Hongwei Gao

Dr. Hongwei Gao


Research Focus & Metrics

Academic Career Length

21Years (2006-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
3 March 2025 in IEEE Transactions on Emerging Topics in Computational Intelligence
Verified
Journal Article
Closed access
CXCaifeng Xia
Caifeng Xia
HGHongwei Gao
Hongwei Gao
WYWei Yang
Wei Yang
JYJIAHUI YU
JIAHUI YU
Multimodal image fusion is a vital technique that integrates images from various sensors to create a comprehensive and coherent representation, with broad applications in surveillance, medical imaging, and autonomous driving. However, current fusion methods struggle with inadequate feature representation, limited global context understanding due to the small receptive fields of convolutional neural networks (CNNs), and the loss of high-frequency information, all of which lead to suboptimal fusion quality. To address these challenges, we propose the Multi-Scale Diffusion Transformer (MSDT), a novel fusion framework that seamlessly combines a latent diffusion model with a transformer-based architecture. MSDT uses a perceptual compression network to encode source images into a low-dimensional latent space, reducing computational complexity while preserving essential features. It also incorporates a multiscale feature fusion mechanism, enhancing both detail and structural understanding. Additionally, MSDT features a self-attention module to extract unique high-frequency features and a cross-attention module to identify common low-frequency features across modalities, improving contextual understanding. Extensive experiments on three datasets show that MSDT significantly outperforms state-of-the-art methods across twelve evaluation metrics, achieving an SSIM score of 0.98. Moreover, MSDT demonstrates superior robustness and generalizability, highlighting the potential of integrating diffusion models with transformer architectures for multimodal image fusion.

20 October 2023
Verified
Conference Paper
Open access
BLBo Liu
Bo Liu
HGHongwei Gao
Hongwei Gao
YJYueqiu Jiang
Yueqiu Jiang
JWJiaxuan Wu
Jiaxuan Wu
Brain-computer interface (BCI) is a new technology, especially non-invasive BCI. A large number of scholars at home and abroad have carried out relevant research. This paper has read a large number of references, first of all, the definition of BCI and its system composition are summarized, and then the classification of BCI, the definition of each part, principles, advantages and disadvantages are introduced. This paper studies and analyzes the research status of non-invasive BCI at home and abroad, and summarizes the common non-invasive BCI systems. Bci can be widely used in medical treatment, entertainment, games, education and training, military and security fields, and plays a very important role in daily life and work.

25 May 2020 in Journal of Control Science and Engineering
Verified
Journal Article
Open access
PDF file
DWDongsheng Wu
Dongsheng Wu
ZJZhenping Ji
Zhenping Ji
JYJian Yang
Jian Yang
HGHongwei Gao
Hongwei Gao
JYJiahui Yu
Jiahui Yu
ZJZhaojie Ju
Zhaojie Ju
Secondary cooling and final electromagnetic stirring (F-EMS) are both key technologies for continuous casting. These parameters are usually optimized and controlled separately which caused internal quality fluctuations in unsteady conditions. In this paper, a coordinated optimal control strategy based on a multiobjective particle swarm optimization (MOPSO) algorithm is proposed for the parameter optimization of secondary cooling and F-EMS, which is solved based on multiobjective particle swarm optimization (MOPSO) algorithm. The solidification and heat transfer model are developed for the computation of billet temperature and the solidification, and the adaptive grid method is used to improve the diversity and robustness of optimal solutions. The secondary cooling water and F-EMS’ stirring current are dynamically controlled based on the optimization results. The results of field trials showed that the maximum carbon segregation and other quality indexes of billets can be improved significantly.

24 July 2024 in Scientific Reports
Verified
Journal Article
Open access
BLBo Liu
Bo Liu
HGHongwei Gao
Hongwei Gao
YJYueqiu Jiang
Yueqiu Jiang
JWJiaxuan Wu
Jiaxuan Wu
Brain-computer interfaces (BCIs) based on steady-state visual evoked potentials (SSVEP) have received widespread attention due to their high information transmission rate, high accuracy, and rich instruction set. However, the performance of its identification methods strongly depends on the amount of calibration data for within-subject classification. Some studies use deep learning (DL) algorithms for inter-subject classification, which can reduce the calculation process, but there is still much room for improvement in performance compared with intra-subject classification. To solve these problems, an efficient SSVEP signal recognition deep learning network model e-SSVEPNet based on the soft saturation nonlinear module is proposed in this paper. The soft saturation nonlinear module uses a similar exponential calculation method for output when it is less than zero, improving robustness to noise. Under the conditions of the SSVEP data set, two sliding time window lengths (1 s and 0.5 s), and three training data sizes, this paper evaluates the proposed network model and compares it with other traditional and deep learning model baseline methods. The experimental results of the nonlinear module were classified and compared. A large number of experimental results show that the proposed network has the highest average accuracy of intra-subject classification on the SSVEP data set, improves the performance of SSVEP signal classification and recognition, and has higher decoding accuracy under short signals, so it has huge potential ability to realize high-speed SSVEP-based for BCI.

1 January 2006
Verified
Conference Paper
Closed access
HGHongwei Gao
Hongwei Gao
GTGong-You Tang
Gong-You Tang
CLChao Li
Chao Li
We consider an optimal disturbance rejection problem for time-delay systems with persistent disturbances. In order to realize disturbance rejection with zero steady-state error, a disturbance compensator is constructed based on the internal model principle. At the same time, the optimal disturbance rejection problem with zero steady-state error is transformed into an optimal control problem for time-delay systems without disturbances. Then by introducing a sensitivity parameter, the original problem is reduced to solving a series of two-point boundary value (TPBV) problems without delay or advance terms. The obtained optimal control law consists of an analytic state feedback term and a compensation term, which is a series sum of the adjoint vectors. By truncating a finite terms of the adjoint vectors series, we obtain a suboptimal feedback control law with zero steady-state error. A simulation example shows the effectiveness of the presented algorithm

8 May 2024 in Research Square
Verified
Preprint
Open access
PDF file
BLBo Liu
Bo Liu
HGHongwei Gao
Hongwei Gao
YJYueqiu Jiang
Yueqiu Jiang
JWJiaxuan Wu
Jiaxuan Wu
Brain-computer interfaces (BCIs) based on steady-state visual evoked potentials (SSVEP) have received widespread attention due to their high information transmission rate, high accuracy, and rich instruction set. However, the performance of its identification methods strongly depends on the amount of calibration data for within-subject classification. Some studies use deep learning (DL) algorithms for inter-subject classification, which can reduce the calculation process, but there is still much room for improvement in performance compared with intra-subject classification. To solve these problems, an efficient SSVEP signal recognition deep learning network model e-SSVEPNet based on the soft saturation nonlinear module is proposed in this paper. The soft saturation nonlinear module uses a similar exponential calculation method for output when it is less than zero, improving robustness to noise. Under the conditions of the SSVEP data set, two sliding time window lengths (1s and 0.5s), and three training data sizes, this paper evaluates the proposed network model and compares it with other traditional and deep learning model baseline methods. The experimental results of the nonlinear module were classified and compared. A large number of experimental results show that the proposed network has the highest average accuracy of inter-subject classification on the SSVEP data set, improves the performance of SSVEP signal classification and recognition, and has higher decoding accuracy under short signals, so it has huge potential ability to realize high-speed SSVEP-based for BCI.