Search

Your search keyword '"Xu, Weifeng"' showing total 1,724 results
Start Over Author "Xu, Weifeng"
1,724 results on '"Xu, Weifeng"'

1. Cross Branch Feature Fusion Decoder for Consistency Regularization-based Semi-Supervised Change Detection

Author

Xing, Yan, Xu, Qi'ao, Zeng, Jingcheng, Huang, Rui, Gao, Sihua, Xu, Weifeng, Zhang, Yuxiang, and Fan, Wei

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Semi-supervised change detection (SSCD) utilizes partially labeled data and a large amount of unlabeled data to detect changes. However, the transformer-based SSCD network does not perform as well as the convolution-based SSCD network due to the lack of labeled data. To overcome this limitation, we introduce a new decoder called Cross Branch Feature Fusion CBFF, which combines the strengths of both local convolutional branch and global transformer branch. The convolutional branch is easy to learn and can produce high-quality features with a small amount of labeled data. The transformer branch, on the other hand, can extract global context features but is hard to learn without a lot of labeled data. Using CBFF, we build our SSCD model based on a strong-to-weak consistency strategy. Through comprehensive experiments on WHU-CD and LEVIR-CD datasets, we have demonstrated the superiority of our method over seven state-of-the-art SSCD methods., Comment: 5 pages, 4 figures, accepted by ICASSP 2024

Published
2024
Full Text
View/download PDF

2. RCDM: Enabling Robustness for Conditional Diffusion Model

Author

Xu, Weifeng, Zhu, Xiang, and Li, Xiaoyong

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

The conditional diffusion model (CDM) enhances the standard diffusion model by providing more control, improving the quality and relevance of the outputs, and making the model adaptable to a wider range of complex tasks. However, inaccurate conditional inputs in the inverse process of CDM can easily lead to generating fixed errors in the neural network, which diminishes the adaptability of a well-trained model. The existing methods like data augmentation, adversarial training, robust optimization can improve the robustness, while they often face challenges such as high computational complexity, limited applicability to unknown perturbations, and increased training difficulty. In this paper, we propose a lightweight solution, the Robust Conditional Diffusion Model (RCDM), based on control theory to dynamically reduce the impact of noise and significantly enhance the model's robustness. RCDM leverages the collaborative interaction between two neural networks, along with optimal control strategies derived from control theory, to optimize the weights of two networks during the sampling process. Unlike conventional techniques, RCDM establishes a mathematical relationship between fixed errors and the weights of the two neural networks without incurring additional computational overhead. Extensive experiments were conducted on MNIST and CIFAR-10 datasets, and the results demonstrate the effectiveness and adaptability of our proposed model.

Published
2024

3. A Saliency Enhanced Feature Fusion based multiscale RGB-D Salient Object Detection Network

Author

Huang, Rui, Zhao, Qingyi, Xing, Yan, Gao, Sihua, Xu, Weifeng, Zhang, Yuxiang, and Fan, Wei

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Multiscale convolutional neural network (CNN) has demonstrated remarkable capabilities in solving various vision problems. However, fusing features of different scales alwaysresults in large model sizes, impeding the application of multiscale CNNs in RGB-D saliency detection. In this paper, we propose a customized feature fusion module, called Saliency Enhanced Feature Fusion (SEFF), for RGB-D saliency detection. SEFF utilizes saliency maps of the neighboring scales to enhance the necessary features for fusing, resulting in more representative fused features. Our multiscale RGB-D saliency detector uses SEFF and processes images with three different scales. SEFF is used to fuse the features of RGB and depth images, as well as the features of decoders at different scales. Extensive experiments on five benchmark datasets have demonstrated the superiority of our method over ten SOTA saliency detectors., Comment: Accpeted by 2024 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2024)

Published
2024

9. An Empirical Study on the Use of Diagnosis Related Group Tools for Grouping Adjustments in Large Public Hospitals

Author

ZHANG Guojie, TAN Xutong, CAI Zhiling, XU Qiang, XU Weifeng, CHEN Yihang, WANG Yating, LIU Jinhan, CHEN Zheng, ZHOU Jiong, and MA Xiaojun

Subjects
diagnosis related group, grouping adjustment, case mix index, hospital resource integration, Medicine
Abstract

ObjectiveTo analyze the disease group structure and its trends in key departments of large public hospitals using diagnosis related group (DRG) data, explore the key points of intervention and optimization of disease groups in departments, and further promote the rational allocation of department resources.MethodsWe retrospectively collected DRG data from two surgical departments in a large public hospital in Beijing from 2017 to 2023. When the case mix index (CMI) of the two surgical departments declined, interventions such as performance appraisal, department education, and hospital publicity were promptly adopted. The changesin CMI values were observed and the trends in disease group weights, time consumption index, cost consumption index, and mortality rate in low-risk groups were analyzed.ResultsAfter the interventions, in surgical department Ⅰ, the proportion of patients with lower-weight diseases, such as major thyroid surgery (KD1), significantly decreased, while that of patients with higher-weight diseases, such as colorectal malignancy surgery (GB2) and pancreatic malignancy surgery (HB1), significantly increased. In surgical department Ⅱ, the proportion of patients with lower-weight diseases, such as chemotherapy (RE1), decreased markedly, while that of patients with higher-weight diseases, including major surgery for malignancy of kidney, ureter, and bladder (LA1), adrenal gland surgery (KC1), surgery for kidney/ureter/bladder except for major malignancy surgery (LB1), and male genital organ malignancy surgery (MA1), increased significantly. Both surgical departments achieved the goal of increasing their CMI values. In terms of efficiency, cost, and quality indicators, the time consumption index and cost consumption index of the two surgical departments were significantly lower than 1, and the mortality rate in low-risk groups was 0.ConclusionBased on actual conditions and development goals, large public hospitals can achieve improvements in CMI values and optimization of disease group structures through reasonable interventions, thereby enhancing medical efficiency and rational utilization of resources.

Published
2024
Full Text
View/download PDF

10. Multi-source DEM accuracy evaluation based on ICESat-2 in Qinghai-Tibet Plateau, China

Author

Xu Weifeng, Li Jun, Peng Dailiang, Jiang Jinge, Xia Hongxuan, Yin Hongyue, and Yang Jun

Subjects
Digital elevation models, ICESat-2 ATL08, vertical accuracy assessment, Qinghai-Tibet Plateau, Mathematical geography. Cartography, GA1-1776
Abstract

ABSTRACTDigital Elevation Models (DEMs) are critical datasets in the field of Earth sciences, essential for accurate measurement and analysis of the Earth's surface. We used ICESat-2 to quantitatively evaluate 5 DEMs (ALOS PALSAR, ASTER GDEM V3, COPERNICUS, NASADEM and TanDEM-X) on the Tibetan Plateau. The research findings indicate that the ALOS exhibits the highest level of accuracy, as evidenced by its root mean square error (RMSE) value of 5.05m. It is closely followed by NASA and COP, which have RMSE values of 6.23m and 8.10m, respectively. In contrast, the ASTER and TDX90 demonstrate comparatively lower levels of accuracy, as indicated by their respective RMSE values of 11.47m and 12.32m. It is worth mentioning that the accuracy of DEMs is significantly influenced by land cover, especially during the transition from areas with low vegetation to those with high vegetation. This transition often results in a decrease in accuracy. Furthermore, the accuracy of DEMs tends to decrease with increasing slope values. Aspect exhibits a notable spatial distribution pattern characterized by a "low in the Southwest direction, high in all other directions" phenomenon. This study offers significant contributions to the evaluation of DEMs accuracy and its suitability in the Qinghai-Tibet Plateau.

Published
2024
Full Text
View/download PDF

11. Federated Coordinate Descent for Privacy-Preserving Multiparty Linear Regression

Author

Leng, Xinlin, Li, Chenxu, Xu, Weifeng, Sun, Yuyan, and Wang, Hongtao

Subjects
Computer Science - Machine Learning, 68Q32, I.2.11
Abstract

Distributed privacy-preserving regression schemes have been developed and extended in various fields, where multiparty collaboratively and privately run optimization algorithms, e.g., Gradient Descent, to learn a set of optimal parameters. However, traditional Gradient-Descent based methods fail to solve problems which contains objective functions with L1 regularization, such as Lasso regression. In this paper, we present Federated Coordinate Descent, a new distributed scheme called FCD, to address this issue securely under multiparty scenarios. Specifically, through secure aggregation and added perturbations, our scheme guarantees that: (1) no local information is leaked to other parties, and (2) global model parameters are not exposed to cloud servers. The added perturbations can eventually be eliminated by each party to derive a global model with high performance. We show that the FCD scheme fills the gap of multiparty secure Coordinate Descent methods and is applicable for general linear regressions, including linear, ridge and lasso regressions. Theoretical security analysis and experimental results demonstrate that FCD can be performed effectively and efficiently, and provide as low MAE measure as centralized methods under tasks of three types of linear regressions on real-world UCI datasets., Comment: 14 pages, 19 figures (Under review)

Published
2022

13. Predicting Code Vulnerability Types via Heterogeneous GNN Learning

Author

Luo, Yu, Xu, Weifeng, Xu, Dianxiang, Goos, Gerhard, Series Editor, Hartmanis, Juris, Founding Editor, Bertino, Elisa, Editorial Board Member, Gao, Wen, Editorial Board Member, Steffen, Bernhard, Editorial Board Member, Yung, Moti, Editorial Board Member, Garcia-Alfaro, Joaquin, editor, Kozik, Rafał, editor, Choraś, Michał, editor, and Katsikas, Sokratis, editor

Published
2024
Full Text
View/download PDF

14. A Mixed Hypergraph Convolutional Network for Session-Based Recommendation

Author

Li, Jianfu, Zhang, Dan, Gao, Sihua, Xu, Weifeng, Goos, Gerhard, Series Editor, Hartmanis, Juris, Founding Editor, Bertino, Elisa, Editorial Board Member, Gao, Wen, Editorial Board Member, Steffen, Bernhard, Editorial Board Member, Yung, Moti, Editorial Board Member, Huang, De-Shuang, editor, Zhang, Chuanlei, editor, and Pan, Yijie, editor

Published
2024
Full Text
View/download PDF

15. A Practical Survey of Data Carving from Non-Functional Android Phones Using Chip-Off Technique

Author

Shaikh, Sarfraz, Deng, Lin, Xu, Weifeng, Kacprzyk, Janusz, Series Editor, Pal, Nikhil R., Advisory Editor, Bello Perez, Rafael, Advisory Editor, Corchado, Emilio S., Advisory Editor, Hagras, Hani, Advisory Editor, Kóczy, László T., Advisory Editor, Kreinovich, Vladik, Advisory Editor, Lin, Chin-Teng, Advisory Editor, Lu, Jie, Advisory Editor, Melin, Patricia, Advisory Editor, Nedjah, Nadia, Advisory Editor, Nguyen, Ngoc Thanh, Advisory Editor, Wang, Jun, Advisory Editor, and Latifi, Shahram, editor

Published
2024
Full Text
View/download PDF

24. Identification of neutrophil extracellular trap-driven gastric cancer heterogeneity and C5AR1 as a therapeutic target

Author

Zhao Jing, Li Xiangyu, Li Liming, Chen Beibei, Xu Weifeng, He Yunduan, and Chen Xiaobing

Subjects
neutrophil extracellular trap, gastric cancer, heterogeneity, personalized therapy, C5AR1, treatment target, Biochemistry, QD415-436, Genetics, QH426-470
Abstract

Neutrophil extracellular traps (NETs) are implicated in gastric cancer (GC) growth, metastatic dissemination, cancer-associated thrombosis, etc. This work is conducted to elucidate the heterogeneity of NETs in GC. The transcriptome heterogeneity of NETs is investigated in TCGA-STAD via a consensus clustering algorithm, with subsequent external verification in the GSE88433 and GSE88437 cohorts. Clinical and molecular traits, the immune microenvironment, and drug response are characterized in the identified NET-based clusters. Based upon the feature genes of NETs, a classifier is built for estimating NET-based clusters via machine learning. Multiple experiments are utilized to verify the expressions and implications of the feature genes in GC. A novel NET-based classification system is proposed for reflecting the heterogeneity of NETs in GC. Two NET-based clusters have unique and heterogeneous clinical and molecular features, immune microenvironments, and responses to targeted therapy and immunotherapy. A logistic regression model reliably differentiates the NET-based clusters. The feature genes C5AR1, CSF1R, CSF2RB, CYBB, HCK, ITGB2, LILRB2, MNDA, MPEG1, PLEK, SRGN, and STAB1 are proven to be aberrantly expressed in GC cells. Specific knockdown of C5AR1 effectively hinders GC cell growth and elicits intracellular ROS accumulation. In addition, its suppression suppresses the aggressiveness and EMT phenotype of GC cells. In all, NETs are the main contributors to intratumoral heterogeneity and differential drug sensitivity in GC, and C5AR1 has been shown to trigger GC growth and metastatic spread. These findings collectively provide a theoretical basis for the use of anti-NETs in GC treatment.

Published
2024
Full Text
View/download PDF

27. Safety, tolerability, and immunogenicity of an adult pneumococcal conjugate vaccine, V116 (STRIDE-3): a randomised, double-blind, active comparator controlled, international phase 3 trial

Author

Abdel-Malek, Michael, Adelglass, Jeffrey, Alpizar, Sady Armad, Anderson, Duane, Andrews, Charles P., Askling, Helena Hervius, Baker, Jeffrey B., Baron, Mira, Barranco-Santana, Elizabeth A., Beckes, John, Bergthold, James H., Bosson, Peter, Buntinx, Erik, Bush, Larry Marc, Butuk, David J., Cartagena Ayala, Edgardo F., Choi, Hee Jung, Choi, Jung Hyun, Choi, Won Suk, Choi, Young Hwa, Clark, James R., Cornely, Oliver A., Cox, Steve E., Daboul, Nizar, De La Cruz, Luis Ignacio, De Silva, Pamela, Degen, Olaf, Ebuh, Valentine, Elzi, Stanton R., Fiel, Thomas, Fitzgibbons, William Patrick, Fraser, Neil J., Freeman, George Hartley, Galloway, Christopher, Garcia-Huidobro, Diego, Garza, Juan Carlos, George, Anil, Griffin, Carl Patrick, Guduri, Sridhar, Guner, Hatice Rahmet, Gunner, Elizabeth, Hanabergh, Enrique, Harper, Charles Harold, Harris, Kathleen, Hemmersmeier, John Matthew, Hill, John M., Huang, Ching-Tai, Huang, Kuo-Chin, Huberty, Chad Roger, Image, Bejamin James, Jandl, Margret, Johnston, William Henry, Jr., Kamerbeek, Jackie Marie, Karabay, Oguz, Ko, Jae-Hoon, Koksal, Iftihar, Kwon, Ki Tae, Leah, Amber, Lee, Dong-Gun, Lee, Jacob, Lenzmeier, Thomas C., Loh, Denissa, Lupercio, Rafael, Luts, Andes Erik, Lutz, Thomas, Martinez, Jose, Mayer, Renae, Miller, Vicki, Morales-Ramirez, Javier O., Munoz, Manuel, Navarro, Jesus Ovidio, Odekirk, Larry Lee, Odia, Patrick, Odio, Alberto J., O'Sullivan, Matthew, Patel, Minesh Arun, Pelayo, Enrique, Perez Cortes, Carlos, Pervaiz, Syed, Probst, Jared, Quinn, Dean Richard, Rigonan, Kathryn R., Rosario-Matos, Nicolas, Rosen, Jeffrey Bruce, Sannesten, Andrej, Schaefer, Axel, Schellberg, Sven, Sharma, Manjoo, Sher, Lawrence D., Shi, Hyejin, Shockey, Gerald R., Singh, Raj, Sjoberg, Folke, Smith, Brian, Smith, Christohper A., Smith, Deon, Smith, Susan Elizabeth, Song, Joon Young, Staben, Jonathan, Steil, Kenneth, Su, Chien-Tien, Ünal, Serhat, Wie, Seong-Heong, Wijewardane, Priyantha, Williams, Michael John, Wolf, Thomas A., Yang, Yi-Ching, Yeom, Joon Sup, Platt, Heather L, Bruno, Christopher, Barranco-Santana, Elizabeth A, Grijalva, Carlos G, Orenstein, Walter A, Morgan, Leslie, Fernsler, Doreen, Xu, Weifeng, Waleed, Muhammad, Li, Jianing, and Buchwald, Ulrike K

Published
2024
Full Text
View/download PDF

41. Forensic Analysis of Contents in Thumbnails Using Transfer Learning

Author

Sayyafzadeh, Shahrzad, Xu, Weifeng, Chi, Hongmei, Kacprzyk, Janusz, Series Editor, Gomide, Fernando, Advisory Editor, Kaynak, Okyay, Advisory Editor, Liu, Derong, Advisory Editor, Pedrycz, Witold, Advisory Editor, Polycarpou, Marios M., Advisory Editor, Rudas, Imre J., Advisory Editor, Wang, Jun, Advisory Editor, and Arai, Kohei, editor

Published
2023
Full Text
View/download PDF

42. Analysis and Application of Investment Cost of 10 kV and Below Power Grid Infrastructure Projects

Author

Xu, Weifeng, Ye, Shangxing, Huang, Jian, Wen, Hongjun, Qiu, Yi, Pan, Xia, Chen, Bowei, Appolloni, Andrea, Series Editor, Caracciolo, Francesco, Series Editor, Ding, Zhuoqi, Series Editor, Gogas, Periklis, Series Editor, Huang, Gordon, Series Editor, Nartea, Gilbert, Series Editor, Ngo, Thanh, Series Editor, Striełkowski, Wadim, Series Editor, Jiang, Yushi, editor, Li, Guangming, editor, and Li, Wilson Xinbao, editor

Published
2023
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results