Liang Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Liang Liu has a total of 18 co-authors. Here are the top co-authors by number of shared publications:
- Shuaiyong Li
Chongqing University of Posts and Telecommunications
7 shared publications - Yongqiang Zhu
Key Laboratory of Industrial Internet of Things and Networked Control, Ministry of Education, Chongqing University of Posts and Telecommunications, Chongqing, China
3 shared publications - Chao Zhang
Key Laboratory of Industrial Internet of Things and Networked Control, Ministry of Education, Chongqing University of Posts and Telecommunications, Chongqing, China
2 shared publications - Song Wang
Southwest University of Science and Technology
2 shared publications - Xuyuntao Zhang
Key Laboratory of Industrial Internet of Things and Networked Control, Ministry of Education, Chongqing University of Posts and Telecommunications, Chongqing, China
2 shared publications - Zhengxu Dai
School of Automation, Chongqing University of Posts and Telecommunications, Chongqing, China
2 shared publications
Latest Publications
- 3 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28 (15209210) , 1601-1612
Modality-Aware Gated Attention Network for Audio-Visual Event Localization
Audio-visual event localization (AVEL) refers to the identification of the category and the corresponding temporal boundaries of an event that is both visually and audibly discernible in unconstrained videos. However, the event-irrelevant background (e.g., ambient noise or visual occlusion) and event-specific modal biases often lead to audio-visual semantic inconsistency. Existing methods utilize modality-guided attention to suppress background interference, but they neglect this attention inevitably introduces redundant or irrelevant information from the other modality. To alleviate this problem, we propose a novel Modality-Aware Gated Attention Network (MAGAN) that focuses on event-relevant visual regions, consolidates informative audio frequencies, and captures event-specific modality biases. Specifically, a cross-modal gated co-attention (CMGCA) scheme is presented for modeling the correspondence between the potential (self-guided) localization maps and the modality-guided localization maps through two gated components, i.e., audio-to-visual attention and visual-to-audio attention. Furthermore, a cross-modal gated co-interaction (CMGCI) mechanism that incorporates both unimodal gated interaction and multimodal gated interaction is introduced to capture event-specific modality biases by considering unimodal independence and multimodal synergy simultaneously. Extensive experiments on the AVE dataset demonstrate the superiority and effectiveness of our model over state-of-the-art approaches in both fully- and weakly-supervised AVE settings.
- 3 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28 (15209210) , 1601-1612
Modality-Aware Gated Attention Network for Audio-Visual Event Localization
Audio-visual event localization (AVEL) refers to the identification of the category and the corresponding temporal boundaries of an event that is both visually and audibly discernible in unconstrained videos. However, the event-irrelevant background (e.g., ambient noise or visual occlusion) and event-specific modal biases often lead to audio-visual semantic inconsistency. Existing methods utilize modality-guided attention to suppress background interference, but they neglect this attention inevitably introduces redundant or irrelevant information from the other modality. To alleviate this problem, we propose a novel Modality-Aware Gated Attention Network (MAGAN) that focuses on event-relevant visual regions, consolidates informative audio frequencies, and captures event-specific modality biases. Specifically, a cross-modal gated co-attention (CMGCA) scheme is presented for modeling the correspondence between the potential (self-guided) localization maps and the modality-guided localization maps through two gated components, i.e., audio-to-visual attention and visual-to-audio attention. Furthermore, a cross-modal gated co-interaction (CMGCI) mechanism that incorporates both unimodal gated interaction and multimodal gated interaction is introduced to capture event-specific modality biases by considering unimodal independence and multimodal synergy simultaneously. Extensive experiments on the AVE dataset demonstrate the superiority and effectiveness of our model over state-of-the-art approaches in both fully- and weakly-supervised AVE settings.
- 11 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-15
Multipath Siamese Neural Network for Intelligent Fault Diagnosis Under Small Samples
The scarcity of fault samples severely limits the performance of traditional fault diagnosis methods under small samples. To address this issue, this article proposes a multipath Siamese neural network (MSN)-based algorithm. MSN ensures the superior fault identification despite limited training samples by integrating information from multiple data types. Specifically, the MSN incorporates a multiscale convolutional module (MCM) and wide convolutional kernels to extract richer feature representations while minimizing the noise interference. In addition, a classification network is integrated into the Siamese network (SN) framework, enabling direct category predictions during testing instead of pairwise similarity computations. A metric neural network (MNN) is also introduced to automatically learn feature similarities during training, thereby improving the model’s generalization capability and accuracy. Experimental evaluations on two distinct datasets demonstrate that the MSN achieves accurate fault detection under small-sample conditions with the robust generalization performance.
- 11 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-15
Multipath Siamese Neural Network for Intelligent Fault Diagnosis Under Small Samples
The scarcity of fault samples severely limits the performance of traditional fault diagnosis methods under small samples. To address this issue, this article proposes a multipath Siamese neural network (MSN)-based algorithm. MSN ensures the superior fault identification despite limited training samples by integrating information from multiple data types. Specifically, the MSN incorporates a multiscale convolutional module (MCM) and wide convolutional kernels to extract richer feature representations while minimizing the noise interference. In addition, a classification network is integrated into the Siamese network (SN) framework, enabling direct category predictions during testing instead of pairwise similarity computations. A metric neural network (MNN) is also introduced to automatically learn feature similarities during training, thereby improving the model’s generalization capability and accuracy. Experimental evaluations on two distinct datasets demonstrate that the MSN achieves accurate fault detection under small-sample conditions with the robust generalization performance.
- 23 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (6) , 9802-9816
Temporal Convolutional Network for Gas Concentration Prediction Based on Weighted Loss and Channel Coupling Attention
Gas sensors form a critical foundation for gas concentration prediction (GCP) tasks, providing the necessary technological support for the transmission and processing of industrial gas data. However, existing deep learning-based gas concentration prediction models often overlook the coupling information between features and have limited generalizability. Consequently, these models exhibit low prediction accuracy when dealing with data characterized by numerous anomalies and significant fluctuations. Therefore, this paper proposes a gas prediction model based on the Weighted Loss channel coupled attentional time convolutional network (WLCCTCN), which aims to capture the coupled information among channels and suppress the abundance of abnormal data with large fluctuations in different environments. The WLCCTCN network primarily consists of three structures: 1) causal dilated convolution structure; 2) channel-coupled attention structure; 3) a weighted loss function. Firstly, the design of the strip matrix of the channel coupled attention module (CCAM) is utilized to capture the complex dependencies between multiple features fully. Secondly, the weighted loss function (EALoss) is employed to suppress abnormal features and fluctuations in gas data. Finally, experimental results with various gas datasets demonstrate that the proposed model captures coupling information between channels in gas concentration data and effectively suppresses data abnormalities and fluctuations. This proves the superior predictive capability of the proposed model compared to existing state-of-the-art methods.
- 23 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (6) , 9802-9816
Temporal Convolutional Network for Gas Concentration Prediction Based on Weighted Loss and Channel Coupling Attention
Gas sensors form a critical foundation for gas concentration prediction (GCP) tasks, providing the necessary technological support for the transmission and processing of industrial gas data. However, existing deep learning-based gas concentration prediction models often overlook the coupling information between features and have limited generalizability. Consequently, these models exhibit low prediction accuracy when dealing with data characterized by numerous anomalies and significant fluctuations. Therefore, this paper proposes a gas prediction model based on the Weighted Loss channel coupled attentional time convolutional network (WLCCTCN), which aims to capture the coupled information among channels and suppress the abundance of abnormal data with large fluctuations in different environments. The WLCCTCN network primarily consists of three structures: 1) causal dilated convolution structure; 2) channel-coupled attention structure; 3) a weighted loss function. Firstly, the design of the strip matrix of the channel coupled attention module (CCAM) is utilized to capture the complex dependencies between multiple features fully. Secondly, the weighted loss function (EALoss) is employed to suppress abnormal features and fluctuations in gas data. Finally, experimental results with various gas datasets demonstrate that the proposed model captures coupling information between channels in gas concentration data and effectively suppresses data abnormalities and fluctuations. This proves the superior predictive capability of the proposed model compared to existing state-of-the-art methods.
- 29 August 2024
- Published by Elsevier in Neurocomputing
- Vol. 609
Polycentric intuitionistic fuzzy weighted least squares twin SVMs
- 29 August 2024
- Published by Elsevier in Neurocomputing
- Vol. 609
Polycentric intuitionistic fuzzy weighted least squares twin SVMs
- 8 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11 (10) , 18593-18604
Gated Transient Fluctuation Dual Attention Unit Network for Long-Term Remaining Useful Life Prediction of Rotating Machinery Using IIoT
The Industrial Internet of Things (IIOT) has greatly facilitated the development of prognostics health management (PHM) for rotating machinery. In the era of IIOT, the long-term prediction of remaining useful life (RUL) can provide a reliable basis for maintenance decisions of rotating machinery. However, most existing RUL prediction methods neglect to learn the transient fluctuation information in the health indicator (HI), which leads to low long-term prediction accuracy. Therefore, a new gated transient fluctuation dual attention unit (GTFDAU) network is designed to achieve sufficient learning of transient fluctuation information, which can improve long-term prediction accuracy. The GTFDAU network mainly comprises three novel gate structures: transient fluctuation gate, historical attention gate and current attention gate. The transient fluctuation gate is used to capture transient fluctuation information, which can deeply mine the historical information. The historical attention gate is used to adaptively focus on transient fluctuation information and the hidden state of the last moment, which can effectively reshape the historical information. The current attention gate is used to adaptively focus on historical information and current input information, which can comprehensively update the current hidden state. The proposed method has superior long-term prediction capability compared to existing state-of-the-art methods without compromising short-term prediction performance through experimental verification of RUL for rolling bearings and fatigue gears.
- 8 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11 (10) , 18593-18604
Gated Transient Fluctuation Dual Attention Unit Network for Long-Term Remaining Useful Life Prediction of Rotating Machinery Using IIoT
The Industrial Internet of Things (IIOT) has greatly facilitated the development of prognostics health management (PHM) for rotating machinery. In the era of IIOT, the long-term prediction of remaining useful life (RUL) can provide a reliable basis for maintenance decisions of rotating machinery. However, most existing RUL prediction methods neglect to learn the transient fluctuation information in the health indicator (HI), which leads to low long-term prediction accuracy. Therefore, a new gated transient fluctuation dual attention unit (GTFDAU) network is designed to achieve sufficient learning of transient fluctuation information, which can improve long-term prediction accuracy. The GTFDAU network mainly comprises three novel gate structures: transient fluctuation gate, historical attention gate and current attention gate. The transient fluctuation gate is used to capture transient fluctuation information, which can deeply mine the historical information. The historical attention gate is used to adaptively focus on transient fluctuation information and the hidden state of the last moment, which can effectively reshape the historical information. The current attention gate is used to adaptively focus on historical information and current input information, which can comprehensively update the current hidden state. The proposed method has superior long-term prediction capability compared to existing state-of-the-art methods without compromising short-term prediction performance through experimental verification of RUL for rolling bearings and fatigue gears.
- 22 January 2024
- Published by Springer Nature
- p. 213-224
Visual Analytics of CO$_2$ Emissions from Individuals’ Daily Travel Based on Large-Scale Taxi Trajectories- 22 January 2024
- Published by Springer Nature
- p. 213-224
Visual Analytics of CO$_2$ Emissions from Individuals’ Daily Travel Based on Large-Scale Taxi Trajectories- 21 May 2023
- Published by Elsevier in Process Safety and Environmental Protection
- Vol. 175, 447-460
A novel DVAPSO-LSTSVM classifier in compressed sensing domain for intelligent pipeline leakage diagnosis
- 21 May 2023
- Published by Elsevier in Process Safety and Environmental Protection
- Vol. 175, 447-460
A novel DVAPSO-LSTSVM classifier in compressed sensing domain for intelligent pipeline leakage diagnosis
- 28 April 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Sustainable Computing
- Vol. 9 (3) , 299-314
Low-Carbon Mixed Traffic Route Recommendation for Community Residents Based on Multilayer Complex Traffic Network
With the proposed “carbon peaking” and “carbon neutral” goals in China, the transportation sector, as the second largest consumer of oil and a major producer of greenhouse gases, is a critical area for energy efficiency and emission reduction actions. However, few studies have focused on the effective combination of solving residents' commuting challenges and low-carbon travel. In this paper, by extracting real traffic flow data from taxi and bike-sharing trajectory data, a multilayer complex traffic network is formed to realize an interactive visual exploration of urban traffic patterns. Based on this network a low-carbon travel route recommendation is implemented using a modified genetic algorithm to reduce personal carbon emission and travel costs. Meanwhile, the trip chain level carbon emission estimation method is defined for city streets and recommended routes. With the integration of the above algorithms, a visual analytics system is designed and implemented to support the joint exploration of urban traffic patterns and the street carbon emission distribution, low-carbon mixed traffic route recommendations for inter-community commuting, and optimization of low-carbon recommended routes by adjusting bike stations. Take the taxi and bike-sharing trajectory data in Xiamen, China as an example, an evaluation analysis of the system shows that the method is effective in reducing commuting costs for community residents while reducing personal travel carbon emission.
- 28 April 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Sustainable Computing
- Vol. 9 (3) , 299-314
Low-Carbon Mixed Traffic Route Recommendation for Community Residents Based on Multilayer Complex Traffic Network
With the proposed “carbon peaking” and “carbon neutral” goals in China, the transportation sector, as the second largest consumer of oil and a major producer of greenhouse gases, is a critical area for energy efficiency and emission reduction actions. However, few studies have focused on the effective combination of solving residents' commuting challenges and low-carbon travel. In this paper, by extracting real traffic flow data from taxi and bike-sharing trajectory data, a multilayer complex traffic network is formed to realize an interactive visual exploration of urban traffic patterns. Based on this network a low-carbon travel route recommendation is implemented using a modified genetic algorithm to reduce personal carbon emission and travel costs. Meanwhile, the trip chain level carbon emission estimation method is defined for city streets and recommended routes. With the integration of the above algorithms, a visual analytics system is designed and implemented to support the joint exploration of urban traffic patterns and the street carbon emission distribution, low-carbon mixed traffic route recommendations for inter-community commuting, and optimization of low-carbon recommended routes by adjusting bike stations. Take the taxi and bike-sharing trajectory data in Xiamen, China as an example, an evaluation analysis of the system shows that the method is effective in reducing commuting costs for community residents while reducing personal travel carbon emission.