Yuzhuo Li
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
Yuzhuo Li has a total of 35 co-authors. Here are the top co-authors by number of shared publications:
- Xisheng Li
Beijing Engineering Research Center of Industrial Spectrum Imaging, School of Automation and Electrical Engineering, University of Science and Technology Beijing, Beijing 100083, China
3 shared publications Hongbing Chen
Research Institute of Urbanization and Urban Safety, School of Future Cities, University of Science and Technology Beijing, Beijing, China
2 shared publications- Jia You
Key Laboratory of Knowledge Automation for Industrial Processes of Ministry of Education, School of Automation and Electrical Engineering, University of Science and Technology Beijing, Beijing, China
2 shared publications - Wenqian Feng
School of Mechanical Engineering, Tiangong University, Tianjin, China
2 shared publications - Xiangdong Ma
School of Automation and Electrical Engineering, University of Science and Technology Beijing, Beijing, China
2 shared publications Xisheng Li
2 shared publications
Latest Publications
- 30 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. PP (1530437X) , 1
Attention-enhanced GRU-based temperature compensation method for fiber-optic gyroscope
Fiber-optic gyroscope temperature compensation represents a critical challenge that is addressed to achieve high-precision applications. Here, we propose an attention-enhanced Gated Recurrent Unit (GRU)-based temperature compensation method for fiber-optic gyroscope. A fiber-optic gyroscope temperature compensation experimental platform was constructed for data acquisition. The temperature and its rate-of-change sequences were fused into a combined vector to serve as the input for training the GRU model. The GRU was enhanced by an attention mechanism for dynamic feature focusing, and the model was optimized through a composite loss function during backpropagation. The final evaluation results indicate that the proposed GRU-based temperature compensation method effectively suppresses the temperature drift of the fiber-optic gyroscope. Over the full temperature range of -40 °C to 60 °C, the temperature-induced bias peak-to-peak value decreased significantly from 4.47 °/h to 0.28 °/h. This represents an approximately 16-fold improvement in temperature stability over the uncompensated performance. Moreover, this method also exhibits notable advantages over the conventional polynomial fitting approach. This research provides theoretical foundation and technical support for the application of artificial intelligence technology in the field of fiber-optic gyroscope temperature compensation, holding significant practical value.
- 30 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. PP (1530437X) , 1
Attention-enhanced GRU-based temperature compensation method for fiber-optic gyroscope
Fiber-optic gyroscope temperature compensation represents a critical challenge that is addressed to achieve high-precision applications. Here, we propose an attention-enhanced Gated Recurrent Unit (GRU)-based temperature compensation method for fiber-optic gyroscope. A fiber-optic gyroscope temperature compensation experimental platform was constructed for data acquisition. The temperature and its rate-of-change sequences were fused into a combined vector to serve as the input for training the GRU model. The GRU was enhanced by an attention mechanism for dynamic feature focusing, and the model was optimized through a composite loss function during backpropagation. The final evaluation results indicate that the proposed GRU-based temperature compensation method effectively suppresses the temperature drift of the fiber-optic gyroscope. Over the full temperature range of -40 °C to 60 °C, the temperature-induced bias peak-to-peak value decreased significantly from 4.47 °/h to 0.28 °/h. This represents an approximately 16-fold improvement in temperature stability over the uncompensated performance. Moreover, this method also exhibits notable advantages over the conventional polynomial fitting approach. This research provides theoretical foundation and technical support for the application of artificial intelligence technology in the field of fiber-optic gyroscope temperature compensation, holding significant practical value.
- 5 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 75 (00189456) , 1-9
High-Accuracy Dynamic North-Finding Method Based on Dual-Polarization Fiber-Optic Angular Accelerometer
North-finding technology is an autonomous orientation technique used to determine the true north direction of carrier platforms, and it holds significant application value in fields such as seismic exploration and aerospace. In this work, we propose, for the first time to the best of our knowledge, a high-accuracy dynamic north-finding method based on a dual-polarization fiber-optic angular accelerometer (DP-FOAA). The angular accelerometer exhibits high sensitivity and a flat response across a broad frequency band. With an enclosed fiber coil area of 12.4 m2, it achieves a sensitivity of 2.7 × 10-⁷ rad/s²/(Hz)1/2 within the frequency range of 0.01 - 100 Hz. A north-finding system was stablished using this angular accelerometer integrated with a high-precision turntable. The novel north-finding method based on the DP-FOAA, together with its effective suppression of turntable vibration noise, has significantly enhanced the accuracy of dynamic north finding. The experimental results show that the north-finding accuracy of this system is better than 52 arcsec, with a total north-finding time of 18 s. Compared to a fiber-optic gyroscope with the same fiber coil, its measurement accuracy is improved by 3.2 times. This research provides a feasible approach for the development of novel high-speed, high-accuracy north finders, thereby advancing their technological progress.
- 5 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 75 (00189456) , 1-9
High-Accuracy Dynamic North-Finding Method Based on Dual-Polarization Fiber-Optic Angular Accelerometer
North-finding technology is an autonomous orientation technique used to determine the true north direction of carrier platforms, and it holds significant application value in fields such as seismic exploration and aerospace. In this work, we propose, for the first time to the best of our knowledge, a high-accuracy dynamic north-finding method based on a dual-polarization fiber-optic angular accelerometer (DP-FOAA). The angular accelerometer exhibits high sensitivity and a flat response across a broad frequency band. With an enclosed fiber coil area of 12.4 m2, it achieves a sensitivity of 2.7 × 10-⁷ rad/s²/(Hz)1/2 within the frequency range of 0.01 - 100 Hz. A north-finding system was stablished using this angular accelerometer integrated with a high-precision turntable. The novel north-finding method based on the DP-FOAA, together with its effective suppression of turntable vibration noise, has significantly enhanced the accuracy of dynamic north finding. The experimental results show that the north-finding accuracy of this system is better than 52 arcsec, with a total north-finding time of 18 s. Compared to a fiber-optic gyroscope with the same fiber coil, its measurement accuracy is improved by 3.2 times. This research provides a feasible approach for the development of novel high-speed, high-accuracy north finders, thereby advancing their technological progress.
- 9 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 44 (4) , 1588-1595
Portable Integrated Transceiver Architecture for Low-Cost Temperature Sensing
Temperature sensing has seen significant advancements across modern scientific applications. While fiber-optic temperature sensors (FOTS) offer multiple advantages, achieving a synergistic combination of cost-effectiveness, compact dimensions, and high integration remains a critical challenge for practical implementations. This work presents an artificial intelligence (AI)-driven solution featuring a miniaturized, cost-efficient fiber-optic leaky-mode temperature speckle sensor with integrated optoelectronic transceiver functionality. The system employs a fiber microsphere at the emission end for light source coupling and collection, while a fiber taper at the detection end facilitates modal leakage from the fiber core. Temperature-sensitive material integrated within the sensing region modulates the refractive index (RI) in response to environmental temperature variations, thereby altering the specklegram characteristics of the leaked modes. A convolutional vision transformer (ConViT)-based regression model was trained to establish quantitative mapping between speckle image features and temperature measurements. Comparative analysis demonstrated superior performance against convolutional neural network (CNN), vision transformer (Vit), and modified CNN architectures, achieving a median absolute error (MedAE) of 0.256 °C. This integrated approach effectively addresses the critical triad of cost reduction, miniaturization, and functional integration, potentially advancing the development of fiber-optic thermometry.
- 9 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 44 (4) , 1588-1595
Portable Integrated Transceiver Architecture for Low-Cost Temperature Sensing
Temperature sensing has seen significant advancements across modern scientific applications. While fiber-optic temperature sensors (FOTS) offer multiple advantages, achieving a synergistic combination of cost-effectiveness, compact dimensions, and high integration remains a critical challenge for practical implementations. This work presents an artificial intelligence (AI)-driven solution featuring a miniaturized, cost-efficient fiber-optic leaky-mode temperature speckle sensor with integrated optoelectronic transceiver functionality. The system employs a fiber microsphere at the emission end for light source coupling and collection, while a fiber taper at the detection end facilitates modal leakage from the fiber core. Temperature-sensitive material integrated within the sensing region modulates the refractive index (RI) in response to environmental temperature variations, thereby altering the specklegram characteristics of the leaked modes. A convolutional vision transformer (ConViT)-based regression model was trained to establish quantitative mapping between speckle image features and temperature measurements. Comparative analysis demonstrated superior performance against convolutional neural network (CNN), vision transformer (Vit), and modified CNN architectures, achieving a median absolute error (MedAE) of 0.256 °C. This integrated approach effectively addresses the critical triad of cost reduction, miniaturization, and functional integration, potentially advancing the development of fiber-optic thermometry.
- 2 December 2025
- Published by Elsevier in Optics & Laser Technology
- Vol. 194
Fraxicon composite fiber probe design for OCT imaging concurrent enhancement
- 2 December 2025
- Published by Elsevier in Optics & Laser Technology
- Vol. 194
Fraxicon composite fiber probe design for OCT imaging concurrent enhancement
- 3 November 2022
- Published by SAGE Publications in Textile Research Journal
- Vol. 93 (7-8) , 1699-1711
Review of cloth modeling and simulation for virtual fitting
Virtual fitting technology is widely used in many fields, such as e-commerce, garment computer-aided design, and film and game production. Cloth modeling and simulation play an important role in virtual fitting. The realism and simulation speed directly affect the user experience and visual appearance. This paper first reviews the history of cloth modeling and simulation methods. Then, it focuses on four perspectives: yarn-structure modeling of the cloth modeling level; multi-resolution grid simulation; the relation between human posture and cloth deformation; and collision problems of the cloth simulation level. Yarn-structure modeling considers a unit cell that incorporates the physical characteristics of the fabric and the fabric structure. A multi-resolution grid can integrate techniques from different research fields and may achieve a breakthrough in cloth modeling. Generative adversarial networks (GANs) have potential for dealing with the relation between human posture and cloth deformation by comparing several typical algorithms with GANs. Solving the collision and friction problem depends on choosing the right envelope box for refining the human body surface. Finally, the paper concludes with an analysis of future research trends that could improve the fidelity and speed of a simulation. The paper may serve as a reference for research into cloth modeling and simulation for virtual fitting.
- 3 November 2022
- Published by SAGE Publications in Textile Research Journal
- Vol. 93 (7-8) , 1699-1711
Review of cloth modeling and simulation for virtual fitting
Virtual fitting technology is widely used in many fields, such as e-commerce, garment computer-aided design, and film and game production. Cloth modeling and simulation play an important role in virtual fitting. The realism and simulation speed directly affect the user experience and visual appearance. This paper first reviews the history of cloth modeling and simulation methods. Then, it focuses on four perspectives: yarn-structure modeling of the cloth modeling level; multi-resolution grid simulation; the relation between human posture and cloth deformation; and collision problems of the cloth simulation level. Yarn-structure modeling considers a unit cell that incorporates the physical characteristics of the fabric and the fabric structure. A multi-resolution grid can integrate techniques from different research fields and may achieve a breakthrough in cloth modeling. Generative adversarial networks (GANs) have potential for dealing with the relation between human posture and cloth deformation by comparing several typical algorithms with GANs. Solving the collision and friction problem depends on choosing the right envelope box for refining the human body surface. Finally, the paper concludes with an analysis of future research trends that could improve the fidelity and speed of a simulation. The paper may serve as a reference for research into cloth modeling and simulation for virtual fitting.
- 4 March 2022
- Published by Taylor & Francis in The Journal of the Textile Institute
- Vol. 114 (3) , 488-495
Research status and prospect of visual image feature point detection in body measurements
To improve visual image feature point detection with two-dimensional human body circumference measurements, this paper analyzes and compares several feature point detection algorithms, algorithm based on human body proportion, and deep-learning-based convolutional neural network algorithms. Among them, the principles and application of the Harris corner detection algorithm are described in detail. The progress in image feature points detection based on human body proportions and regions is then summarized and reviewed. Based on the convolutional neural network method, a method of calculating feature points based on the relatively mature human joint point recognition algorithm is proposed, so as to solve the problem of inaccurate positioning in traditional algorithms. Finally, the challenge of feature point detection for clothing volumes is explained. Feature point detection from visual images has significant development prospects in the two-dimensional measurement of body circumference. Related research provides a reliable theoretical reference for realizing the accurate and rapid detection of human feature points in visual images.
- 4 March 2022
- Published by Taylor & Francis in The Journal of the Textile Institute
- Vol. 114 (3) , 488-495
Research status and prospect of visual image feature point detection in body measurements
To improve visual image feature point detection with two-dimensional human body circumference measurements, this paper analyzes and compares several feature point detection algorithms, algorithm based on human body proportion, and deep-learning-based convolutional neural network algorithms. Among them, the principles and application of the Harris corner detection algorithm are described in detail. The progress in image feature points detection based on human body proportions and regions is then summarized and reviewed. Based on the convolutional neural network method, a method of calculating feature points based on the relatively mature human joint point recognition algorithm is proposed, so as to solve the problem of inaccurate positioning in traditional algorithms. Finally, the challenge of feature point detection for clothing volumes is explained. Feature point detection from visual images has significant development prospects in the two-dimensional measurement of body circumference. Related research provides a reliable theoretical reference for realizing the accurate and rapid detection of human feature points in visual images.