A5100721041- Image Enhancement Techniques
- Image and Signal Denoising Methods
- Acoustic Wave Resonator Technologies
- Image Processing and 3D Reconstruction
- Advanced Sensor and Control Systems
- Advanced Neural Network Applications
- Advanced Image Fusion Techniques
- Medical Imaging Techniques and Applications
- Ferroelectric and Piezoelectric Materials
- Radiomics and Machine Learning in Medical Imaging
- Advanced Steganography and Watermarking Techniques
- Modular Robots and Swarm Intelligence
- Advanced Algorithms and Applications
- Optical Polarization and Ellipsometry
- Medical Image Segmentation Techniques
- Advanced MRI Techniques and Applications
- Advanced Data Compression Techniques
- Mineral Processing and Grinding
- Simulation and Modeling Applications
- Industrial Technology and Control Systems
- Advanced Vision and Imaging
- Human Pose and Action Recognition
- Iterative Learning Control Systems
- Advanced Image and Video Retrieval Techniques
- Image Processing Techniques and Applications
Shanghai Institute of Microsystem and Information Technology
2021-2025
Chinese Academy of Sciences
2005-2025
University of Chinese Academy of Sciences
2021-2025
Air Force Medical University
2024
Chinese University of Hong Kong
2021-2024
Nanjing University of Aeronautics and Astronautics
2022-2023
China Telecom
2023
Xi'an Jiaotong University
2005-2023
Qilu University of Technology
2021-2023
Shandong Academy of Sciences
2023
In this work, we report the set-up and results of Liver Tumor Segmentation Benchmark (LiTS), which was organized in conjunction with IEEE International Symposium on Biomedical Imaging (ISBI) 2017 Conferences Medical Image Computing Computer-Assisted Intervention (MICCAI) 2018. The image dataset is diverse contains primary secondary tumors varied sizes appearances various lesion-to-background levels (hyper-/hypo-dense), created collaboration seven hospitals research institutions. Seventy-five...
In this work, we report the set-up and results of Liver Tumor Segmentation Benchmark (LiTS), which was organized in conjunction with IEEE International Symposium on Biomedical Imaging (ISBI) 2017 Conferences Medical Image Computing Computer-Assisted Intervention (MICCAI) 2018. The image dataset is diverse contains primary secondary tumors varied sizes appearances various lesion-to-background levels (hyper-/hypo-dense), created collaboration seven hospitals research institutions. Seventy-five...
Traditional lidar techniques mainly rely on the backscattering/echo light intensity and spectrum as information sources. In contrast, polarization (P-lidar) expands dimensions of detection by utilizing physical property polarization. By incorporating parameters such degree, angle, ellipticity, P-lidar enhances richness obtained from target objects, providing advantages for subsequent analysis. Over past five decades, application fields have rapidly expanded, starting its early use in...
In this Letter, we present a Stokes imaging-based method to restore objects and enhance image contrast in turbid water. the system, light source illuminates with two orthometric polarization states; based on new decomposition model, recorded images are converted maps subsequently restored clear image, free of reflections scattered lights. A mathematical model has been developed explain how undesired lights rejected. Imaging experiments have devised performed different objects, e.g., metals...
Machine learning models are widely deployed in production cloud to provide online inference services. Efficiently deploying services requires careful tuning of hardware and runtime configurations (e.g., GPU type, memory, batch size), which can significantly improve the model serving performance reduce cost. However, existing autoconfiguration approaches for general workloads, such as Bayesian optimization white-box prediction, inefficient navigating high-dimensional configuration space...
Vertebra segmentation from biplanar whole-spine radiographs is highly demanded in the quantitative assessment of scoliosis and resultant sagittal deformities. However, automatic vertebra extremely challenging due to low contrast, blended boundaries, superimposition many layers, especially plane. To alleviate these problems, we propose a lightweight pyramid attention quick refinement network (LPAQR-Net) for efficient accurate segmentation. The LPAQR-Net consists three components: (1) backbone...
Polarization characteristics are significantly crucial for tasks in various fields, including the remote sensing of oceans and atmosphere, as well polarization LIDAR polarimetric imaging scattering media. Many metrics (such degree polarization, angle diattenuation, depolarization) have been proposed to enrich characterization improve task performance media; yet, their related efficacy is limited, especially high turbidity conditions. The indices purity (IPPs), three different depolarization...
Promoting infrared (IR) response is crucial to boost the short‐circuit current density ( J SC ) of thinner silicon heterojunction (SHJ) solar cells. Herein, tantalum‐doped tin oxide (TaTO) film not only a transparent conductive but also used as rear reflector stacked with multiple‐doped indium (IMO) promote IR quantum efficiency SHJ Stack films (IMO/TaTO) are prepared and their structural optical–electrical properties were studied. IMO/TaTO stack layer polycrystalline IMO covered by...
Undersampling -space data in magnetic resonance imaging (MRI) reduces scan time but pose challenges image reconstruction. Considerable progress has been made reconstructing accelerated MRI. However, restoration of high-frequency details highly undersampled remains challenging. To address this issue, we propose CAMP-Net, an unrolling-based Consistency-Aware Multi-Prior Network for MRI CAMP-Net leverages complementary multi-prior knowledge and multi-slice information from various domains to...
Teeth are one of the most important materials for anthropological studies because they likely to be preserved in ancient remains. While frequencies dental characteristics can provide clues phylogeny populations, genetic at individual level further reveal biological mechanisms and evolutionary context characteristics. In this study, by analyzing 38 242 Xinjiang Uyghur individuals, we found that (i) Uyghurs showed evidence admixture between European East Asian populations. The proportions were...
Microhelices with unique three-dimensional (3D) helical structures have attracted great attention due to applications in various fields, especially magnetic microhelices can be applied as microrobots for removal of clogging substance microchannels, cargo transport, cell manipulation, and so on. Here, a facile flexible strategy is developed controllably fabricate droplet microfluidics. On-flow fabrication simply achieved by generating monodisperse droplets first, transforming the spherical...
In this work, a novel heterogeneous substrate of lithium tantalate thin film on silicon carbide (LiTaO <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</inf> -on-SiC, LTOSiC) was proposed and prepared. Since the SiC features an extremely-high phase velocity, low acoustic loss small thermal expansion coefficient (CTE), surface wave devices with high frequency, TCF were achieved LTOSiC substrate. The fabricated shear horizontal (SH-SAW) resonator...
Mobile platforms are now becoming a more and important medium for providing information to travelers on the move. To improve accuracy relevance of mobile traveler information, context awareness has become active research topic. In this paper, we describe algorithms services provided by Path2Go, multimodal system developed California PATH, UC Berkeley. The Path2Go activity detection algorithm, core context-aware design, made following contributions: (1) it uses rule-based multi-hypothesis...
Abstract Bulk Nb superconducting radio-frequency (SRF) cavities are widely used in accelerators, and their accelerating gradient general performance limited by the superheating field ( B sh ). To push theoretical limit of , new multilayer structures required. We fabricated FeSe-coated films using pulsed laser deposition, performed structural characterizations, measured transport magnetic properties for this superconductor-superconductor bilayer structure with smooth surface. Additionally, c1...
Fitness yoga is now a popular form of national fitness and sportive physical therapy. At present, Microsoft Kinect, depth sensor, other applications are widely used to monitor guide performance, but they inconvenient use still little expensive. To solve these problems, we propose spatial–temporal self-attention enhanced graph convolutional networks (STSAE-GCNs) that can analyze RGB video data captured by cameras or smartphones. In the STSAE-GCN, build module (STSAM), which effectively...
A novel method for online monitoring of ethanol concentration has been proposed by successfully developing an ethanol-responsive microfluidic membrane device.
This letter presents a new class of High-overtone Bulk Acoustic Resonators (HBARs) with only top electrodes but excellent performance using lithium niobate (LN) thin films on conductive silicon carbide (SiC). The 4-inch single-crystalline film SiC (LNCSiC) substrate was prepared by ion-cutting process. Since the serves as both resonant cavity and floating electrode, HBARs based LNCSiC were designed fabricated. devices show wide frequency response span exceeding 2.0 GHz, spacing 10 MHz, an f...
In the limestone/gypsum wet flue gas desulfurization technology, control process of absorber slurry pH value is a significantly non-linear, time-varying process. This paper proposed combined strategy for dealing with nonlinear control, which composed neural predictive controller (by controlling flow limestone to value) and feedback loop outlet SO2 concentration). introducing design implement steps results show that system has good dynamic performance robustness.
In this work, the phase velocity (V <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">p</inf> ) and electromechanical coupling factors (k xmlns:xlink="http://www.w3.org/1999/xlink">t</inf> <sup xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> of SH0 modes excited in X-cut LiNbO xmlns:xlink="http://www.w3.org/1999/xlink">3</inf> (LN) films were studied by finite element analysis, both them show strong dispersion when LN are bonded to foreign...
Early detection of incipient faults is a challenging task in the field chemical process monitoring. For this problem, paper proposes new data-driven monitoring method called stream data projection transformation analysis (SDPTA). First, we determine set vectors, orthogonal basis based on original to solve problem that space vector has relevance. Then, use sliding window project onto vectors obtain components which defined as transform (PTCs). In way, local sequence information can be...