Feature selection is always an important issue in the visual SLAM （simultaneous location and mapping） literature. Considering that the location estimation can be improved by tracking features with larger value of vi...Feature selection is always an important issue in the visual SLAM （simultaneous location and mapping） literature. Considering that the location estimation can be improved by tracking features with larger value of visible time, a new feature selection method based on motion estimation is proposed. First, a k-step iteration algorithm is presented for visible time estimation using an affme motion model; then a delayed feature detection method is introduced for efficiently detecting features with the maximum visible time. As a means of validation for the proposed method, both simulation and real data experiments are carded out. Results show that the proposed method can improve both the estimation performance and the computational performance compared with the existing random feature selection method.展开更多
Based on the compressive sensing,a novel algorithm is proposed to solve reconstruction problem under sparsity assumptions.Instead of estimating the reconstructed data through minimizing the objective function,the auth...Based on the compressive sensing,a novel algorithm is proposed to solve reconstruction problem under sparsity assumptions.Instead of estimating the reconstructed data through minimizing the objective function,the authors parameterize the problem as a linear combination of few elementary thresholding functions,which can be solved by calculating the linear weighting coefficients.It is to update the thresholding functions during the process of iteration.The advantage of this method is that the optimization problem only needs to be solved by calculating linear coefficients for each time.With the elementary thresholding functions satisfying certain constraints,a global convergence of the iterative algorithm is guaranteed.The synthetic and the field data results prove the effectiveness of the proposed algorithm.展开更多
KSSOLV(Kohn-Sham Solver)is a MATLAB(Matrix Laboratory)toolbox for solving the Kohn-Sham density functional theory(KS-DFT)with the plane-wave basis set.In the KS-DFT calculations,the most expensive part is commonly the...KSSOLV(Kohn-Sham Solver)is a MATLAB(Matrix Laboratory)toolbox for solving the Kohn-Sham density functional theory(KS-DFT)with the plane-wave basis set.In the KS-DFT calculations,the most expensive part is commonly the diagonalization of Kohn-Sham Hamiltonian in the self-consistent field(SCF)scheme.To enable a personal computer to perform medium-sized KS-DFT calculations that contain hundreds of atoms,we present a hybrid CPU-GPU implementation to accelerate the iterative diagonalization algorithms implemented in KSSOLV by using the MATLAB built-in Parallel Computing Toolbox.We compare the performance of KSSOLV-GPU on three types of GPU,including RTX3090,V100,and A100,with conventional CPU implementation of KSSOLV respectively and numerical results demonstrate that hybrid CPU-GPU implementation can achieve a speedup of about 10 times compared with sequential CPU calculations for bulk silicon systems containing up to 128 atoms.展开更多
A family of two-order Hermite vector-interpolating subdivision schemes is proposed and its convergence and con- tinuity are analyzed. The iterative level can be estimated for given error. The sufficient conditions of ...A family of two-order Hermite vector-interpolating subdivision schemes is proposed and its convergence and con- tinuity are analyzed. The iterative level can be estimated for given error. The sufficient conditions of C2 continuity are proved. Geometric features of subdivision curves, such as line segments, cusps and inflection points, are obtained by appending some conditions to initial vectorial Hermite sequence. An algorithm is presented for generating geometric features. For an initial se- quence of two-order Hermite elements from unit circle, the numerical error of the 4th subdivided level is O(10?4).展开更多
The auxiliary principle technique is extended to study a class of generalized set-valued strongly nonlinear mixed variational-like type inequalities. Firstly, the existence of solutions to the auxiliary problems for t...The auxiliary principle technique is extended to study a class of generalized set-valued strongly nonlinear mixed variational-like type inequalities. Firstly, the existence of solutions to the auxiliary problems for this class of generalized set-valued strongly nonlinear mixed variational-like type inequalities is shown. Secondly, the iterative algorithm for solving this class of generalized set-valued strongly nonlinear mixed variational-like type inequalities is given by using this existence result. Finally, the strong convergence of iterative sequences generated by the algorithm is proven. The present results improve, generalize and modify the earlier and recent ones obtained previously by some authors in the literature.展开更多
Reasonable selection and optimization of a filter used in model estimation for a multiple model structure is the key to improve tracking accuracy of maneuvering target.Combining with the cubature Kalman filter with it...Reasonable selection and optimization of a filter used in model estimation for a multiple model structure is the key to improve tracking accuracy of maneuvering target.Combining with the cubature Kalman filter with iterated observation update and the interacting multiple model method,a novel interacting multiple model algorithm based on the cubature Kalman filter with observation iterated update is proposed.Firstly,aiming to the structural features of cubature Kalman filter,the cubature Kalman filter with observation iterated update is constructed by the mechanism of iterated observation update.Secondly,the improved cubature Kalman filter is used as the model filter of interacting multiple model,and the stability and reliability of model identification and state estimation are effectively promoted by the optimization of model filtering step.In the simulations,compared with classic improved interacting multiple model algorithms,the theoretical analysis and experimental results show the feasibility and validity of the proposed algorithm.展开更多
Based on the characteristics of impulse noises, the authors establish a new filter, Iterative Adaptive Median Filter （IAMF）. Acccording to the characteristics of images polluted by impulse noises, they establish wei...Based on the characteristics of impulse noises, the authors establish a new filter, Iterative Adaptive Median Filter （IAMF）. Acccording to the characteristics of images polluted by impulse noises, they establish weight function combined with iterative algorithm to eliminate noises. In IAMF filter process, because the noise sixes do not participate in the computation, they do not influence the normal points in the image, therefore IAMF can retain the detail well, maintain the good clarity after processing image, and simultaneously reduce the computation. Experiments showed that IAMF have ideal denoising effect for the images polluted by the impulse noises; especially when the noise rates are more than 0.5, IAMF is mote prominent, even when the noise rotes are more than 0.9, IAMF can achieve a satisfactory results.展开更多
文摘Feature selection is always an important issue in the visual SLAM （simultaneous location and mapping） literature. Considering that the location estimation can be improved by tracking features with larger value of visible time, a new feature selection method based on motion estimation is proposed. First, a k-step iteration algorithm is presented for visible time estimation using an affme motion model; then a delayed feature detection method is introduced for efficiently detecting features with the maximum visible time. As a means of validation for the proposed method, both simulation and real data experiments are carded out. Results show that the proposed method can improve both the estimation performance and the computational performance compared with the existing random feature selection method.
文摘Based on the compressive sensing,a novel algorithm is proposed to solve reconstruction problem under sparsity assumptions.Instead of estimating the reconstructed data through minimizing the objective function,the authors parameterize the problem as a linear combination of few elementary thresholding functions,which can be solved by calculating the linear weighting coefficients.It is to update the thresholding functions during the process of iteration.The advantage of this method is that the optimization problem only needs to be solved by calculating linear coefficients for each time.With the elementary thresholding functions satisfying certain constraints,a global convergence of the iterative algorithm is guaranteed.The synthetic and the field data results prove the effectiveness of the proposed algorithm.
基金supported by the National Natural Science Foundation of China (No.21688102,No.21803066,and No.22003061)the Chinese Academy of Sciences Pioneer Hundred Talents Program (KJ2340000031,KJ2340007002)+7 种基金the National Key Research and Development Program of China(2016YFA0200604)the Anhui Initiative in Quantum Information Technologies (AHY090400)the Strategic Priority Research of Chinese Academy of Sciences(XDC01040100)CAS Project for Young Scientists in Basic Research (YSBR-005)the Fundamental Research Funds for the Central Universities (WK2340000091,WK2060000018)the Hefei National Laboratory for Physical Sciences at the Microscale (SK2340002001)the Research Start-Up Grants (KY2340000094)the Academic Leading Talents Training Program(KY2340000103) from University of Science and Technology of China
文摘KSSOLV(Kohn-Sham Solver)is a MATLAB(Matrix Laboratory)toolbox for solving the Kohn-Sham density functional theory(KS-DFT)with the plane-wave basis set.In the KS-DFT calculations,the most expensive part is commonly the diagonalization of Kohn-Sham Hamiltonian in the self-consistent field(SCF)scheme.To enable a personal computer to perform medium-sized KS-DFT calculations that contain hundreds of atoms,we present a hybrid CPU-GPU implementation to accelerate the iterative diagonalization algorithms implemented in KSSOLV by using the MATLAB built-in Parallel Computing Toolbox.We compare the performance of KSSOLV-GPU on three types of GPU,including RTX3090,V100,and A100,with conventional CPU implementation of KSSOLV respectively and numerical results demonstrate that hybrid CPU-GPU implementation can achieve a speedup of about 10 times compared with sequential CPU calculations for bulk silicon systems containing up to 128 atoms.
文摘A family of two-order Hermite vector-interpolating subdivision schemes is proposed and its convergence and con- tinuity are analyzed. The iterative level can be estimated for given error. The sufficient conditions of C2 continuity are proved. Geometric features of subdivision curves, such as line segments, cusps and inflection points, are obtained by appending some conditions to initial vectorial Hermite sequence. An algorithm is presented for generating geometric features. For an initial se- quence of two-order Hermite elements from unit circle, the numerical error of the 4th subdivided level is O(10?4).
基金the Teaching and Research Award Fund for Outstanding Young Teachers in Higher Education Institutions of MOE,Chinathe Dawn Program Foundationin Shanghai
文摘The auxiliary principle technique is extended to study a class of generalized set-valued strongly nonlinear mixed variational-like type inequalities. Firstly, the existence of solutions to the auxiliary problems for this class of generalized set-valued strongly nonlinear mixed variational-like type inequalities is shown. Secondly, the iterative algorithm for solving this class of generalized set-valued strongly nonlinear mixed variational-like type inequalities is given by using this existence result. Finally, the strong convergence of iterative sequences generated by the algorithm is proven. The present results improve, generalize and modify the earlier and recent ones obtained previously by some authors in the literature.
基金Supported by the National Nature Science Foundations of China(No.61300214,U1204611,61170243)the Science and Technology Innovation Team Support Plan of Education Department of Henan Province(No.13IRTSTHN021)+3 种基金the Science and Technology Research Key Project of Education Department of Henan Province(No.13A413066)the Basic and Frontier Technology Research Plan of Henan Province(No.132300410148)the Funding Scheme of Young Key Teacher of Henan Province Universitiesthe Key Project of Teaching Reform Research of Henan University(No.HDXJJG2013-07)
文摘Reasonable selection and optimization of a filter used in model estimation for a multiple model structure is the key to improve tracking accuracy of maneuvering target.Combining with the cubature Kalman filter with iterated observation update and the interacting multiple model method,a novel interacting multiple model algorithm based on the cubature Kalman filter with observation iterated update is proposed.Firstly,aiming to the structural features of cubature Kalman filter,the cubature Kalman filter with observation iterated update is constructed by the mechanism of iterated observation update.Secondly,the improved cubature Kalman filter is used as the model filter of interacting multiple model,and the stability and reliability of model identification and state estimation are effectively promoted by the optimization of model filtering step.In the simulations,compared with classic improved interacting multiple model algorithms,the theoretical analysis and experimental results show the feasibility and validity of the proposed algorithm.
基金supported by Shandong Prvince Natural Science Foundation(Y2008G31)
文摘Based on the characteristics of impulse noises, the authors establish a new filter, Iterative Adaptive Median Filter （IAMF）. Acccording to the characteristics of images polluted by impulse noises, they establish weight function combined with iterative algorithm to eliminate noises. In IAMF filter process, because the noise sixes do not participate in the computation, they do not influence the normal points in the image, therefore IAMF can retain the detail well, maintain the good clarity after processing image, and simultaneously reduce the computation. Experiments showed that IAMF have ideal denoising effect for the images polluted by the impulse noises; especially when the noise rates are more than 0.5, IAMF is mote prominent, even when the noise rotes are more than 0.9, IAMF can achieve a satisfactory results.
