Longru Chen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Longru Chen has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Gaoyuan Dai
School of Electronic and Information Engineering, South China University of Technology, Guangzhou 510642, China
2 shared publications - Dehuan Wan
Guangdong University of Finance
1 shared publication - Dongping An
China Telecom Stocks Co., Ltd., Beijing 100033, China
1 shared publication - Fei Ji
South China University of Technology, Guangzhou, China
1 shared publication - Jianxian Lu
Guangdong Branch of China Telecom Co., Ltd., Guangzhou 510180, China
1 shared publication - Jing Yuan
School of Information Engineering, East China Jiaotong University, Nanchang 330013, China
1 shared publication
Latest Publications
- Open Access
- 3 October 2024
- Published by MDPI AG in Electronics
- Vol. 13 (19) , 3919
CECL-Net: Contrastive Learning and Edge-Reconstruction-Driven Complementary Learning Network for Image Forgery Localization
While most current image forgery localization (IFL) deep learning models focus primarily on the foreground of tampered images, they often neglect the essential complementary background semantic information. This oversight tends to create significant gaps in these models’ ability to thoroughly interpret and understand a tampered image, thereby limiting their effectiveness in extracting critical tampering traces. Given the above, this paper presents a novel contrastive learning and edge-reconstruction-driven complementary learning network (CECL-Net) for image forgery localization. CECL-Net enhances the understanding of tampered images by employing a complementary learning strategy that leverages foreground and background features, where a unique edge extractor (EE) generates precise edge artifacts, and edge-guided feature reconstruction (EGFR) utilizes the edge artifacts to reconstruct a fully complementary set of foreground and background features. To carry out the complementary learning process more efficiently, we also introduce a pixel-wise contrastive supervision (PCS) method that attracts consistent regions in features while repelling different regions. Moreover, we propose a dense fusion (DF) strategy that utilizes multi-scale and mutual attention mechanisms to extract more discriminative features and improve the representational power of CECL-Net. Experiments conducted on two benchmark datasets, one Artificial Intelligence (AI)-manipulated dataset and two real challenge datasets, indicate that our CECL-Net outperforms seven state-of-the-art models on three evaluation metrics.
- 3 October 2024
- Published by MDPI AG in Electronics
- Vol. 13 (19) , 3919
Open AccessCECL-Net: Contrastive Learning and Edge-Reconstruction-Driven Complementary Learning Network for Image Forgery Localization
While most current image forgery localization (IFL) deep learning models focus primarily on the foreground of tampered images, they often neglect the essential complementary background semantic information. This oversight tends to create significant gaps in these models’ ability to thoroughly interpret and understand a tampered image, thereby limiting their effectiveness in extracting critical tampering traces. Given the above, this paper presents a novel contrastive learning and edge-reconstruction-driven complementary learning network (CECL-Net) for image forgery localization. CECL-Net enhances the understanding of tampered images by employing a complementary learning strategy that leverages foreground and background features, where a unique edge extractor (EE) generates precise edge artifacts, and edge-guided feature reconstruction (EGFR) utilizes the edge artifacts to reconstruct a fully complementary set of foreground and background features. To carry out the complementary learning process more efficiently, we also introduce a pixel-wise contrastive supervision (PCS) method that attracts consistent regions in features while repelling different regions. Moreover, we propose a dense fusion (DF) strategy that utilizes multi-scale and mutual attention mechanisms to extract more discriminative features and improve the representational power of CECL-Net. Experiments conducted on two benchmark datasets, one Artificial Intelligence (AI)-manipulated dataset and two real challenge datasets, indicate that our CECL-Net outperforms seven state-of-the-art models on three evaluation metrics.
- Open Access
- 9 November 2023
- Published by MDPI AG in Electronics
- Vol. 12 (22) , 4577
Towards Flawless Designs: Recent Progresses in Non-Orthogonal Multiple Access Technology
High effectiveness and high reliability are two fundamental concerns in data transmission. Non-orthogonal multiple-access (NOMA) technology presents a promising solution for high-speed data transmission, which has long been pursued by academia and industry. However, there is still a significant road ahead for it to effectively support a wide range of applications. This paper provides a comprehensive study, comparison, and classification of the current advanced NOMA schemes from the perspectives of single-carrier (SC) systems, multicarrier (MC) systems, reconfigurable-intelligent-surface (RIS)-assisted systems, and deep-learning (DL)-assisted systems. Specifically, system implementation issues such as the transition from SC-NOMA to MC-NOMA, the relaxation of distinct channel gains, the consideration of imperfect channel knowledge, and the mitigation of error propagation/intra-group interference are involved. To begin with, we present an overview of the state-of-the-art developments related to the advanced design of SC-NOMA. Subsequently, a generalized MC-NOMA framework that provides the diversity–multiplexing gain by enhancing users’ signal-to-interference-plus-noise ratio (SINR) is proposed for better system performance. Moreover, we delve into discussions on RIS-assisted NOMA systems, where the receiver’s SINR can be enhanced by intelligently reconfiguring the reflected signal propagations. Finally, we analyze designs that combine NOMA/RIS-NOMA with DL to achieve highly efficient data transmission. We also identify key trends and future directions in deep-learning-based NOMA frameworks, providing valuable insights for researchers in this field.
- 9 November 2023
- Published by MDPI AG in Electronics
- Vol. 12 (22) , 4577
Open AccessTowards Flawless Designs: Recent Progresses in Non-Orthogonal Multiple Access Technology
High effectiveness and high reliability are two fundamental concerns in data transmission. Non-orthogonal multiple-access (NOMA) technology presents a promising solution for high-speed data transmission, which has long been pursued by academia and industry. However, there is still a significant road ahead for it to effectively support a wide range of applications. This paper provides a comprehensive study, comparison, and classification of the current advanced NOMA schemes from the perspectives of single-carrier (SC) systems, multicarrier (MC) systems, reconfigurable-intelligent-surface (RIS)-assisted systems, and deep-learning (DL)-assisted systems. Specifically, system implementation issues such as the transition from SC-NOMA to MC-NOMA, the relaxation of distinct channel gains, the consideration of imperfect channel knowledge, and the mitigation of error propagation/intra-group interference are involved. To begin with, we present an overview of the state-of-the-art developments related to the advanced design of SC-NOMA. Subsequently, a generalized MC-NOMA framework that provides the diversity–multiplexing gain by enhancing users’ signal-to-interference-plus-noise ratio (SINR) is proposed for better system performance. Moreover, we delve into discussions on RIS-assisted NOMA systems, where the receiver’s SINR can be enhanced by intelligently reconfiguring the reflected signal propagations. Finally, we analyze designs that combine NOMA/RIS-NOMA with DL to achieve highly efficient data transmission. We also identify key trends and future directions in deep-learning-based NOMA frameworks, providing valuable insights for researchers in this field.