Xiangguang Xiong
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Xiangguang Xiong has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
Guangyun Yang
5 shared publicationsXinhui Lu
5 shared publicationsYu Lu
4 shared publications- Siyao Zhong
School of Big Data and Computer Science, Guizhou Normal University, Guiyang, China
2 shared publications Yanxin Li
2 shared publications- Yu Lu
School of Big Data and Computer Science, Guizhou Normal University, Guiyang, China
2 shared publications
Latest Publications
- 2 July 2026
- Published by Elsevier in Neurocomputing
- Vol. 699
Robust zero-watermarking method for medical images via DINOv3-ViT and multi-scale neighborhood Zernike moments
- 2 July 2026
- Published by Elsevier in Neurocomputing
- Vol. 699
Robust zero-watermarking method for medical images via DINOv3-ViT and multi-scale neighborhood Zernike moments
- 22 May 2026
- Published by Elsevier in Digital Signal Processing
- Vol. 181
Robust zero-watermarking algorithm for multiple medical images based on guided filtering fusion and FFST-Daisy
- 22 May 2026
- Published by Elsevier in Digital Signal Processing
- Vol. 181
Robust zero-watermarking algorithm for multiple medical images based on guided filtering fusion and FFST-Daisy
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28, 14065-14078
ResiMark: A Spatial-Temporal Enhanced Video Watermarking Network Balancing Robustness and Imperceptibility
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28, 14065-14078
ResiMark: A Spatial-Temporal Enhanced Video Watermarking Network Balancing Robustness and Imperceptibility
- Open Access
- 12 December 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (24) , 13097
Efficient Pulsar Candidate Recognition Algorithm Under a Multi-Scale DenseNet Framework
The exponential growth of candidate data from large-scale radio pulsar surveys has created a pressing need for efficient and accurate classification methods. This paper presents a novel hybrid pulsar candidate recognition algorithm that integrates diagnostic plot images and structured numerical features using a multi-scale DenseNet framework. The proposed model combines convolutional neural networks (CNNs) for extracting spatial patterns from pulsar diagnostic plots and feedforward neural networks (FNNs) for processing scalar features such as SNR, DM, and pulse width. By fusing these multimodal representations, the model achieves superior classification performance, particularly in class-imbalanced settings standard to pulsar survey data. Evaluated on a synthesized dataset constructed from FAST and HTRU survey characteristics, the model demonstrates robust performance, achieving an F1-score of 0.904 and AUC-ROC of 0.978. Extensive ablation and cross-validation analyses confirm the contribution of each data modality and the model’s generalizability. Furthermore, the system maintains low inference latency (4.2 ms per candidate) and a compact architecture (~2.3 million parameters), indicating potential for real-time deployment once validated on real observational datasets. The proposed approach offers a scalable and interpretable multimodal framework for automated pulsar classification and provides a foundation for future validation and potential integration into observatories such as FAST and the Square Kilometre Array (SKA).
- 12 December 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (24) , 13097
Open AccessEfficient Pulsar Candidate Recognition Algorithm Under a Multi-Scale DenseNet Framework
The exponential growth of candidate data from large-scale radio pulsar surveys has created a pressing need for efficient and accurate classification methods. This paper presents a novel hybrid pulsar candidate recognition algorithm that integrates diagnostic plot images and structured numerical features using a multi-scale DenseNet framework. The proposed model combines convolutional neural networks (CNNs) for extracting spatial patterns from pulsar diagnostic plots and feedforward neural networks (FNNs) for processing scalar features such as SNR, DM, and pulse width. By fusing these multimodal representations, the model achieves superior classification performance, particularly in class-imbalanced settings standard to pulsar survey data. Evaluated on a synthesized dataset constructed from FAST and HTRU survey characteristics, the model demonstrates robust performance, achieving an F1-score of 0.904 and AUC-ROC of 0.978. Extensive ablation and cross-validation analyses confirm the contribution of each data modality and the model’s generalizability. Furthermore, the system maintains low inference latency (4.2 ms per candidate) and a compact architecture (~2.3 million parameters), indicating potential for real-time deployment once validated on real observational datasets. The proposed approach offers a scalable and interpretable multimodal framework for automated pulsar classification and provides a foundation for future validation and potential integration into observatories such as FAST and the Square Kilometre Array (SKA).
- 8 December 2025
- Published by Elsevier in Computer Standards & Interfaces
- Vol. 97
Robust zero-watermarking method for multi-medical images based on Chebyshev–Fourier moments and Contourlet-FFT
- 8 December 2025
- Published by Elsevier in Computer Standards & Interfaces
- Vol. 97
Robust zero-watermarking method for multi-medical images based on Chebyshev–Fourier moments and Contourlet-FFT
- 18 August 2025
- Published by Elsevier in Journal of Information Security and Applications
- Vol. 93
Robust zero-watermarking method for medical images based on FFST and Daisy descriptor
- 18 August 2025
- Published by Elsevier in Journal of Information Security and Applications
- Vol. 93
Robust zero-watermarking method for medical images based on FFST and Daisy descriptor
- 11 March 2025
- Published by Elsevier in Journal of Information Security and Applications
- Vol. 90
Robust zero-watermarking method for multiple medical images using wavelet fusion and DTCWT-QR
- 11 March 2025
- Published by Elsevier in Journal of Information Security and Applications
- Vol. 90
Robust zero-watermarking method for multiple medical images using wavelet fusion and DTCWT-QR
- 1 March 2025
- Published by Springer Nature in Circuits, Systems, and Signal Processing
- Vol. 44 (7) , 4796-4827
Double Robust Zero-Watermarking Algorithms Based on Fractional-Order Jacobi-Fourier Moments and FFST-Schur
- 1 March 2025
- Published by Springer Nature in Circuits, Systems, and Signal Processing
- Vol. 44 (7) , 4796-4827
Double Robust Zero-Watermarking Algorithms Based on Fractional-Order Jacobi-Fourier Moments and FFST-Schur
- 13 June 2024
- Published by Elsevier in Biomedical Signal Processing and Control
- Vol. 96
Robust zero-watermarking algorithm for multi-medical images based on FFST-Schur and Tent mapping
- 13 June 2024
- Published by Elsevier in Biomedical Signal Processing and Control
- Vol. 96
Robust zero-watermarking algorithm for multi-medical images based on FFST-Schur and Tent mapping
- 22 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11 (11) , 20088-20098
An Adaptive Bit Allocation Strategy for Minimizing Embedding Distortion in Interpolated Images Used for Reversible Data Hiding
The social and commercial value of digital data in the Internet of Medical Things (IoMT) is currently receiving unprecedented attention, and the security of the data is increasingly being challenged. In recent years, reversible data hiding (RDH) based on interpolation technology (IT) has emerged as an area of activity due to its high bit per pixel (BPP) and peak signal-to-noise ratio (PSNR). However, the PSNR of conventional embedding strategies may not be optimal for a given BPP because they sequentially embed data concerning the difference between interpolated and selected pixels, without considering that embedding fewer bits of data could produce higher PSNR. As such, we propose an adaptive strategy for IT-based RDH that provides the ability to embed into interpolated pixels with fewer bits of data preferentially. First, an adaptive bit allocation strategy (ABAS) with lower computational complexity is developed for a given BPP. Second, we propose an adaptive RDH method using the proposed ABAS and data recoding. The results on BossBase and BOWS-2 data sets showed that the proposed method surpassed several IT-based RDH methods, with average PSNR increasing by ~6.4 dB. Our method also achieved superior performance compared with non-IT-based RDH methods with an average PSNR increase of ~2.6 dB. Moreover, the proposed ABAS was applied to four typical IT-based RDH methods. The results showed that the improved methods increased the average PSNR by ~7.4 dB compared to the original methods. Our method is an expected solution to the security issues of sensitive data transmission in IoMT.
- 22 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11 (11) , 20088-20098
An Adaptive Bit Allocation Strategy for Minimizing Embedding Distortion in Interpolated Images Used for Reversible Data Hiding
The social and commercial value of digital data in the Internet of Medical Things (IoMT) is currently receiving unprecedented attention, and the security of the data is increasingly being challenged. In recent years, reversible data hiding (RDH) based on interpolation technology (IT) has emerged as an area of activity due to its high bit per pixel (BPP) and peak signal-to-noise ratio (PSNR). However, the PSNR of conventional embedding strategies may not be optimal for a given BPP because they sequentially embed data concerning the difference between interpolated and selected pixels, without considering that embedding fewer bits of data could produce higher PSNR. As such, we propose an adaptive strategy for IT-based RDH that provides the ability to embed into interpolated pixels with fewer bits of data preferentially. First, an adaptive bit allocation strategy (ABAS) with lower computational complexity is developed for a given BPP. Second, we propose an adaptive RDH method using the proposed ABAS and data recoding. The results on BossBase and BOWS-2 data sets showed that the proposed method surpassed several IT-based RDH methods, with average PSNR increasing by ~6.4 dB. Our method also achieved superior performance compared with non-IT-based RDH methods with an average PSNR increase of ~2.6 dB. Moreover, the proposed ABAS was applied to four typical IT-based RDH methods. The results showed that the improved methods increased the average PSNR by ~7.4 dB compared to the original methods. Our method is an expected solution to the security issues of sensitive data transmission in IoMT.