Yu 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
Research Interests
Yu Liu has a total of 35 co-authors. Here are the top co-authors by number of shared publications:
- Guo-Qiang Tang
State Key Laboratory of Lithospheric and Environmental Coevolution, Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China
3 shared publications - Qiu-Li Li
State Key Laboratory of Lithospheric and Environmental Coevolution, Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China
3 shared publications - Ping Zhong
National Key Laboratory of Automatic Target Recognition, College of Electronic Science and Technology, National University of Defense Technology, Changsha 410073, China
2 shared publications - Wenlin Liu
National Key Laboratory of Automatic Target Recognition, College of Electronic Science and Technology, National University of Defense Technology, Changsha, China
2 shared publications - Xian-Hua Li
State Key Laboratory of Lithospheric and Environmental Coevolution, Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China
2 shared publications - Xikun Hu
National Key Laboratory of Science and Technology on Automatic Target Recognition, National University of Defense Technology, Changsha, China
2 shared publications
Latest Publications
- 11 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-14
Discriminative Latent-Space Learning for Fine-Grained Object Detection in Remote Sensing Images
Abstract—Fine-grained object detection (FOD) is essential in many remote sensing image interpretation tasks. Existing FOD methods have achieved remarkable progress in modeling discriminative features for FOD in remote sensing images. However, they receive unsatisfactory recognition accuracy due to the curse of dimensionality (CoD) problem. In this paper, we propose an orthogonal constraint-based discriminative latent-space learning (DLL) method to address the CoD problem. We first optimize a sparse optimization paradigm with convex relaxation to extract shared features between fine-grained objects into a low-dimensional latent-space. Then, we solve the orthogonal space of the latent-space for extracting irrelevant features related to common features from input features, i.e., discriminative features. The sparse optimization paradigm reprojects the underlying trends hidden in high-dimensional feature space into a low-dimensional latent space and hence addresses the CoD problem. We use neural parameters with latent-space and orthogonal constraints to approximately solve the proposed DLL, which can be efficiently optimized under convex programming tools. We theoretically prove the effectiveness of the proposed DLL. By adding our DLL to the existing deep learning-based object detection method, extensive experiments conducted on two datasets demonstrate that our method achieves superior performance compared with other state-of-the-art methods.
- 11 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-14
Discriminative Latent-Space Learning for Fine-Grained Object Detection in Remote Sensing Images
Abstract—Fine-grained object detection (FOD) is essential in many remote sensing image interpretation tasks. Existing FOD methods have achieved remarkable progress in modeling discriminative features for FOD in remote sensing images. However, they receive unsatisfactory recognition accuracy due to the curse of dimensionality (CoD) problem. In this paper, we propose an orthogonal constraint-based discriminative latent-space learning (DLL) method to address the CoD problem. We first optimize a sparse optimization paradigm with convex relaxation to extract shared features between fine-grained objects into a low-dimensional latent-space. Then, we solve the orthogonal space of the latent-space for extracting irrelevant features related to common features from input features, i.e., discriminative features. The sparse optimization paradigm reprojects the underlying trends hidden in high-dimensional feature space into a low-dimensional latent space and hence addresses the CoD problem. We use neural parameters with latent-space and orthogonal constraints to approximately solve the proposed DLL, which can be efficiently optimized under convex programming tools. We theoretically prove the effectiveness of the proposed DLL. By adding our DLL to the existing deep learning-based object detection method, extensive experiments conducted on two datasets demonstrate that our method achieves superior performance compared with other state-of-the-art methods.
- 27 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-14
MSP-O: A Two-Stage Mask Pretraining Method With Object Focus for Remote Sensing Object Detection
Pre-training is essential for enhancing performance across various downstream tasks, yet pre-training techniques for remote sensing image object detection remain limited. This is due to the distinct characteristics of remote sensing images, such as fixed imaging angles, complex backgrounds, and diverse target categories, which make them different from natural images. To address these challenges, this paper introduces MSP-O, an innovative two-stage mask pre-training method. Traditional single-stage pre-training frameworks aim to extract intrinsic features from large-scale datasets, improving generalization across downstream tasks. MSP-O builds upon this by adding a second-stage pre-training phase, which optimizes the model’s ability to focus on features crucial for detecting objects in downstream tasks. This additional stage allows the model to persistently acquire features beneficial for remote sensing object detection. Specifically, MSP-O incorporates a pixel-level restoration task focused on the objects of interest, guiding the model to prioritize the most critical areas for detection. This approach significantly enhances the model’s ability to discern key target features, improving both recognition precision and localization accuracy in subsequent detection tasks. Extensive experiments on the DOTA and DIOR datasets demonstrate the effectiveness of MSP-O, showing substantial improvements in detection performance compared to conventional methods.
- 27 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-14
MSP-O: A Two-Stage Mask Pretraining Method With Object Focus for Remote Sensing Object Detection
Pre-training is essential for enhancing performance across various downstream tasks, yet pre-training techniques for remote sensing image object detection remain limited. This is due to the distinct characteristics of remote sensing images, such as fixed imaging angles, complex backgrounds, and diverse target categories, which make them different from natural images. To address these challenges, this paper introduces MSP-O, an innovative two-stage mask pre-training method. Traditional single-stage pre-training frameworks aim to extract intrinsic features from large-scale datasets, improving generalization across downstream tasks. MSP-O builds upon this by adding a second-stage pre-training phase, which optimizes the model’s ability to focus on features crucial for detecting objects in downstream tasks. This additional stage allows the model to persistently acquire features beneficial for remote sensing object detection. Specifically, MSP-O incorporates a pixel-level restoration task focused on the objects of interest, guiding the model to prioritize the most critical areas for detection. This approach significantly enhances the model’s ability to discern key target features, improving both recognition precision and localization accuracy in subsequent detection tasks. Extensive experiments on the DOTA and DIOR datasets demonstrate the effectiveness of MSP-O, showing substantial improvements in detection performance compared to conventional methods.
- 11 September 2022
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Solid Earth
- Vol. 127 (9)
Desilicification Rims of Zircon Xenocrysts Record the Timing of Kimberlite Emplacement
Kimberlites are deep and relatively rare mantle‐derived igneous rocks, which provide key insights into the lithospheric mantle composition beneath continents (i.e., from the entrained mantle materials) at the time of magma emplacement. In addition, the temporal distribution of kimberlites throughout Earth's history may reveal secular changes in global geodynamics. However, alteration and abundant exotic material (i.e., xenoliths and xenocrysts) of different ages, both typical of kimberlites, often hinder reliable dating of these rocks. In this study, we investigated zircon xenocrysts and their replacement by desilicification reaction rims (DSRs), which consist mainly of baddeleyite, in the metamorphosed Kimozero kimberlites from the Karelian Craton, Russia. Similar DSRs have been identified previously on mantle‐derived zircons in numerous kimberlites, but the precise timing and location of formation of these rims remain unclear. U–Pb isotope dating of mantle zircon xenocrysts in the Kimozero kimberlites indicates that these rocks were emplaced at ca. 1.97 Ga. DSRs with similar textures and baddeleyite chemistry occur on both the 1.97 Ga mantle and 2.70–2.38 Ga crustal zircon xenocrysts in these kimberlites. Pb–Pb isotope dating of the DSRs yields the same age (i.e., ca. 1.97 Ga) as the mantle zircons. The data constrain the formation of DSRs to crustal levels and, more importantly, indicate that these rims record the timing of kimberlite emplacement. Therefore, we suggest that such baddeleyite reaction rims are a new and promising material for kimberlite geochronology.
- 11 September 2022
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Solid Earth
- Vol. 127 (9)
Desilicification Rims of Zircon Xenocrysts Record the Timing of Kimberlite Emplacement
Kimberlites are deep and relatively rare mantle‐derived igneous rocks, which provide key insights into the lithospheric mantle composition beneath continents (i.e., from the entrained mantle materials) at the time of magma emplacement. In addition, the temporal distribution of kimberlites throughout Earth's history may reveal secular changes in global geodynamics. However, alteration and abundant exotic material (i.e., xenoliths and xenocrysts) of different ages, both typical of kimberlites, often hinder reliable dating of these rocks. In this study, we investigated zircon xenocrysts and their replacement by desilicification reaction rims (DSRs), which consist mainly of baddeleyite, in the metamorphosed Kimozero kimberlites from the Karelian Craton, Russia. Similar DSRs have been identified previously on mantle‐derived zircons in numerous kimberlites, but the precise timing and location of formation of these rims remain unclear. U–Pb isotope dating of mantle zircon xenocrysts in the Kimozero kimberlites indicates that these rocks were emplaced at ca. 1.97 Ga. DSRs with similar textures and baddeleyite chemistry occur on both the 1.97 Ga mantle and 2.70–2.38 Ga crustal zircon xenocrysts in these kimberlites. Pb–Pb isotope dating of the DSRs yields the same age (i.e., ca. 1.97 Ga) as the mantle zircons. The data constrain the formation of DSRs to crustal levels and, more importantly, indicate that these rims record the timing of kimberlite emplacement. Therefore, we suggest that such baddeleyite reaction rims are a new and promising material for kimberlite geochronology.
- 1 June 2021
- Published by Elsevier in Geochimica et Cosmochimica Acta
- Vol. 302, 175-192
Radiogenic Pb mobilization induced by shock metamorphism of zircons in the Apollo 72255 Civet Cat norite clast
- 1 June 2021
- Published by Elsevier in Geochimica et Cosmochimica Acta
- Vol. 302, 175-192
Radiogenic Pb mobilization induced by shock metamorphism of zircons in the Apollo 72255 Civet Cat norite clast
- 1 March 2020
- Published by Elsevier in Geochimica et Cosmochimica Acta
- Vol. 272, 198-217
Origin of Al-rich chondrules in CV chondrites: Incorporation of diverse refractory components into the ferromagnesian chondrule-forming region
- 1 March 2020
- Published by Elsevier in Geochimica et Cosmochimica Acta
- Vol. 272, 198-217
Origin of Al-rich chondrules in CV chondrites: Incorporation of diverse refractory components into the ferromagnesian chondrule-forming region
- 1 November 2018
- Published by Elsevier in Journal of Asian Earth Sciences
- Vol. 167, 197-208
Constraints from zircon Hf-O-Li isotopic compositions on the genesis of slightly low-δ18O alkaline granites in the Taohuadao area, Zhejiang Province, SE China
- 1 November 2018
- Published by Elsevier in Journal of Asian Earth Sciences
- Vol. 167, 197-208
Constraints from zircon Hf-O-Li isotopic compositions on the genesis of slightly low-δ18O alkaline granites in the Taohuadao area, Zhejiang Province, SE China
- 25 June 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 20 (8) , 831-834
Bregman Iteration Based Efficient Algorithm for MR Image Reconstruction From Undersampled K-Space Data
It is difficult to solve Magnetic Resonance (MR) image reconstruction problems with linear combinations of total variation and ℓ1 norm regularization terms. In order to solve these compound regularization problems, we propose an efficient algorithm in this letter. The proposed algorithm adopts the Bregman iteration technique to convert the original constrained problem to a sequence of unconstrained problems, which are then solved using operator splitting and variable splitting techniques. Simulation experiments demonstrate that significant improvement of the quality of reconstructed images is achieved by the proposed algorithm when compared to previous methods.
- 25 June 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 20 (8) , 831-834
Bregman Iteration Based Efficient Algorithm for MR Image Reconstruction From Undersampled K-Space Data
It is difficult to solve Magnetic Resonance (MR) image reconstruction problems with linear combinations of total variation and ℓ1 norm regularization terms. In order to solve these compound regularization problems, we propose an efficient algorithm in this letter. The proposed algorithm adopts the Bregman iteration technique to convert the original constrained problem to a sequence of unconstrained problems, which are then solved using operator splitting and variable splitting techniques. Simulation experiments demonstrate that significant improvement of the quality of reconstructed images is achieved by the proposed algorithm when compared to previous methods.