, the internal structures of hand hygiene action are important in hand hygiene assessment. Therefore, we propose a novel fine-grained learning framework to perform step segmentation and key action scorer in a joint manner for accurate hand hygiene assessment. Existing temporal segmentation methods usually employ multi-stage convolutional network to improve the segmentation robustness, but easily lead to over-segmentation due to the lack of the long-range dependence. To address this issue, we design a multi-stage convolution-transformer network for step segmentation. Based on the observation that each hand-washing step involves several key actions which determine the hand-washing quality, we design a set of key action scorers to evaluate the quality of key actions in each step. In addition, there lacks a unified dataset in hand hygiene assessment. Therefore, under the supervision of medical staff, we contribute a video dataset that contains 300 video sequences with fine-grained annotations. Extensive experiments on the dataset suggest that our method well assesses hand hygiene videos and achieves outstanding performance. Copyright © 2022, The Authors. All rights reserved.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Live cell single molecule-guided Bayesian localization super resolution microscopy

引用

细胞研究（英文版） 2017年第5期27卷 713-716页

作者： Mingshu Zhang Wenting He Zhiyong Liu Fa Zhang Jennifer Lippincott-Schwartz Key Laboratory of RNA Biology Institute of Biophysics Chinese Academy of Sciences Beijing 100101 China Beijing Key Laboratory of Noncoding RNA Institute of Biophysics Chinese Academy of Sciences Beijing 100101 China Key Lab of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing 100190 China Janelia Research Campus Howard Hughes Medical Institute Ashburn Virginia 20147 USA

Dear Editor,Many current super resolution (SR) microscopic techniques [1-6] have been successfully applied to image cellular dynamics in living cells,but their applications have remained technically *** cell stimulated emission depletion (STED)/reversible saturable optical linear fluorescence transitions (RESOLFT) microscopy and structured illumination microscopy (SIM)/nonlinear SIM require sophisticated expensive optical setups and specialized expertise for accurate optical *** cell photo-activated localization microscopy (PALM)/stochastic optical reconstruction microscopy (STORM)use less complicated setup;however,a scientific complementary metal-oxide-semiconductor (sCMOS) camera,whose pixel-dependent noise must be characterized and calibrated before use [7],is required for extremely high acquisition speed over tens of thousands of ***,wide field-based SR microscopies have been developed to improve temporal resolution using much fewer time-lapse images (hundreds to thousands) than PALM/STORM [8,9].One of them,Bayesian analysis of the blinking and bleaching (3B) [8],offers enormous potential to resolve ultrastructure and fast cellular dynamics in living cells beyond the diffraction *** its potential,3B analysis is impractical when imaging the nanoscale dynamics in large fields of view over long time periods,as the calculation is extremely time-consuming [8] and/or the analysis consumes large amounts of web resources [10].Another major problem of 3B imaging is the artificial thinning and thickening of structures both in simulated image data and in experimental data [8,10].

关键词： @@

来源：评论

学校读者我要写书评

暂无评论

Research on Data Augmentation for Garbage Classification with Target Detection Algorithm

Research on Data Augmentation for Garbage Classification wit...

引用

International Applied Computational Electromagnetics Society Symposium (ACES)

作者： Jiansong Chen Shikai Zuo Yun Lin Kai Tang Jianhang Zou Naixing Feng Yijun Cai Smart Sensing Integrated Circuit Engineering Research Center of Universities in Fujian Province Xiamen University of Technology Xiamen China Xiamen Wish Information Technology Co. Ltd Xiamen China Key Laboratory of Intelligent Computing & Signal Processing Ministry of Education Anhui University Hefei China

Garbage classification is an essential work in daily life. With the development of artificial intelligence (AI), we have begun to use object detection to achieve garbage classification. However, the stacking and occlusion of garbage often make the detection performance unsatisfactory. In this paper, YOLOv5 algorithm is used for garbage detection and classification. YOLOv5 has three detection heads, which can detect objects at multiple scales. At the same time, Mosaic and Coarse Dropout image augmentation techniques are used to process our private data set of garbage, which improves the diversity of images. In this way, the target classification accuracy of the detection model can reach 75.7%, which is significantly improved compared with 69.9% without data augmentation. This paper provides a new technical support for the scene classification of stacked garbage based on deep learning.

关键词： Training Deep learning Head Stacking Object detection Computational electromagnetics Data models

来源：评论

学校读者我要写书评

暂无评论

Highly isolated two-elements ultra-wideband mimo fractal antenna with multi band-notched characteristics

引用

Progress In Electromagnetics Research C 2021年 116卷 13-24页

作者： Cai, Yong Cheng, Guangshang Ren, Xingang Wu, Jie Ren, Hao Song, Kaihong Huang, Zhixiang Wu, Xianliang Information Materials and Intelligent Sensing Laboratory of Anhui Province Institute of Physical Science and Information Technology Anhui University Hefei230039 China Anhui Province Key Laboratory of Target Recognition and Feature Extraction Luan237000 China Key Laboratory of Intelligent Computing & Signal Processing Ministry of Education Institute of Physical Science and Information Technology Anhui University Hefei230039 China Key Laboratory of Electromagnetic Environmental Sensing Department of Education of Anhui Province Anhui University Hefei230039 China

This work presents a high isolation UWB-MIMO antenna with a bandwidth up to 8.6 GHz based on a Minkowski fractal structure. The proposed antenna is fed by microstrip and comprises two orthogonal monopole antennas, which delivers a decent isolation effect. Moreover, the ground is designed as two separated blocks with an I-shaped branch for improving the isolation degree between the units. The resultant isolation degree of this antenna is greater than 25 dB. Besides, the electromagnetic interference in the partial frequency band (such as Wi-Max band (3.45–4.45 GHz), WLAN band (5.1– 5.8 GHz), and X-band (7.25–7.75 GHz)) is further prevented through etching a split-ring resonator (SRR) and C-slot on the unit. The antenna reflection coefficient of the UWB-MIMO antenna at the notch is 3.5 dB, which indicates that the antenna has a conspicuousness anti-interference effect. Through the above judicious design, the proposed UWB-MIMO antenna possesses a relative bandwidth of 113% (up to 8.6 GHz). The envelope correlation coefficient between antenna units is less than 0.005, and the antenna radiation efficiency is up to 80%. The results indicate that the proposed MIMO antenna meets UWB applications. © 2021, Electromagnetics Academy. All rights reserved.

关键词： Monopole antennas

来源：评论

学校读者我要写书评

暂无评论

AUCSeg: AUC-oriented Pixel-level Long-tail Semantic Segmentation

arXiv

引用

arXiv 2024年

作者： Han, Boyu Xu, Qianqian Yang, Zhiyong Bao, Shilong Wen, Peisong Jiang, Yangbangyan Huang, Qingming Key Lab. of Intelligent Information Processing Institute of Computing Technology CAS China School of Computer Science and Tech. University of Chinese Academy of Sciences China Peng Cheng Laboratory China Key Laboratory of Big Data Mining and Knowledge Management CAS China

The Area Under the ROC Curve (AUC) is a well-known metric for evaluating instance-level long-tail learning problems. In the past two decades, many AUC optimization methods have been proposed to improve model performance under long-tail distributions. In this paper, we explore AUC optimization methods in the context of pixel-level long-tail semantic segmentation, a much more complicated scenario. This task introduces two major challenges for AUC optimization techniques. On one hand, AUC optimization in a pixel-level task involves complex coupling across loss terms, with structured inner-image and pairwise inter-image dependencies, complicating theoretical analysis. On the other hand, we find that mini-batch estimation of AUC loss in this case requires a larger batch size, resulting in an unaffordable space complexity. To address these issues, we develop a pixel-level AUC loss function and conduct a dependency-graph-based theoretical analysis of the algorithm’s generalization ability. Additionally, we design a Tail-Classes Memory Bank (T-Memory Bank) to manage the significant memory demand. Finally, comprehensive experiments across various benchmarks confirm the effectiveness of our proposed AUCSeg method. The code is available at https://***/boyuh/AUCSeg. © 2024, CC BY.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

An image compression method based on block truncation coding and linear regression

引用

Journal of Information Hiding and Multimedia signal processing 2016年第1期7卷 198-215页

作者： Lyu, Wan-Li Chang, Chin-Chen Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education School of Computer Science and Technology Anhui University Hefei230039 China Department of Information Engineering and Computer Science Feng Chia University 100 Wenhwa Rd Seatwen Taichung40724 Taiwan Department of Computer Science and Information Engineering Asia University Taichung41354 Taiwan

This paper describes an image compression method based on block truncation coding (BTC) and linear regression coding (LRC) hybrid strategy. BTC is simple and easy to use, but only two representative values can not adequately represent the visual feature of image blocks. However, LRC can obtain different values of points with the line function by recording the coefficient of the regression line while the data distribution shows a linear relationship. The proposed method designs a hybrid strategy to combine the advantages of the two methods to improve image quality. In the paper, first, models are designed to rearrange the data of image blocks to satisfy a linear distribution, and then these models are used to code the image blocks. The experimental results indicated that the proposed method is superior to the BTC scheme in terms of image quality, and it can be used for image compression applications. © 2016.

关键词： Linear regression

来源：评论

学校读者我要写书评

暂无评论

Hybrid color image steganography method used for copyright protection and content authentication

引用

Journal of Information Hiding and Multimedia signal processing 2015年第4期6卷 686-696页

作者： Lyu, Wan-Li Chang, Chin-Chen Chou, Yeh-Chieh Lin, Chia-Chen Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education School of Computer Science and Technology Anhui University Hefei230039 China Department of Information Engineering and Computer Science Feng Chia University 100 Wenhwa Rd Seatwen Taichung40724 Taiwan Department of Computer Science and Information Management Providence University Taichung43301 Taiwan

In this paper, we proposed a hybrid image data hiding method used for watermarking applications in order to make sure the hidden dual watermarks survive after the image manipulation and enhancement processes. To achieve this goal, the proposed method is divided into two phases. In the first phase, the YCbCr color space is exploited, and, then, LSB techniques are used to embed the first level watermark bits in the YCbCr channel. In the second phase, the modification direction method is used to embed the second level watermark bits in the RGB channel. The experimental results confirmed that the proposed method resists most geometric attacks more effectively than Lusson et al.'s method. Therefore, the method is suitable for protecting the ownership of images and integrity of the potential tampered images. © 2015 ISSN.

关键词： Steganography

来源：评论

学校读者我要写书评

暂无评论

Visual Alignment Constraint for Continuous Sign Language Recognition

Visual Alignment Constraint for Continuous Sign Language Rec...

引用

International Conference on Computer Vision (ICCV)

作者： Yuecong Min Aiming Hao Xiujuan Chai Xilin Chen Key Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS) Institute of Computing Technology CAS Beijing China University of Chinese Academy of Sciences Beijing China Agricultural Information Institute Chinese Academy of Agricultural Sciences Beijing China

ISBN: (纸本)9781665428132

Vision-based Continuous Sign Language Recognition (CSLR) aims to recognize unsegmented signs from image streams. Overfitting is one of the most critical problems in CSLR training, and previous works show that the iterative training scheme can partially solve this problem while also costing more training time. In this study, we revisit the iterative training scheme in recent CSLR works and realize that sufficient training of the feature extractor is critical to solving the overfitting problem. Therefore, we propose a Visual Alignment Constraint (VAC) to enhance the feature extractor with alignment supervision. Specifically, the proposed VAC comprises two auxiliary losses: one focuses on visual features only, and the other enforces prediction alignment between the feature extractor and the alignment module. Moreover, we propose two metrics to reflect overfitting by measuring the prediction inconsistency between the feature extractor and the alignment module. Experimental results on two challenging CSLR datasets show that the proposed VAC makes CSLR networks end-to-end trainable and achieves competitive performance.

关键词： Training Measurement Visualization Image recognition Costing Gesture recognition Assistive technologies

来源：评论

学校读者我要写书评

暂无评论

Linear projective non-negative matrix factorization

引用

ICIC Express Letters 2013年第8期7卷 2285-2291页

作者： Hu, Lirui Wu, Jianguo Wang, Lei Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education Anhui University No. 3 Feixi Road Hefei 230039 China School of Computer Science and Technology Nantong University No. 9 Seyuan Road Nantong 226019 China School of Computer Science and Technology Anhui University No. 3 Feixi Road Hefei 230039 China

In order to solve the problem that the iterative method for Linear Projection- Based Non-negative Matrix Factorization (LPBNMF) is complex, a method, called Linear Projective Non-negative Matrix Factorization (LP-NMF), is proposed. In LP-NMF, from projection and linear transformation angle, an objective function of Frobenius norm is defined. The Taylor series expansion is used. An iterative algorithm for basis matrix and linear transformation matrix is derived, and a proof of algorithm convergence is provided. Experimental results show that the algorithm is convergent;relative to Nonnegative Matrix Factorization (NMF), the orthogonality and the sparseness of the basis matrix are better;in face recognition, there is higher recognition accuracy. The method for LP-NMF is effective. © 2013 ICIC International.

关键词： Linear transformations

来源：评论

学校读者我要写书评

暂无评论

Efficient Robust Principal Component Analysis via Block Krylov Iteration and CUR Decomposition

Efficient Robust Principal Component Analysis via Block Kryl...

引用

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Shun Fang Zhengqin Xu Shiqian Wu Shoulie Xie School of Information Science and Engineering Wuhan University of Science and Technology China Institute of Robotics and Intelligent Systems Wuhan University of Science and Technology China MoE Key Lab of Artificial Intelligence AI Institute Shanghai Jiao Tong University China Hubei Province Key Laboratory of Intelligent Information Processing and Real-Time Industrial Systems Signal Processing RF & Optical Dept. Institute for Infocomm Research A*STAR Singapore

Robust principal component analysis (RPCA) is widely studied in computer vision. Recently an adaptive rank estimate based RPCA has achieved top performance in low-level vision tasks without the prior rank, but both the rank estimate and RPCA optimization algorithm involve singular value decomposition, which requires extremely huge computational resource for large-scale matrices. To address these issues, an efficient RPCA (eRPCA) algorithm is proposed based on block Krylov iteration and CUR decomposition in this paper. Specifically, the Krylov iteration method is employed to approximate the eigenvalue decomposition in the rank estimation, which requires $O(ndrq+n(rq)^{2})$ for an $(n\times d)$ input matrix, in which $q$ is a parameter with a small value, $r$ is the target rank. Based on the estimated rank, CUR decomposition is adopted to replace SVD in updating low-rank matrix component, whose complexity reduces from $O(rnd)$ to $O(r^{2}n)$ per iteration. Experimental results verify the efficiency and effectiveness of the proposed eRPCA over the state-of-the-art methods in various low-level vision applications.

关键词：

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：