Hongfeng Li
Hongfeng Li has a total of 60 co-authors. Here are the top co-authors by number of shared publications:
- Hongbin Han
Peking University Health Science Center
3 shared publications - Hua Huang
Wuhan College
3 shared publications - Huizhen Li
Wuhan Textile University
3 shared publications - Bengong Zhang
Wuhan Textile University
2 shared publications - Baiying Lei
Shenzhen University
1 shared publication Caixia Wu
Department of Nuclear Medicine, Peking University First Hospital, Beijing, China
1 shared publication
Latest Publications
- 25 July 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 332
Quaternion-based multi-branch attention network for hyperspectral image classification
- 25 July 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 332
Quaternion-based multi-branch attention network for hyperspectral image classification
- 19 March 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 320
MemTTA: Cluster-guided continual test-time adaptation for cross-domain segmentation
- 19 March 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 320
MemTTA: Cluster-guided continual test-time adaptation for cross-domain segmentation
- 13 February 2024
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 171, 108133
Quantitative analysis of molecular transport in the extracellular space using physics-informed neural network
- 13 February 2024
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 171, 108133
Quantitative analysis of molecular transport in the extracellular space using physics-informed neural network
- 25 January 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Signal Processing
- Vol. 70 (1053587X) , 701-713
Hyperspectral Image Classification Using Adaptive Weighted Quaternion Zernike Moments
Hyperspectral image classification (HSI) has been widely used in many fields. However, image noise, atmospheric conditions, material distribution and other factors seriously degrade the classification accuracy of HSIs. To alleviate these issues, a new approach, namely adaptive weighted quaternion Zernike moments (AWQZM), is proposed, which extracts effective spatial-spectral features for pixels in HSI classification. The main contributions and novelties of the method are as follows: 1) the AWQZM can adaptively set weights for each pixel in the neighborhood, which not only can flexibly search for homogeneous regions of HSIs, but also can strengthen the similarity of pixels from the same class and the distinctiveness of pixels from different classes; 2) the AWQZM can be constructed in a small subset of bands through a grouping strategy, thereby reducing the computational complexity; and 3) the introduction of quaternions can preserve the spatial correlation among bands and reduce the loss of data information, and the use of quaternion phase information makes the extracted features more informative and discriminative. Moreover, the spectral features and spatial features are combined to achieve better HSI classification results. Experimental results on three benchmark data sets demonstrate that the proposed approach achieves better classification performance than other related approaches.
- 25 January 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Signal Processing
- Vol. 70 (1053587X) , 701-713
Hyperspectral Image Classification Using Adaptive Weighted Quaternion Zernike Moments
Hyperspectral image classification (HSI) has been widely used in many fields. However, image noise, atmospheric conditions, material distribution and other factors seriously degrade the classification accuracy of HSIs. To alleviate these issues, a new approach, namely adaptive weighted quaternion Zernike moments (AWQZM), is proposed, which extracts effective spatial-spectral features for pixels in HSI classification. The main contributions and novelties of the method are as follows: 1) the AWQZM can adaptively set weights for each pixel in the neighborhood, which not only can flexibly search for homogeneous regions of HSIs, but also can strengthen the similarity of pixels from the same class and the distinctiveness of pixels from different classes; 2) the AWQZM can be constructed in a small subset of bands through a grouping strategy, thereby reducing the computational complexity; and 3) the introduction of quaternions can preserve the spatial correlation among bands and reduce the loss of data information, and the use of quaternion phase information makes the extracted features more informative and discriminative. Moreover, the spectral features and spatial features are combined to achieve better HSI classification results. Experimental results on three benchmark data sets demonstrate that the proposed approach achieves better classification performance than other related approaches.
- 3 July 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 30 (2) , 539-552
Neural-Response-Based Extreme Learning Machine for Image Classification
This paper proposes a novel and simple multilayer feature learning method for image classification by employing the extreme learning machine (ELM). The proposed algorithm is composed of two stages: the multilayer ELM (ML-ELM) feature mapping stage and the ELM learning stage. The ML-ELM feature mapping stage is recursively built by alternating between feature map construction and maximum pooling operation. In particular, the input weights for constructing feature maps are randomly generated and hence need not be trained or tuned, which makes the algorithm highly efficient. Moreover, the maximum pooling operation enables the algorithm to be invariant to certain transformations. During the ELM learning stage, elastic-net regularization is proposed to learn the output weight. Elastic-net regularization helps to learn more compact and meaningful output weight. In addition, we preprocess the input data with the dense scale-invariant feature transform operation to improve both the robustness and invariance of the algorithm. To evaluate the effectiveness of the proposed method, several experiments are conducted on three challenging databases. Compared with the conventional deep learning methods and other related ones, the proposed method achieves the best classification results with high computational efficiency.
- 3 July 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 30 (2) , 539-552
Neural-Response-Based Extreme Learning Machine for Image Classification
This paper proposes a novel and simple multilayer feature learning method for image classification by employing the extreme learning machine (ELM). The proposed algorithm is composed of two stages: the multilayer ELM (ML-ELM) feature mapping stage and the ELM learning stage. The ML-ELM feature mapping stage is recursively built by alternating between feature map construction and maximum pooling operation. In particular, the input weights for constructing feature maps are randomly generated and hence need not be trained or tuned, which makes the algorithm highly efficient. Moreover, the maximum pooling operation enables the algorithm to be invariant to certain transformations. During the ELM learning stage, elastic-net regularization is proposed to learn the output weight. Elastic-net regularization helps to learn more compact and meaningful output weight. In addition, we preprocess the input data with the dense scale-invariant feature transform operation to improve both the robustness and invariance of the algorithm. To evaluate the effectiveness of the proposed method, several experiments are conducted on three challenging databases. Compared with the conventional deep learning methods and other related ones, the proposed method achieves the best classification results with high computational efficiency.
- 1 November 2014
- Published by Elsevier in Neurocomputing
- Vol. 144, 198-207
Sparse-based neural response for image classification
- 1 November 2014
- Published by Elsevier in Neurocomputing
- Vol. 144, 198-207
Sparse-based neural response for image classification
- 23 November 2011
- Published by SPIE-Intl Soc Optical Eng
- Vol. 8200, 820018-820018-8
A color image segmentation algorithm by integrating watershed with automatic seeded region growing and merging
A color image segmentation algorithm which integrates watershed with automatic seeded region growing and merging is proposed in the paper. Firstly, the image was transformed from RGB color space to HSI space. Next, watershed algorithm was applied to the image to obtain initial segmentation effect. And then, Based on the results of watershed segmentation, some regions in the image were selected as seeded regions automatically for seeds growth algorithm by making use of color differences and relative Euclidean distance. Finally, a region combining algorithm was executed to avoid excessive segmentation. The proposed method combines the advantages of watershed and region growing approach, and in accord with the human vision segmentation strategy. This algorithm was applied to segment some images, the experimental results confirm its effectiveness and efficiency.
- 23 November 2011
- Published by SPIE-Intl Soc Optical Eng
- Vol. 8200, 820018-820018-8
A color image segmentation algorithm by integrating watershed with automatic seeded region growing and merging
A color image segmentation algorithm which integrates watershed with automatic seeded region growing and merging is proposed in the paper. Firstly, the image was transformed from RGB color space to HSI space. Next, watershed algorithm was applied to the image to obtain initial segmentation effect. And then, Based on the results of watershed segmentation, some regions in the image were selected as seeded regions automatically for seeds growth algorithm by making use of color differences and relative Euclidean distance. Finally, a region combining algorithm was executed to avoid excessive segmentation. The proposed method combines the advantages of watershed and region growing approach, and in accord with the human vision segmentation strategy. This algorithm was applied to segment some images, the experimental results confirm its effectiveness and efficiency.