Search

Your search keyword '"Wu, Tianbo"' showing total 43 results
Start Over Author "Wu, Tianbo"
43 results on '"Wu, Tianbo"'

1. Micro-Expression Recognition Based on Attribute Information Embedding and Cross-modal Contrastive Learning

Author

Song, Yanxin, Wang, Jianzong, Wu, Tianbo, Huang, Zhangcheng, and Xiao, Jing

Subjects
Computer Science - Computer Vision and Pattern Recognition, Computer Science - Artificial Intelligence, Computer Science - Machine Learning
Abstract

Facial micro-expressions recognition has attracted much attention recently. Micro-expressions have the characteristics of short duration and low intensity, and it is difficult to train a high-performance classifier with the limited number of existing micro-expressions. Therefore, recognizing micro-expressions is a challenge task. In this paper, we propose a micro-expression recognition method based on attribute information embedding and cross-modal contrastive learning. We use 3D CNN to extract RGB features and FLOW features of micro-expression sequences and fuse them, and use BERT network to extract text information in Facial Action Coding System. Through cross-modal contrastive loss, we embed attribute information in the visual network, thereby improving the representation ability of micro-expression recognition in the case of limited samples. We conduct extensive experiments in CASME II and MMEW databases, and the accuracy is 77.82% and 71.04%, respectively. The comparative experiments show that this method has better recognition effect than other methods for micro-expression recognition., Comment: This paper has been accepted by IJCNN2022

Published
2022

4. Joint Intent Detection And Slot Filling Based on Continual Learning Model

Author

Hui, Yanfei, Wang, Jianzong, Cheng, Ning, Yu, Fengying, Wu, Tianbo, and Xiao, Jing

Subjects
Computer Science - Computation and Language, Electrical Engineering and Systems Science - Audio and Speech Processing
Abstract

Slot filling and intent detection have become a significant theme in the field of natural language understanding. Even though slot filling is intensively associated with intent detection, the characteristics of the information required for both tasks are different while most of those approaches may not fully aware of this problem. In addition, balancing the accuracy of two tasks effectively is an inevitable problem for the joint learning model. In this paper, a Continual Learning Interrelated Model (CLIM) is proposed to consider semantic information with different characteristics and balance the accuracy between intent detection and slot filling effectively. The experimental results show that CLIM achieves state-of-the-art performace on slot filling and intent detection on ATIS and Snips., Comment: Accepted to ICASSP 2021

Published
2021

5. Analysis on Closing Inrush Current of No-Load Transformers in ±800 kV Tianshan Converter Station

Author

Liu, Lei, Zhang, Fei, Liao, Wenfeng, Zhou, Libing, Wu, Tianbo, Ge, Zhijie, Du, Zhaoguang, Lv, Dongjian, Angrisani, Leopoldo, Series Editor, Arteaga, Marco, Series Editor, Panigrahi, Bijaya Ketan, Series Editor, Chakraborty, Samarjit, Series Editor, Chen, Jiming, Series Editor, Chen, Shanben, Series Editor, Chen, Tan Kay, Series Editor, Dillmann, Rüdiger, Series Editor, Duan, Haibin, Series Editor, Ferrari, Gianluigi, Series Editor, Ferre, Manuel, Series Editor, Hirche, Sandra, Series Editor, Jabbari, Faryar, Series Editor, Jia, Limin, Series Editor, Kacprzyk, Janusz, Series Editor, Khamis, Alaa, Series Editor, Kroeger, Torsten, Series Editor, Li, Yong, Series Editor, Liang, Qilian, Series Editor, Martín, Ferran, Series Editor, Ming, Tan Cher, Series Editor, Minker, Wolfgang, Series Editor, Misra, Pradeep, Series Editor, Möller, Sebastian, Series Editor, Mukhopadhyay, Subhas, Series Editor, Ning, Cun-Zheng, Series Editor, Nishida, Toyoaki, Series Editor, Oneto, Luca, Series Editor, Pascucci, Federica, Series Editor, Qin, Yong, Series Editor, Seng, Gan Woon, Series Editor, Speidel, Joachim, Series Editor, Veiga, Germano, Series Editor, Wu, Haitao, Series Editor, Zamboni, Walter, Series Editor, Zhang, Junjie James, Series Editor, Xie, Kaigui, editor, Hu, Jianlin, editor, Yang, Qingxin, editor, and Li, Jian, editor

Published
2023
Full Text
View/download PDF

6. Study on Inrush Current and Overvoltage Caused by AC Filter Circuit Breakers Closing in ±1100 kV UHVDC Converter Station

Author

Zhang, Fei, Liu, Lei, Shan, Huaping, Zhang, Weining, Wu, Tianbo, Li, Longfei, Du, Zhaoguang, Gao, Hongguang, Angrisani, Leopoldo, Series Editor, Arteaga, Marco, Series Editor, Panigrahi, Bijaya Ketan, Series Editor, Chakraborty, Samarjit, Series Editor, Chen, Jiming, Series Editor, Chen, Shanben, Series Editor, Chen, Tan Kay, Series Editor, Dillmann, Rüdiger, Series Editor, Duan, Haibin, Series Editor, Ferrari, Gianluigi, Series Editor, Ferre, Manuel, Series Editor, Hirche, Sandra, Series Editor, Jabbari, Faryar, Series Editor, Jia, Limin, Series Editor, Kacprzyk, Janusz, Series Editor, Khamis, Alaa, Series Editor, Kroeger, Torsten, Series Editor, Li, Yong, Series Editor, Liang, Qilian, Series Editor, Martín, Ferran, Series Editor, Ming, Tan Cher, Series Editor, Minker, Wolfgang, Series Editor, Misra, Pradeep, Series Editor, Möller, Sebastian, Series Editor, Mukhopadhyay, Subhas, Series Editor, Ning, Cun-Zheng, Series Editor, Nishida, Toyoaki, Series Editor, Oneto, Luca, Series Editor, Pascucci, Federica, Series Editor, Qin, Yong, Series Editor, Seng, Gan Woon, Series Editor, Speidel, Joachim, Series Editor, Veiga, Germano, Series Editor, Wu, Haitao, Series Editor, Zamboni, Walter, Series Editor, Zhang, Junjie James, Series Editor, Xie, Kaigui, editor, Hu, Jianlin, editor, Yang, Qingxin, editor, and Li, Jian, editor

Published
2023
Full Text
View/download PDF

10. Chinese Punctuation Prediction with Adaptive Attention and Dependency Tree

Author

Yan, Zelong, Wang, Jianzong, Cheng, Ning, Wu, Tianbo, Xiao, Jing, Filipe, Joaquim, Editorial Board Member, Ghosh, Ashish, Editorial Board Member, Prates, Raquel Oliveira, Editorial Board Member, Zhou, Lizhu, Editorial Board Member, Chen, Huajun, editor, Liu, Kang, editor, Sun, Yizhou, editor, Wang, Suge, editor, and Hou, Lei, editor

Published
2021
Full Text
View/download PDF

11. IDRiD: Diabetic Retinopathy – Segmentation and Grading Challenge

Author

Porwal, Prasanna, Pachade, Samiksha, Kokare, Manesh, Deshmukh, Girish, Son, Jaemin, Bae, Woong, Liu, Lihong, Wang, Jianzong, Liu, Xinhui, Gao, Liangxin, Wu, TianBo, Xiao, Jing, Wang, Fengyan, Yin, Baocai, Wang, Yunzhi, Danala, Gopichandh, He, Linsheng, Choi, Yoon Ho, Lee, Yeong Chan, Jung, Sang-Hyuk, Li, Zhongyu, Sui, Xiaodan, Wu, Junyan, Li, Xiaolong, Zhou, Ting, Toth, Janos, Baran, Agnes, Kori, Avinash, Chennamsetty, Sai Saketh, Safwan, Mohammed, Alex, Varghese, Lyu, Xingzheng, Cheng, Li, Chu, Qinhao, Li, Pengcheng, Ji, Xin, Zhang, Sanyuan, Shen, Yaxin, Dai, Ling, Saha, Oindrila, Sathish, Rachana, Melo, Tânia, Araújo, Teresa, Harangi, Balazs, Sheng, Bin, Fang, Ruogu, Sheet, Debdoot, Hajdu, Andras, Zheng, Yuanjie, Mendonça, Ana Maria, Zhang, Shaoting, Campilho, Aurélio, Zheng, Bin, Shen, Dinggang, Giancardo, Luca, Quellec, Gwenolé, and Mériaudeau, Fabrice

Published
2020
Full Text
View/download PDF

19. Experimental research on short electric arc milling machining based on the mechanical-fluid coupling effect.

Author

Li, Xuezhi, Li, Zewen, Wu, Tianbo, and Zhou, Jianping

Abstract

Short electric arc milling (SEAM) involves the complex coupling between the high-density heat output of the arc, the high-speed rotation of the tool electrode, and the high-pressure flushing fluid inside the tubular electrode. To improve the arc discharge utilization, this study investigates the effects of mechanical action, fluid action, and mechanical-fluid coupling action on the arc breakage of short arcs via experiments. It is found that the high-speed rotation of the electrode reduces the adsorption of the working fluid to the electrode surface, which can help to reduce the recast of molten metal on the workpiece surface and improve the surface quality. The coupling of the two makes a greater contribution to arc breaking and molten metal removal, while also improving the spatial continuity of the arc discharge. Moreover, after SEAM based on the mechanical-fluid coupling effect, the nickel-based alloy GH4169 is found to have lower surface roughness, lower tool electrode wear, a higher material removal rate, and a thinner recast layer. [ABSTRACT FROM AUTHOR]

Published
2023
Full Text
View/download PDF

20. Research on the Policy Effects and Impact Mechanisms of the Belt and Road Initiative on China's Forest Products Trade.

Author

Wu, Tianbo, Zhou, Dan, Cheng, Hao, Zhang, Keqiu, and Jiang, Yihao

Abstract

The Belt and Road Initiative, as an important measure for China in terms of opening up and participating in international economic and trade cooperation, has become a new driving force for the sustainable development of China's forest products trade. This paper takes the Belt and Road Initiative as a policy event and evaluates its policy effects on the development of China's forest products trade from the causal level through the difference-in-differences model (DID), explores the policy effect in detail from the perspectives of product heterogeneity and regional heterogeneity, and clarifies the specific impact mechanism. The main results are as follows: (1) there is a significant policy promotion effect of the Belt and Road Initiative on the growth of the bilateral trade scale of forest products between China and the countries along the route. (2) In terms of product structure, the policy promotion effect of the Belt and Road Initiative is mainly manifested in processed wood products. (3) In terms of regional distribution, the policy promotion effects of the Belt and Road Initiative are mainly concentrated in Europe, Africa, and Asia. (4) The "logistics performance, political partnership with China, and Internet penetration" of trading countries play a significant positive mediating role in the policy effects of the Belt and Road Initiative. Therefore, in view of the significant role of the Belt and Road Initiative in promoting the development of bilateral forest products trade, China should promote more countries to participate in the joint construction of the Belt and Road and tap new momentum for the development of the forest products trade by focusing on key countries, priority areas, and key products. [ABSTRACT FROM AUTHOR]

Published
2023
Full Text
View/download PDF

24. Joint Intent Detection and Slot Filling Based on Continual Learning Model

Author

Cheng Ning, Jianzong Wang, Yu Fengying, Xiao Jing, Hui Yanfei, and Wu Tianbo

Subjects
FOS: Computer and information sciences, Signal processing, Computer Science - Computation and Language, Computer science, business.industry, Natural language understanding, Continual learning, Semantics, computer.software_genre, Machine learning, Field (computer science), Audio and Speech Processing (eess.AS), FOS: Electrical engineering, electronic engineering, information engineering, Artificial intelligence, Semantic information, Joint (audio engineering), business, Computation and Language (cs.CL), computer, Natural language, Electrical Engineering and Systems Science - Audio and Speech Processing
Abstract

Slot filling and intent detection have become a significant theme in the field of natural language understanding. Even though slot filling is intensively associated with intent detection, the characteristics of the information required for both tasks are different while most of those approaches may not fully aware of this problem. In addition, balancing the accuracy of two tasks effectively is an inevitable problem for the joint learning model. In this paper, a Continual Learning Interrelated Model (CLIM) is proposed to consider semantic information with different characteristics and balance the accuracy between intent detection and slot filling effectively. The experimental results show that CLIM achieves state-of-the-art performace on slot filling and intent detection on ATIS and Snips., Comment: Accepted to ICASSP 2021

Published
2021

28. Chinese Punctuation Prediction with Adaptive Attention and Dependency Tree

Author

Wu Tianbo, Cheng Ning, Yan Zelong, Xiao Jing, and Wang Jianzong

Subjects
Conditional random field, Standardization, business.industry, Computer science, media_common.quotation_subject, computer.software_genre, Punctuation, Tree (data structure), Dependency grammar, Word representation, Key (cryptography), Artificial intelligence, Dependency tree, business, computer, Natural language processing, media_common
Abstract

Punctuation prediction, as a key step of connecting speech recognition and natural language processing, has a profound impact on subsequent tasks. Although methods of bidirectional long short term memory and conditional random fields (BiLSTM+CRF) are proposed and remain advanced for a long time, it still suffers from the lack of ability of capturing long-distance interactions among words and extracting useful semantic information, especially in Chinese punctuation prediction. In this paper, considering the characteristic of Chinese punctuation symbols, we propose a novel method of punctuation standardization. In our BiLSTM+CRF based network, adaptive attention and dependency parsing tree are utilized to capture the long distance interactions and extract useful semantic information, and thus enhancing the word representation. As for the performance, the first proposal of Chinese punctuation prediction outperforms BiLSTM+CRF with a gain of 0.292% and 0.127% on accuracy in two datasets respectively. The second proposal outperforms existing methods with a gap of above 4.5% of accuracy and reaches state-of-the-art performance in two datasets.

Published
2021

31. Automatic Annotation of Text Classification Data Set in Specific Field Using Named Entity Recognition

Author

Wu Tianbo, Chen Haiqiang, Zhou Xin, He Xiaohai, and Yang Qiang

Subjects
Set (abstract data type), Annotation, Information retrieval, Text mining, Machine translation, Named-entity recognition, business.industry, Computer science, Question answering, business, computer.software_genre, computer, Field (computer science)
Abstract

Natural language processing has set off another development boom in the field of artificial intelligence. Intelligent question answering, machine translation, and dialogue systems are all important research contents in the field of natural language processing. Among them, text mining is an important branch in the field of natural language processing, including text classification, named entity recognition and other technologies. In actual research and engineering projects, obtaining a large number of reliable text data sets is the guarantee for these tasks. In the industrial field, the production of text data sets is usually done by manual sorting and labeling, but manual labeling consumes a lot of manpower and time, making the text data set not ideal owing to the different subjective thoughts of each person. Consequently, it can not meet the needs of the actual industrial field. This paper proposes an automatic annotation method based on the number of named entities for text datasets, which can quickly obtain a large number of text datasets in the field of terrorism, and ensure the smooth progress of research and engineering projects.

Published
2019

32. Study on the Time Constant of DC Component and Its Influence on Short-circuit Breaking Capacity of Circuit Breaker

Author

Wu Tianbo, Zhou Libing, Li Shan, Bai Yunxiao, and Wentao Chen

Subjects
Physics, Electric power system, Breaking capacity, Control theory, Component (UML), Direct current, Time constant, Fault (power engineering), Circuit breaker, DC bias
Abstract

By introducing the time constant of the direct current(DC) component, which is specified by the specification, and the derivation of the DC component, we concluded that the time constant of DC component will affect the short-circuit breaking capacity. We introduced the influences of the time constant of DC component are the structure, the parameters and the loss of the power system, as well as the type of the fault. Combined with an accident of a power plant and some other cases, we proved that time constant which is beyond the specification will affect the short-circuit breaking capacity, and may further affect the safety of the power system.

Published
2018

36. Design of low-voltage switch phase-sensitive protector based on STM32

Author

Wu Tianbo, Wu Yaohui, Liu Si, and Wu Haozhen

Subjects
business.industry, Computer science, Electrical engineering, Power factor, Timer, AC power, business, Low voltage, Induction motor, Overcurrent, Voltage, Power (physics)
Abstract

To distinguish the short circuit fault overcurrent and the asynchronous motor starting overcurrent in power supply system protection, this paper described a phase-sensitive detection protector based on signal-input-capture function of the timer in STM32 chip. Through many theories, experiments and analyses, the range of the power factor used for discriminant of asynchronous motor starting was set to 0.4–0.5 in protector. The protector used the power factor as a discrimination criterion. The voltage and current sinusoidal signals were transformed to recognizable signals by the hardware, which were input to the timer and then the relevant data were gotten. At last, the data were dealt by the system task in software, gotten was the power factor of system voltage and current, which was afterwards compared with the set value to distinguish the two kinds of overcurrent. By testing, the protector measured the power factors of system accurately, which met the protection requirements.

Published
2017

38. The economic strength of four EC countries

Author

Wu Tianbo

Subjects
Business, European Union -- Economic aspects
Abstract

It is essential for the European countries to unite to gain economic power since they have to compete with Japan and the United States. Germany, France Italy and Britain want to quicken the process of integration, but their efforts are hampered because each country has its own problems. Germany has to rebuild itself after unification, and is finding it difficult to compete economically with the Japanese whose commodities are superior in quality. France and Italy are not able to main high exchange rates and Britain's economy is at its lowest due to recession. These problems have to be overcome for the integration and economic unity of EC countries.

Published
1993

39. Experimental study in SEAM machining performance of W-Cu alloy electrode materials.

Author

Li, Xuezhi, Zhou, Jianping, Xu, Yan, Zhu, Chenguang, Wang, Kedian, and Wu, Tianbo

Subjects
MACHINE performance, SCANNING electron microscopes, ELECTRIC arc, ALLOYS, ELECTRODES, MATERIAL erosion, ELECTRIC metal-cutting
Abstract

To solve problems such as insufficient energy and effective arc burning during the injection of a working medium in a short arc discharge channel, the application of a W-Cu alloy as the electrode material in short electrical arc machining (SEAM) is explored in this paper. Using the surface roughness, material removal rate, and tool electrode wear rate as technical indices, the effects of voltage, frequency, and duty cycle on W-Cu alloy materials are studied to explore machining characteristics in SEAM. Contrast experiments on a graphite electrode were conducted using a SU8010 scanning electron microscope and energy spectrum analysis to study the erosion mechanism of the W-Cu alloy material. This paper addresses applications of copper-based composites in SEAM and provides a theoretical basis for the development of electrode materials in the SEAM. [ABSTRACT FROM AUTHOR]

Published
2019
Full Text
View/download PDF

40. Arc characteristics in short electrical arc high-efficiency milling for GH4169

Author

Li, Xuezhi, Zhou, Jianping, Wang, Kedian, Wu, Tianbo, Xu, Yan, and Hu, Guoyu

Abstract

To solve core problems of slow arc response and unstable combustion, arc movement characteristics and arc action rules in short electrical arc milling are studied using a high-speed acquisition system and the Phantom high-speed camera in this article. To improve arc reignition, the characteristic parameters of arc discharge are determined as a discharge gap of 0.053 mm and discharge time of 2180 μs. A short electrical arc milling experiment with nickel-based superalloy GH4169 is carried out, confirming that the continuous stable discharge of the arc can easily produce a material removal rate of 40 g/min and a tool wear ratio of 0.8%. An SU8010 scanning electron microscope and energy spectrum analyzer are then used to study the erosion mechanism of GH4169 in short electrical arc milling. This article aims to expand the processing range of short electrical arc machining in the field of special material processing to provide experimental and theoretical bases for the development of short electrical arc milling technology theory and methods.

Published
2024
Full Text
View/download PDF

43. Enhancement method with naturalness preservation and artifact suppression based on an improved Retinex variational model for color retinal images.

Author

Han R, Tang C, Xu M, Liang B, Wu T, and Lei Z

Subjects
Artifacts, Image Interpretation, Computer-Assisted methods, Image Enhancement methods, Algorithms
Abstract

Retinal images are widely used for the diagnosis of various diseases. However, low-quality retinal images with uneven illumination, low contrast, or blurring may seriously interfere with diagnosis by ophthalmologists. This study proposes an enhancement method for low-quality retinal color images. In this paper, an improved variational Retinex model for color retinal images is first proposed and applied to each channel of the RGB color space to obtain the illuminance and reflectance layers. Subsequently, the Naka-Rushton equation is introduced to correct the illumination layer, and an enhancement operator is constructed to improve the clarity of the reflectance layer. Finally, the corrected illuminance and enhanced reflectance are recombined. Contrast-limited adaptive histogram equalization is introduced to further improve the clarity and contrast. To demonstrate the effectiveness of the proposed method, this method is tested on 527 images from four publicly available datasets and 40 local clinical images from Tianjin Eye Hospital (China). Experimental results show that the proposed method outperforms the other four enhancement methods and has obvious advantages in naturalness preservation and artifact suppression.

Published
2023
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results