A5060014765- Face and Expression Recognition
- Photoacoustic and Ultrasonic Imaging
- Face recognition and analysis
- Remote-Sensing Image Classification
- Nanoplatforms for cancer theranostics
- Video Surveillance and Tracking Methods
- Infrared Target Detection Methodologies
- Emotion and Mood Recognition
- Image and Video Stabilization
- Optical Imaging and Spectroscopy Techniques
- Remote Sensing and Land Use
- Advanced Image Fusion Techniques
- Advanced Computing and Algorithms
- Industrial Vision Systems and Defect Detection
- Image Enhancement Techniques
- Human Pose and Action Recognition
- Infrared Thermography in Medicine
- Thermography and Photoacoustic Techniques
- Extracellular vesicles in disease
- Biometric Identification and Security
- Angiogenesis and VEGF in Cancer
- Random lasers and scattering media
- Advanced Neural Network Applications
- Optical Coherence Tomography Applications
- Advanced Computational Techniques and Applications
Jiangxi Science and Technology Normal University
2016-2025
Shenzhen Institutes of Advanced Technology
2017-2024
Chinese Academy of Sciences
2011-2023
South China Agricultural University
2020
Wuhan University of Technology
2015
Changchun Institute of Optics, Fine Mechanics and Physics
2011
Jiangxi University of Finance and Economics
2008
Hunan University
2005-2007
Litchi is often harvested by clamping and cutting the branches, which are small can easily be damaged picking robot. Therefore, detection of litchi branches particularly significant. In this article, an fully convolutional neural network-based semantic segmentation algorithm proposed to semantically segment branches. First, DeepLabV3+ model combined with Xception depth separable convolution feature. Second, transfer learning data enhancement used accelerate convergence improve robustness...
Objectives:The objective of this study was to demonstrate the feasibility using noninvasive photoacoustic imaging technology along with novel semiconducting polymer nanoparticles for in vivo identifying inflammatory components carotid atherosclerosis and assessing severity inflammation mouse models.Methods Results: Healthy arteries atherosclerotic were imaged by system.Molecular probes PBD-CD36 used label cells show information imaging.In experiments, we observed maximum signal enhancement...
Semiconducting polymer nanoparticles (PDPPTBZ NPs) with a high mass extinction coefficient of 43 mL mg<sup>−1</sup> cm<sup>−1</sup> at 1064 nm have been devised as contrast agent for deep photoacoustic imaging gliomas under ultralow laser fluence (4 mJ cm<sup>−2</sup>).
The synthesis of sMoSe<sub>2</sub>–ICG NSs for highly efficient tumor photoacoustic imaging guided photothermal therapy.
Myocardial infarctions are most often caused by the so-called vulnerable plaques, usually featured as non-obstructive lesions with a lipid-rich necrotic core, thin-cap fibroatheroma, and large plaque size. The identification quantification of these characteristics keys to evaluate vulnerability. However, single modality intravascular methods, such ultrasound, optical coherence tomography photoacoustic, can hardly achieve all comprehensive information satisfy clinical needs. In this paper,...
Rupture or erosion of inflammatory atherosclerotic vulnerable plaque is essential to acute coronary events, while the target intervene very challenging, due relatively small volume, high hemodynamic shear stress, and multifactorial nature lesion foci. Herein, we utilize biological functionality neutrophil versatility microbubble in acoustic field, form Neu-balloon through CD11b antibody binding. The inherits advantage neutrophils on firming endothelium adhesion even at stress up 16 dyne/cm2...
Phase demodulation is the core of fringe projection profilometry systems. However, current U-Net-based phase approaches demonstrate deficiencies in global context propagation, adversely affecting wrapped extraction precision. To this end, paper proposes a deep-learning-based model for single-shot extraction, named full-scale connection and attention enhancement network (SEC-UNet3+). The mitigates limitations traditional U-Net architecture by introducing cross-layer feature integration module...
Intravascular photoacoustic (IVPA) imaging technology enables the visualization of pathological characteristics (such as inflammation activities, lipid deposition) artery wall. Blood flushing is a necessary step in improving quality vivo IVPA imaging. But limited speed systems stretches their time, which an important obstacle clinical translations. In this paper, we report improvement IVPA/IVUS to 100 frames per second. The high-speed demonstrated rabbit , visualizing nanoparticles...
Current intravascular imaging modalities face hurdles in the molecular evaluation of progressed plaques. This study aims to construct a novel hybrid system (intravascular ultrasound/intravascular photoacoustic, IVPA/IVUS) via RGDfk peptide-targeted nanoparticles for monitoring angiogenesis atherosclerotic plaques rabbit model. An model was induced by abdominal aorta balloon de-endothelialization followed high-fat diet. A human serum albumin (HSA)-based nanoprobe modified with peptide...
A micro-expression (ME), as a spontaneous facial expression, usually occurs instantaneously and is difficult to disguise after an emotion-evoking event. Numerous convolutional neural network (CNN)-based models have been widely explored recognize MEs for their strong local feature representation ability on images. However, the main drawback of current methods inability fully extracting holistic contextual information from ME To achieve efficient learning diverse perspectives, this paper uses...
The endometrium microvessel system, responsible for supplying oxygen and nutrients to the embryo, holds significant importance in evaluating endometrial receptivity (ER). Visualizing this system directly can significantly enhance ER evaluation. Currently, clinical methods like Narrow-band hysteroscopy Color Doppler ultrasound are commonly used uterine blood vessel examination, but they have limitations depth or resolution. Endoscopic Photoacoustic Imaging (PAE) has proven effective...
To address the challenges of excessive model parameters and low detection accuracy in printed circuit board (PCB) defect detection, this paper proposes a novel PCB based on improved RTDETR (Real-Time Detection, Embedding Tracking) method, named MFAD-RTDETR. Specifically, proposed introduces designed Detail Feature Retainer (DFR) into original backbone to capture retain local details. Subsequently, Mamba architecture, Visual State Space (VSS) module is integrated enhance global attention...
Abstract The power grid dispatching system is a critical component in the development of industry and technological innovation smart grids. However, as systems expand, traditional approach to managing through telephone communication no longer sufficient meet growing data processing demands Consequently, this paper introduces an algorithm named MKConvLFormer (multi-scale residual convolutional neural network Linformer). streaming speech recognition designed for semantic environment grid,...
The traditional LBP Histogram representation extracts the local micro-patterns and assigns same weight all micro-patterns. To combine different contribution to face recognition, this paper proposes a weighted histogram based on Weber's law. Firstly, inspired by psychological law, intensity of micro-pattern is defined ratio between two terms: one relative differences central pixel against its neighbors, other pixel. Secondly, regarding as weight, constructed with weight. Finally, make full...