16 results on '"Kong, Heng"'
Search Results
2. Jointly sparse fast hashing with orthogonal learning for large-scale image retrieval
- Author
-
Xu, Honghao, Lai, Zhihui, and Kong, Heng
- Published
- 2023
- Full Text
- View/download PDF
3. Two-dimensional jointly sparse robust discriminant regression
- Author
-
Lai, Zhihui, Yu, Zhuozhen, Kong, Heng, and Shen, Linlin
- Published
- 2021
- Full Text
- View/download PDF
4. Robust supervised sparse representation for face recognition
- Author
-
Mi, Jian-Xun, Sun, Yueru, Lu, Jia, and Kong, Heng
- Published
- 2020
- Full Text
- View/download PDF
5. The MLS based numerical manifold method for bending analysis of thin plates on elastic foundations.
- Author
-
Zhao, Shuaixing, Kong, Heng, and Zheng, Hong
- Subjects
- *
ELASTIC foundations , *ELASTIC plates & shells , *LEAST squares , *FINITE element method , *DEGREES of freedom - Abstract
Compared with the finite element method, H2-regularity in the Galerkin based approximation to the Kirchhoff thin plate model can be easily realized using either the moving least squares (MLS) or the generalized moving least squares (GMLS), which take the Lagrange form and the Hermite form, respectively. Coupling (G)MLS with the numerical manifold method (NMM) can greatly improve numerical properties of NMM in the treatment of plates of complicated shape, thereby denoted by MLS-NMM and GMLS-NMM. In the (G)MLS-NMM, the mathematical cover is composed of simply connected and partially overlapped mathematical patches that are the influence domains of (G)MLS nodes. GMLS-NMM appears to better fit to the Kirchhoff plate because it is equipped with rotation angle degrees of freedom. Through numerical tests and theoretical analysis in solving problems of thin plates on elastic foundations, however, this study shows that MLS-NMM is much more advantageous over GMLS-NMM from the aspects of both accuracy and memory usage. [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF
6. The numerical manifold method for harmonic wave propagation in unbounded domains.
- Author
-
Wang, Fangyi, Kong, Heng, and Zheng, Hong
- Subjects
- *
THEORY of wave motion , *ACOUSTIC surface waves , *WATER waves , *INFINITE element method , *FINITE element method , *ELASTIC wave propagation - Abstract
The numerical manifold method (NMM) has been applied successfully to a wide variety of problems, including time-independent exterior problems, where problem domains are unbounded. However, of the utmost interest and significance is the analysis of wave propagation in unbounded domains, which is still an open issue because it is far more difficult than time-independent exterior problems. This study aims to fill partly this gap from the prospective of NMM. A new type of infinite patches and the local approximations are designed to construct the Galerkin approximation that satisfies the asymptotic behavior of waves at infinite. Different from the infinite element technique in the finite element method (FEM), where the shape functions are required to satisfy not only the continuity between finite and infinite elements but also the attenuation behavior of solution while approaching infinite, the NMM is demonstrated more straightforward and elegant in the construction of the Galerkin approximation. Some examples in propagation of elastic wave and surface water wave are investigated to verify the accuracy and applicability of the proposed method. [ABSTRACT FROM AUTHOR]
- Published
- 2022
- Full Text
- View/download PDF
7. Global analysis approach of stability of deep foundation pit slopes reinforced by underground diaphragm walls and prestressed anchor cables.
- Author
-
Kong, Heng, Dong, Miao, Cao, Xitailang, Lin, Shan, Zhao, ShuaiXing, and Zheng, Hong
- Subjects
- *
BUILDING foundations , *BORED piles , *DIAPHRAGM walls , *SAFETY factor in engineering , *SLOPE stability , *CABLE structures - Abstract
The stability evaluation of deep foundation pit slopes is crucial, including the solution of the factor of safety and the location of the critical slip surface. Prestressed anchor cables and underground diaphragm walls are commonly used to reinforce pit slopes. By means of the dimensional increase technique and the global analysis approach of slope stability, a nonlinear optimization problem is defined for the stability analysis of pit slopes. In the optimization problem, the ordinates of discrete points on the slip surface and the factor of safety are both the decision variables, the objective function is the factor of security, and the constraints are the equilibrium equations and the convexity of the slip surface. Because the objective function is linear and the constraint functions are polynomials of degree three at most, the optimization problem is a classical optimization one that conventional optimization techniques can solve without recourse to modern optimization techniques such as the AI technique. Examples suggest that the proposed procedure is far more efficient and stable than the optimization models based on the methods of slides. At last, applications are demonstrated to ongoing deep foundation pit slopes situated at Tong Zhou – Beijing sub-center. [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF
8. Characteristic discriminative prototype network with detailed interpretation for classification.
- Author
-
Wen, Jiajun, Kong, Heng, Lai, Zhihui, and Zhu, Zhijie
- Subjects
- *
SIMILARITY (Physics) , *DEEP learning , *LEARNING strategies , *PROTOTYPES , *PRIOR learning - Abstract
Existing prototype learning methods provide limited interpretation on which patches from input images are similar to the corresponding prototypes. Moreover, these methods do not consider the diversities among the prototypes, which leads to low classification accuracy. To address these problems, this paper proposes Characteristic Prototype Network (CDPNet) with clear interpretation of local regions and characteristic. The network designs the feature prototype to represent the discriminative feature and the characteristic prototype to characterize the prototype's properties among different individuals. In addition, two novel strategies, dynamic region learning and similarity score minimization among similar intra-class prototypes, are designed to learn the prototypes so as to improve their diversity. Therefore, CDPNet can explain which kind of characteristic within the image is the most important one for classification tasks. The experimental results on well-known datasets show that CDPNet can provide clearer interpretations and obtain state-of-the-art classification performance in prototype learning. • We propose a novel interpretable prototype learning method for classification. • The prototype learning mechanism in the previous research is analyzed and the drawbacks are revealed. • This paper presents a new strategy to learn the prototype within the dynamic region by similarity score. [ABSTRACT FROM AUTHOR]
- Published
- 2025
- Full Text
- View/download PDF
9. Breast cancer discriminant feature analysis for diagnosis via jointly sparse learning.
- Author
-
Kong, Heng, Lai, Zhihui, Wang, Xu, and Liu, Feng
- Subjects
- *
BREAST cancer diagnosis , *COMPUTER algorithms , *ITERATIVE methods (Mathematics) , *PROBLEM solving , *COMPUTER systems - Abstract
Breast cancer is one of the most common malignant tumors among female. How to effectively discriminate the category of the cancers using the key features/factors is very important in the diagnosis and prediction. In this paper, Jointly Sparse Discriminant Analysis (JSDA) is proposed to explore the key factors in breast cancer and extract the key features for improving the accuracy in diagnosis and prediction. JSDA introduces the jointly sparse regular term (i.e. L 2 , 1 -norms term) to the criterion. A convergent iterative algorithm is designed to solve the optimization problem. It is shown that the proposed JSDA algorithm not only can learn the jointly sparse discriminant vectors to explore the key factors of the breast cancer in cancer pathologic diagnosis, but also can improve the diagnosis accuracy compared with the classical feature extraction and discriminant analysis algorithm. Experimental results on breast cancer datasets indicate that JSDA outperforms some well-known subspace learning algorithms in prediction accuracy, not matter they are non-sparse or sparse, particularly in the cases of small sample sizes. Data analysis shows that the key factors of the breast cancer explored by the JSDA are consistent with the practical experience. [ABSTRACT FROM AUTHOR]
- Published
- 2016
- Full Text
- View/download PDF
10. Cryoreservation of alginate–fibrin beads involving bone marrow derived mesenchymal stromal cells by vitrification
- Author
-
Bhakta, Gajadhar, Lee, Kong Heng, Magalhães, Raquel, Wen, Feng, Gouk, Sok Siam, Hutmacher, Dietmar W., and Kuleshova, Lilia L.
- Subjects
- *
CRYOPRESERVATION of cells , *BONE marrow , *BIOMEDICAL materials , *REGENERATIVE medicine , *SOLIDIFICATION , *BONE regeneration , *SCANNING electron microscopy , *BIOCOMPATIBILITY - Abstract
Abstract: Application of cell-–biomaterial systems in regenerative medicine can be facilitated by their successful low temperature preservation. Vitrification, which avoids ice crystal formation by amorphous solidification, is an emerging approach to cryopreservation. Developing vitrification strategy, effective cryopreservation of alginate–fibrin beads with porcine mesenchymal stromal cells has been achieved in this study. The cell–biomaterial constructs were pre-cultured for 20 days before cryopreservation, allowing for cell proliferation and construct stabilization. Ethylene glycol (EG) was employed as the basic cryoprotectant for two equilibration solutions. Successful cryopreservation of the constructs was achieved using vitrification solution composed of penetrating (EG MW 62Da) and non-penetrating (sucrose MW 342Da) cryoprotectants. Stepwise procedure of introduction to and removal of cryoprotectants was brief; direct plunging into liquid nitrogen was applied. Cell viability, evaluated by combining live/death staining and confocal laser microscopy, was similar for both control and vitrified cells in the beads. No detectable damage of microstructure of cryopreserved beads was found as shown by scanning electron microscopy. Both osteogenically induced control and vitrified cells in the constructs were equally capable of mineral production and deposition. There was no statistically significant difference in metabolic activity and proliferation between both groups during the entire culture period. Our study leads to the conclusion that the developed cryopreservation protocol allowed to maintain the integrity of the beads while preserving the ability of the pig bone marrow derived mesenchymal stromal cells to proliferate and subsequently differentiate; demonstrating that vitrification is a promising approach for cryopreser-vation of “ready-to-use” cell–biomaterial constructs. [Copyright &y& Elsevier]
- Published
- 2009
- Full Text
- View/download PDF
11. LPRR: Locality preserving robust regression based jointly sparse feature extraction.
- Author
-
Zhu, Yufei, Wen, Jiajun, Lai, Zhihui, Zhou, Jie, and Kong, Heng
- Subjects
- *
FEATURE extraction , *COMPUTATIONAL complexity , *REGRESSION analysis , *ALGORITHMS - Abstract
Jointly sparse projection learning attracts considerable attention due to its strong interpretability in feature extraction. To address the challenges related to weak discriminating representation in supervised feature extraction, we propose a more powerful regression framework. Based on the framework, we exhibit a new regression model called locality preserving robust regression (LPRR). In LPRR, we first combine the reconstruction error minimization and the projection variance maximization to explore the structured information of the data. Then, the label information is utilized and the low rank representation can be learned to explore the latent correlation structures among different classes. Furthermore, L 2 , 1 -norm is applied to measure the loss function and regularization terms, enhancing the robustness of the model and ensuring the joint sparsity of the projection matrix. An iterative algorithm is elaborately designed to achieve the optimal solutions of LPRR, in which the subproblem of LPRR can be regarded as a general quadratic problem on the Stiefel manifold. The convergence and the computational complexity of LPRR are analyzed rigorously. Finally, comprehensive experiments demonstrate the competitive performance of the proposed algorithm. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
12. A joint learning framework for optimal feature extraction and multi-class SVM.
- Author
-
Lai, Zhihui, Liang, Guangfei, Zhou, Jie, Kong, Heng, and Lu, Yuwu
- Subjects
- *
FEATURE extraction , *FISHER discriminant analysis , *PATTERN recognition systems , *SUPPORT vector machines , *CLASSIFICATION - Abstract
In high-dimensional data classification, effectively extracting discriminative features while eliminating redundancy is crucial for enhancing the performances of classifiers, such as Support Vector Machine (SVM). However, previous studies have decoupled the process of feature extraction from the development of SVM, leading to suboptimal classification accuracy. To address this problem, we propose a novel joint learning framework that combines optimal feature extraction and multi-class SVM, incorporating a generalized regression form to learn a discriminative latent subspace. The projected data in this subspace are more likely to have a larger margin between different classes and align with the properties of the SVM classification mechanism, enhancing the overall classification performance. Three iterative algorithms were presented to obtain optimal solutions with guaranteed convergence, and theoretical analyses were also conducted to reveal their fundamental nature. The optimal linear projection subspace is equivalent to that obtained from Linear Discriminant Analysis (LDA) in some special cases. We conducted extensive experiments using diverse datasets to evaluate the performances of the proposed algorithms. Our algorithms achieved an accuracy improvement of up to 7.55% compared to other conventional methods. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
13. Robust jointly sparse embedding for dimensionality reduction.
- Author
-
Lai, Zhihui, Chen, Yudong, Mo, Dongmei, Wen, Jiajun, and Kong, Heng
- Subjects
- *
ROBUST control , *EMBEDDINGS (Mathematics) , *DIMENSIONAL reduction algorithms , *LINEAR systems , *ORTHOGRAPHIC projection - Abstract
Abstract As a famous linear manifold learning method, orthogonal neighborhood preserving projections (ONPP) is able to provide a set of orthogonal projections for dimensionality reduction. However, a problem of ONPP is that it takes the L 2 -norm as the basic measurement and therefore tends to be sensitive to the outliers or the variations of the data. Aiming at strengthening the robustness of the conventional method ONPP, in this paper, a robust and sparse dimensionality reduction method based on linear reconstruction, called Robust Jointly Sparse Embedding (RJSE), is proposed by introducing L 2, 1 -norm as the basic measurement and regularization term. We design a simple iterative algorithm to obtain the optimal solution of the proposed robust and sparse dimensionality reduction model. Experiments on four benchmark data sets demonstrate the competitive performance of the proposed method compared with the state-of-the-art methods. [ABSTRACT FROM AUTHOR]
- Published
- 2018
- Full Text
- View/download PDF
14. Generalized multiview regression for feature extraction.
- Author
-
Lai, Zhihui, Lin, Yiling, Zheng, Jiacan, Zhou, Jie, and Kong, Heng
- Subjects
- *
FEATURE extraction , *COMPUTATIONAL complexity , *ENCYCLOPEDIAS & dictionaries , *STATISTICAL correlation - Abstract
Multiview learning (MVL) has attracted considerable attention since an object can be observed from various views. To use the consistency of multiple views, canonical correlation analysis (CCA) is used as a basic technique for analyzing correlated subspaces. However, CCA-based methods are designed to extract the correlation information between each pair of views. In addition, these methods ignore view-specific geometric structures, which may provide effective complementary information. To address the limitations of CCA-based methods and improve the performance of multiview subspace learning methods, a novel MVL framework called generalized multiview regression (GMR) is proposed. It aims at finding a common subspace to preserve the complementary information of each view and maintain consistency among all the views. Specifically, to preserve the view-specific structures, GMR first considers data reconstruction and local geometrical structures. Subsequently, by introducing an orthogonal dictionary, GMR captures the discriminative consistency suitable for classification tasks. Finally, it uses L 2 , 1 as the basic norm to measure errors and regularization, which facilitates robustness and sparsity for feature extraction and selection. An iterative algorithm is designed to solve the proposed GMR. In addition, the convergence and complexity are analyzed theoretically. Extensive experiments on benchmark datasets are conducted to compare the GMR with state-of-the-art or some available multiview methods. The competitive performance implies that GMR is an effective multiview method for learning sparse projections and extracting discriminative and robust features. • A multiview method, GMR, is proposed for feature extraction, which considers view consistency and complementary information simultaneously. • GMR explores the complementary properties using view-specific geometric structure and achieves the view consistency using label information. • The L2,1 norm is imposed on objective function and regularization term to eliminate redundant information so as to extract robust features. • An iterative algorithm is presented to obtain the local optimal solution, and the convergence and computational complexity are given. • Extensive experiments are conducted on several well-known datasets to illustrate the competitive performance of the GMR. [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF
15. Multiview Jointly Sparse Discriminant Common Subspace Learning.
- Author
-
Lin, Yiling, Lai, Zhihui, Zhou, Jie, Wen, Jiajun, and Kong, Heng
- Subjects
- *
FISHER discriminant analysis , *DISCRIMINANT analysis , *AUGMENTED reality , *COMPUTATIONAL complexity , *PROBLEM solving - Abstract
• A multiview method, GRMDA, is proposed for extracting the robust and sparse features, which reconstructs within-class scatter term and between-class scatter term in multiview scenario using L 2 , 1 norm to enhance robustness. • We conduct the analysis of small-class problem and adopt another criterion inspired by MMC to solve this problem. In terms of overfitting problem, the L 2 , 1 norm is also imposed on the regularization term for extracting sparse features. • The paper presents an iterative algorithm to compute optimal linear transform for each view. The poof of convergence is provided, and the computational complexity is analyzed. Several experiments were conducted to evaluate the performance of GRMDA Multiview data leads to the demand for classifying samples from various views, and the large gap between different views makes the classification task challenging. Recently, researchers have extended linear discriminant analysis (LDA) to multi-view scenarios. However, the extended methods are generally associated with the small-class problem, that is, the projection size is limited by the number of classes. In addition, they are sensitive to variations in images or outliers. To solve these problems, this study proposes a generalized robust multiview discriminant analysis (GRMDA) to obtain a linear transform for each view and for learning multiview jointly sparse discriminant common subspace. GRMDA aims to achieve both maximal between-class and minimal within-class variation for data from multiple views in a common space. Instead of formulating the ratio trace problem, we reformulate GRMDA inspired by maximum margin criterion (MMC) to address the small-class problem. Moreover, the proposed method achieves stronger robustness by reconstructing the within-class and between-class scatter terms from the definition of L 2 , 1 norm. Furthermore, GRMDA ensures joint sparsity using the L 2 , 1 norm-based regularization term. Additionally, we present an iterative algorithm, convergence proof, and complexity analysis. Experiments on six popular databases, that is, COIL100, USPS/MNIST, Extended Yale Face B, AR, BBCSport, and multiple feature datasets, were conducted to evaluate the performance of GRMDA against the state-of-the-art multiview methods. The experimental results demonstrate that the proposed method can achieve a significant performance with strong robustness and fast convergence. [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF
16. An analytical model for face stability of shield tunnel in dry cohesionless soils with different buried depth.
- Author
-
Zhang, Yu, Tao, Lianjin, Zhao, Xu, Kong, Heng, Guo, Fei, and Bian, Jin
- Subjects
- *
TUNNELS , *INTERNAL friction , *SOILS , *TERRITORIAL partition , *REFERENCE values - Abstract
[Display omitted] For deep-buried shield tunnels in dry cohesionless soils, it is critical to determine the support pressure acting on a tunnel face due to the significant soil arching effect during excavation. In this paper, a new analytical model of the multi-layer parabolic bearing arch is proposed to predict the limit support pressure based on the limit equilibrium method and the wedge model. The model consists of the multi-layer parabolic bearing arch, the friction arch and the wedge. According to the different buried depth of tunnel, the tunnel state is divided into shallow buried zone, transition zone and deep buried zone. By assuming the multi-layer parabolic soil arch as a three-hinged structural arch with reasonable arch axis, the load transfer expression of multi-layer parabolic soil arch is derived in transition zone and deep buried zone respectively. Then, the rationality and effectiveness of the proposed model were verified by comparing with model tests, existing theoretical methods and numerical model. Finally, the effects from the internal friction angle φ of soil on the inclination angle β , tunnel state and limit support pressure were also discussed. It is shown that with the increase of internal friction angle φ , the boundaries in between the shallow zone and the deep buried zone declines linearly, the inclination angle β increases linearly and the limit support pressure decreases nonlinearly. This study provides a significant reference for predicting the value of limit support pressure acting on the tunnel face in dry cohesionless soils. [ABSTRACT FROM AUTHOR]
- Published
- 2022
- Full Text
- View/download PDF
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.