Dong Liu

Dong Liu

Auto-generated profile, based on publicly available information

Dong Liu

Department of Mathematics, Xian University of Technology, Xian, China

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

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

Representative Publications

View All Publications
14 March 2024 in IEEE Transactions on Multimedia
Verified
Journal Article
Closed access
DLDong Liu
Dong Liu
XWXiaofeng Wang
Xiaofeng Wang
RHRuidong Han
Ruidong Han
NBNingning Bai
Ningning Bai
JHJianpeng Hou
Jianpeng Hou
SPShanmin Pang
Shanmin Pang
The object of image super-resolution reconstruction is to overcome the limitations imposed by hardware imaging conditions and patterns, aiming to restore high-frequency details in images through signal processing techniques. Recently, deep learning-based single-image super-resolution reconstruction (SISR) has achieved remarkable performance. However, the current methods exhibit inadequate performance in the reconstruction of texture details, thereby posing a challenge for further enhancing the accuracy of super-resolution reconstruction. In this study, we propose a novel contextual texture enhancement network (CTE-Net) aimed at improving the level of texture details in image super-resolution. The CTE-Net comprises of two crucial components: the multi-level feature aggregation module (MFAM) and the contextual information enhancement module (CIEM). The MFAM integrates global and local low-resolution (LR) features from both the pixel space and channel dimensions, thereby enhancing the feature representation capability of the network. The CIEM is deployed to enhance the network's learning capacity by integrating a meticulously designed context-attention mechanism, which effectively explores the adjacent contextual information of images and thereby amplifies the expressive capability of the generated features. Moreover, we utilize local binary patterns (LBP) to guide the feature selection strategies for MFAM and CIEM, thereby prioritizing the network's decision logic towards the recovery of texture details. The extensive experiments demonstrate that our method yields satisfactory results. In comparison to the state-of-the-art approaches, our method exhibits superior performance on the benchmark datasets.

3 June 2023 in Entropy
Verified
Journal Article
Open access
PDF file
JWJunfeng Wu
Junfeng Wu
JZJialu Zhang
Jialu Zhang
DLDong Liu
Dong Liu
XWXiaofeng Wang
Xiaofeng Wang
Ensuring the privacy and secrecy of digital medical images has become a pressing issue as a result of the quick development of smart medical technology and the exponential growth in the quantity of medical images transmitted and stored in networks. The lightweight multiple-image encryption approach for medical images that is suggested in this research can encrypt/decrypt any number of medical photos of varied sizes with just one encryption operation and has a computational cost that is similar to encrypting a single image. The plaintext images with different sizes are filled at the right and bottom of the image to ensure that the size of all plaintext images is uniform; then, all the filled images are stacked to obtain a superimposed image. The initial key, which is generated using the SHA-256 technique, is then used as the starting value of the linear congruence algorithm to create the encryption key sequence. The cipher picture is then created by encrypting the superimposed image with the encryption key and DNA encoding. The algorithm can be made even more secure by implementing a decryption mechanism that decrypts the image independently in order to reduce the possibility of information leaking during the decryption process. The outcomes of the simulation experiment demonstrate the algorithm’s strong security and resistance to interference such as noise pollution and lost image content.

17 January 2024 in Expert Systems with Applications
Verified
Journal Article
Closed access
JHJianpeng Hou
Jianpeng Hou
XWXiaofeng Wang
Xiaofeng Wang
RHRuidong Han
Ruidong Han
MJMao Jia
Mao Jia
DLDong Liu
Dong Liu
SPShanmin Pang
Shanmin Pang

22 December 2025 in Aerospace Science and Technology
Verified
Journal Article
Closed access
LDLiheng Dong
Liheng Dong
CTChengyang Tao
Chengyang Tao
ZZZHANG ZHAOXIANG
ZHANG ZHAOXIANG
NUNorthwestern University
Northwestern University
XYXu YueLei
Xu YueLei
DLDong Liu
Dong Liu

14 November 2025 in IEEE Transactions on Multimedia
Verified
Journal Article
Closed access
DLDong Liu
Dong Liu
XWXiaofeng Wang
Xiaofeng Wang
RHRuidong Han
Ruidong Han
JLJianghua Li
Jianghua Li
SPShanmin Pang
Shanmin Pang
Image inpainting has attracted considerable attention in computer vision and image processing due to its wide range of applications. While deep learning-based methods have shown promising potential, accurately recovering pixel-level details remains a significant challenge, particularly in the presence of large and irregular missing regions. Furthermore, existing methods are limited by unidirectional semantic guidance and a localized understanding of global structural context. In this study, we propose a mask-guided dual-branch Transformer-based framework, named MDT-FI, which effectively balances local detail restoration and global contextual reasoning by explicitly modeling long-range dependencies. MDT-FI consists of three key components: the Interactive Attention Module (IAM), the Spectral Harmonization Module (SHM), and the Lateral Adaptation Network (LAN). The model integrates multi-scale feature interaction, frequency-domain information fusion, and a mask-guided attention mechanism to progressively build cross-level feature associations. This design facilitates multi-level representation learning and optimization, thereby enhancing local texture synthesis while preserving global structural consistency. To further improve perceptual quality, a feature augmenter is employed to assess the fidelity of both texture and structure in the generated results. Extensive experiments on CelebA-HQ, Places2, and Paris Street View demonstrate that MDT-FI significantly outperforms state-of-the-art methods.

27 April 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
DLDong Liu
Dong Liu
YXYuelei Xu
Yuelei Xu
XWXiaofeng Wang
Xiaofeng Wang
ZZZhaoxiang Zhang
Zhaoxiang Zhang
SXShangjie Xue
Shangjie Xue
SPShanmin Pang
Shanmin Pang
RGB-D salient object detection (SOD) relies on effective synergy between RGB and depth modalities for precise localization of salient objects. However, semantic inconsistencies between modalities weaken structural and edge information during fusion, destabilizing salient object representations across scales and limiting detection performance. To address this, we propose a cross-modal edge-guided consistency learning framework (CMCL-Net), stabilizing semantic distributions by jointly modeling cross-modal consistency and structural integrity. Specifically, the semantic coordination and consistency fusion (SCCF) module employs a content-aware weighting mechanism to adaptively align cross-modal features and enhance complementarity. The fine-grained edge guidance (FEGM) module leverages explicit edge supervision to strengthen structural perception of object boundaries, improving contour integrity and edge clarity. To handle large scale variations of salient objects, the multi-scale salient feature perception (MSFP) module integrates global semantic context with local details, enhancing multi-scale salient region modeling. Extensive experiments demonstrate that CMCL-Net outperforms state-of-the-art methods across multiple RGB-D SOD benchmarks.