http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
News Frame Classification based on Word Embedding Techniques
Xuebin Shi(스슈에빈) 한국정보기술학회 2022 Proceedings of KIIT Conference Vol.2022 No.6
전 세계에서 매 순간 수많은 사건들이 일어나고 있으며, 그 사건들은 서로 다른 언어로 기록되고 보고된다. 그러나 이러한 뉴스가 대부분은 책임 귀속, 갈등, 인간 이익, 경제적 결과 및 도덕의 다섯 가지 프레임을 따른다. 뉴스의 핵심적인 관점과 입장은 종종 위의 다섯 가지 프레임에서 제시된다. 본 논문에서는 단어 임베딩 기술을 적용하여 다른 언어의 인쇄를 벡터로 변환한 후 클러스터 분석을 수행하여 프레임을 자동으로 추출하고 분류한다. 이 접근 방식은 수동 분석보다 더 많은 시간과 비용을 절약할 것이다. There are countless events happening all over the world every moment, which are recorded and reported in different languages. However, most of these news follow five frames: attribution of responsibility, conflict, human interest, economic consequences, and morality. The core viewpoints and positions of a news are often presented in the above five frames. In this paper, we adopt the word embedding technology to convert different languages press into vectors and then perform cluster analysis to automatically extract and classify frames. This approach saves more time and money than manual analysis.
Synthesis and luminescence properties of LiMg1-xMnxPO4 solid solutions
Xuebin Qiao 한양대학교 세라믹연구소 2013 Journal of Ceramic Processing Research Vol.14 No.S1
Mn2+-doped LiMg1-xMnxPO4 (x = 0.0035-1) solid solutions were prepared by conventional solid state reaction. A systematic structural of the solid solution series was carried out by X-ray powder diffraction. The emission and excitation spectra were employed to characterize the synthesized phosphors. The XRD results reveal that the samples adopt an olivine (Mg2SiO4) type structure with the space group Pnma. With increasing Mn2+ concentration, the XRD patterns shift systemically to lower angle in indexes of (200), (131), and (211) with all other lines, confirming the formation of solid solutions. The great red-shift of Mn2+ emission with increasing Mn2+-concentration from 0.35 to 100 mol% in LiMg1-xMnxPO4 is observed. The CIE coordinates and the emission shift of Mn2+ ions in LiMg1-xMnxPO4 are discussed in relation to the structural properties of LiMg1-xMnxPO4.
A Tuberculosis Detection Method Using Attention and Sparse R-CNN
Xuebin Xu,Jiada Zhang,Xiaorui Cheng,Longbin Lu,Yuqing Zhao,Zongyu Xu,Zhuangzhuang Gu 한국인터넷정보학회 2022 KSII Transactions on Internet and Information Syst Vol.16 No.7
To achieve accurate detection of tuberculosis (TB) areas in chest radiographs, we design a chest X-ray TB area detection algorithm. The algorithm consists of two stages: the chest X-ray TB classification network (CXTCNet) and the chest X-ray TB area detection network (CXTDNet). CXTCNet is used to judge the presence or absence of TB areas in chest X-ray images, thereby excluding the influence of other lung diseases on the detection of TB areas. It can reduce false positives in the detection network and improve the accuracy of detection results. In CXTCNet, we propose a channel attention mechanism (CAM) module and combine it with DenseNet. This module enables the network to learn more spatial and channel features information about chest X-ray images, thereby improving network performance. CXTDNet is a design based on a sparse object detection algorithm (Sparse R-CNN). A group of fixed learnable proposal boxes and learnable proposal features are using for classification and location. The predictions of the algorithm are output directly without non-maximal suppression post-processing. Furthermore, we use CLAHE to reduce image noise and improve image quality for data preprocessing. Experiments on dataset TBX11K show that the accuracy of the proposed CXTCNet is up to 99.10%, which is better than most current TB classification algorithms. Finally, our proposed chest X-ray TB detection algorithm could achieve AP of 45.35% and AP50 of 74.20%. We also establish a chest X-ray TB dataset with 304 sheets. And experiments on this dataset showed that the accuracy of the diagnosis was comparable to that of radiologists. We hope that our proposed algorithm and established dataset will advance the field of TB detection.
Xuebin Xu,Kan Meng,Xiaomin Xing,Chen Chen 한국인터넷정보학회 2022 KSII Transactions on Internet and Information Syst Vol.16 No.3
Palmprint recognition has drawn increasingly attentions in the past decade due to its uniqueness and reliability. Traditional palmprint recognition methods usually use high-resolution images as the identification basis so that they can achieve relatively high precision. However, high-resolution images mean more computation cost in the recognition process, which usually cannot be guaranteed in mobile computing. Therefore, this paper proposes an improved low-resolution palmprint image recognition method based on residual networks. The main contributions include: 1) We introduce a channel attention mechanism to refactor the extracted feature maps, which can pay more attention to the informative feature maps and suppress the useless ones. 2) The ResStage group structure proposed by us divides the original residual block into three stages, and we stabilize the signal characteristics before each stage by means of BN normalization operation to enhance the feature channel. Comparison experiments are conducted on a public dataset provided by the Hong Kong Polytechnic University. Experimental results show that the proposed method achieve a rank-1 accuracy of 98.17% when tested on low-resolution images with the size of 12dpi, which outperforms all the compared methods obviously.
Qiao, Xuebin,Seo, Hyo Jin Elsevier 2014 JOURNAL OF ALLOYS AND COMPOUNDS Vol.615 No.-
<P><B>Abstract</B></P> <P>This paper reports on the luminescence studies of Eu<SUP>3+</SUP>-doped strontium fluoroapatite. The goal of this paper is to establish and characterize the possible sites for Eu<SUP>3+</SUP> substitution in Sr<SUB>5</SUB>(PO<SUB>4</SUB>)<SUB>3</SUB>F lattice. Eu<SUP>3+</SUP>-doped Sr<SUB>5</SUB>(PO<SUB>4</SUB>)<SUB>3</SUB>F phosphor was prepared by the solid state reaction method. The X-ray powder diffraction result of as-synthesized powder phosphor reveals the single phase Sr<SUB>5</SUB>(PO<SUB>4</SUB>)<SUB>3</SUB>F and it also indicates that the incorporation of Eu<SUP>3+</SUP> ions does not affect the crystal structure. Photoluminescence (PL) studies of Eu<SUP>3+</SUP>-doped Sr<SUB>5</SUB>(PO<SUB>4</SUB>)<SUB>3</SUB>F phosphor are performed to study spectral properties of the sample. Site-selective excitation and emission spectra together with the decay curves are investigated by site-selective laser-excitation spectroscopy. The three crystallographic sites for Eu<SUP>3+</SUP> ions are identified in the <SUP>7</SUP>F<SUB>0</SUB> → <SUP>5</SUP>D<SUB>0</SUB> excitation spectra by using a pulsed, tunable, and narrowband dye laser. The luminescence due to the <SUP>5</SUP>D<SUB>0</SUB> → <SUP>7</SUP>F<I> <SUB>J</SUB> </I> (<I>J</I> =1, 2) transitions under excitation at each crystallographic site exhibits its own spectral features. Three crystallographic sites in Sr<SUB>5</SUB>(PO<SUB>4</SUB>)<SUB>3</SUB>F give rise to different crystal-field splits of the <SUP>7</SUP>F<SUB>1</SUB> and <SUP>7</SUP>F<SUB>2</SUB> multiplets. Heterovalent substitution by rare-earth ions (Eu<SUP>3+</SUP>) for Ca<SUP>2+</SUP> positions in the hexagonal crystal lattice requires charge compensation. The charge-compensation mechanism, site symmetry, and the crystal-field strength on Eu<SUP>3+</SUP> in Sr<SUB>5</SUB>(PO<SUB>4</SUB>)<SUB>3</SUB>F are discussed for better understanding of preferential substitution of Eu<SUP>3+</SUP> in the Sr<SUB>5</SUB>(PO<SUB>4</SUB>)<SUB>3</SUB>F lattice.</P> <P><B>Highlights</B></P> <P> <UL> <LI> Site-selective excitation and emission spectra were investigated. </LI> <LI> The three crystallographic sites for Eu<SUP>3+</SUP> ions were identified. </LI> <LI> Three crystallographic sites give rise to different site symmetry. </LI> <LI> Charge compensation and the crystal-field strength on Eu<SUP>3+</SUP> sites were discussed. </LI> </UL> </P>
Human Face Recognition using Multi-Class Projection Extreme Learning Machine
Xu, Xuebin,Wang, Zhixiao,Zhang, Xinman,Yan, Wenyao,Deng, Wanyu,Lu, Longbin The Institute of Electronics and Information Engin 2013 IEIE Transactions on Smart Processing & Computing Vol.2 No.6
An extreme learning machine (ELM) is an efficient learning algorithm that is based on the generalized single, hidden-layer feed-forward networks (SLFNs), which perform well in classification applications. Many studies have demonstrated its superiority over the existing classical algorithms: support vector machine (SVM) and BP neural network. This paper presents a novel face recognition approach based on a multi-class project extreme learning machine (MPELM) classifier and 2D Gabor transform. First, all face image features were extracted using 2D Gabor filters, and the MPELM classifier was used to determine the final face classification. Two well-known face databases (CMU-PIE and ORL) were used to evaluate the performance. The experimental results showed that the MPELM-based method outperformed the ELM-based method as well as other methods.