Guanghui Zeng
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Guanghui Zeng has a total of 4 co-authors. Here are the top co-authors by number of shared publications:
- Chaoqun Hong
College of Computer and Information Engineering, Xiamen University of Technology, Xiamen, China
2 shared publications - Lijuan Liu
Xiamen University of Technology
1 shared publication - Weixin Su
Xiamen University of Technology
1 shared publication - Yiqing Fan
Key Laboratory of Fujian Universities for Virtual Reality and 3D Visualization, Xiamen University of Technology, Xiamen, China
1 shared publication
Latest Publications
- Open Access
- 29 July 2024
- Published by Springer Nature in Research Square
Ensemble Learning with Feature Optimization for Credit Risk Assessment
Credit risk assessment stands as a cornerstone in financial decision-making, with significant implications for economic stability and growth. This paper highlights the transformative advantages of credit big data over traditional methods, particularly in enhancing the creditworthiness evaluation of small and medium-sized enterprises (SMEs). We delineate the distinctive features of the big data financial innovation model across six economic dimensions, showcasing its potential to reshape financial practices. To address the inefficiencies of traditional expert-driven approaches, we introduce an innovative 'Feature Selector-classifier Optimization Framework' that streamlines the credit risk prediction process. This framework not only refines the accuracy and efficiency of predictions but also integrates seamlessly with economic analysis, offering a robust tool for financial decision-makers. Our ensemble classifier delivers remarkable performance, exemplified by its high accuracy and AUC scores across multiple datasets, thereby validating the framework's efficacy in enhancing predictive power while ensuring operational efficiency.
- 29 July 2024
- Published by Springer Nature in Research Square
Open AccessEnsemble Learning with Feature Optimization for Credit Risk Assessment
Credit risk assessment stands as a cornerstone in financial decision-making, with significant implications for economic stability and growth. This paper highlights the transformative advantages of credit big data over traditional methods, particularly in enhancing the creditworthiness evaluation of small and medium-sized enterprises (SMEs). We delineate the distinctive features of the big data financial innovation model across six economic dimensions, showcasing its potential to reshape financial practices. To address the inefficiencies of traditional expert-driven approaches, we introduce an innovative 'Feature Selector-classifier Optimization Framework' that streamlines the credit risk prediction process. This framework not only refines the accuracy and efficiency of predictions but also integrates seamlessly with economic analysis, offering a robust tool for financial decision-makers. Our ensemble classifier delivers remarkable performance, exemplified by its high accuracy and AUC scores across multiple datasets, thereby validating the framework's efficacy in enhancing predictive power while ensuring operational efficiency.
- Open Access
- 11 February 2024
- Published by Springer Nature in Neural Processing Letters
- Vol. 56 (1) , 1-20
A Deep Convolutional Encoder–Decoder–Restorer Architecture for Image Deblurring
The accuracy of many computer vision tasks is reduced by blurred images, so deblur is important. More details of the image can be captured by a common multi-stage network, but the computational complexity of this method is higher compared with a single-stage network. However, a single-stage network cannot capture multi-scale information well. To tackle the problem, a novel convolutional encoder–decoder–restorer architecture is proposed. In this architecture, a multi-scale input structure is used in the encoder. Improved supervised attention module is inserted into the encoder for enhanced feature acquisition. In decoder, information supplement block is proposed to fuse multi-scale features. Finally, the fused features are used for image recovery in the restorer. In order to optimise the model in multiple domains, the loss function is calculated separately in the spatial and frequency domains. Our method is compared with existing methods on the GOPRO dataset. In addition, to verify the applications of our proposed method, we conduct experiments on the Real image dataset, the VOC2007 dataset and the LFW dataset. Experimental results show that our proposed method outperforms state-of-the-art deblurring methods and improves the accuracy of different vision tasks.
- 11 February 2024
- Published by Springer Nature in Neural Processing Letters
- Vol. 56 (1) , 1-20
Open AccessA Deep Convolutional Encoder–Decoder–Restorer Architecture for Image Deblurring
The accuracy of many computer vision tasks is reduced by blurred images, so deblur is important. More details of the image can be captured by a common multi-stage network, but the computational complexity of this method is higher compared with a single-stage network. However, a single-stage network cannot capture multi-scale information well. To tackle the problem, a novel convolutional encoder–decoder–restorer architecture is proposed. In this architecture, a multi-scale input structure is used in the encoder. Improved supervised attention module is inserted into the encoder for enhanced feature acquisition. In decoder, information supplement block is proposed to fuse multi-scale features. Finally, the fused features are used for image recovery in the restorer. In order to optimise the model in multiple domains, the loss function is calculated separately in the spatial and frequency domains. Our method is compared with existing methods on the GOPRO dataset. In addition, to verify the applications of our proposed method, we conduct experiments on the Real image dataset, the VOC2007 dataset and the LFW dataset. Experimental results show that our proposed method outperforms state-of-the-art deblurring methods and improves the accuracy of different vision tasks.