A5100746164- Photoacoustic and Ultrasonic Imaging
- Optical Imaging and Spectroscopy Techniques
- Medical Imaging Techniques and Applications
- Ultrasound Imaging and Elastography
- Medical Image Segmentation Techniques
- Non-Invasive Vital Sign Monitoring
- Remote-Sensing Image Classification
- Wireless Signal Modulation Classification
- Advanced MRI Techniques and Applications
- Advanced Image Fusion Techniques
- Hydrocarbon exploration and reservoir analysis
- Image and Signal Denoising Methods
- Remote Sensing and Land Use
- Spectroscopy Techniques in Biomedical and Chemical Research
- Ultrasonics and Acoustic Wave Propagation
- Elasticity and Material Modeling
- Ultrasound and Hyperthermia Applications
- Image Retrieval and Classification Techniques
- Medical Imaging and Analysis
- Speech and Audio Processing
- Advanced Image and Video Retrieval Techniques
- Advanced X-ray and CT Imaging
- Domain Adaptation and Few-Shot Learning
- Infrared Thermography in Medicine
- Radiomics and Machine Learning in Medical Imaging
Xidian University
2016-2025
North Minzu University
2017-2025
ShanghaiTech University
2024
Tsinghua University
2010-2024
Jiangnan University
2024
East China Normal University
2024
China Geological Survey
2015-2024
China University of Geosciences
2023-2024
Institute of Psychology, Chinese Academy of Sciences
2024
Beijing Children’s Hospital
2024
Understanding the private car aggregation effect is conducive to a broad range of applications, from intelligent transportation management urban planning. However, this work challenging, especially on weekends, due inefficient representations spatiotemporal features for such and considerable randomness mobility weekends. In article, we propose deep learning framework attention network (STANet) with neural algorithm logic unit (NALU), so-called STANet-NALU, understand dynamic cars...
Hyperspectral images (HSIs) have gained high spectral resolution due to recent advances in imaging technologies. This incurs problems, such as an increased data scale and number of bands for HSIs, which results a complex correlation between different bands. In the applications remote sensing earth observation, ground objects represented by each HSI pixel are composed physical chemical non-Euclidean structures, classification (HIC) is becoming more challenging task. To solve above we propose...
Deep learning has become a mainstream method of hyperspectral image (HSI) classification. Many DL-based methods exploit spatial-spectral features to achieve better classification results. However, due the complex backgrounds in HSIs, existing usually show unsatisfactory performance for class pixels located on land-cover category boundary area. In large part, this is because network susceptible interference by irrelevant information around target pixel training stage, resulting inaccurate...
In the field of hyperspectral image (HSI) classification, deep learning has helped achieve great successes. However, most these achievements are made with very large amounts labeled training data. Manual annotation HSIs is labor intensive and time consuming. practical HSI there may only be a few samples available. To perform classification small number samples, new few-shot model based on adaptive subspaces featurewise transformation proposed in this article. First, we design 3-D local...
Brain-inspired algorithms have become a new trend in next-generation artificial intelligence. Through research on brain science, the intelligence of remote sensing can be effectively improved. This paper summarizes and analyzes essential properties cognise learning recent advance interpretation. Firstly, this introduces structural composition brain. Then, five represent brain-inspired are studied, including multiscale geometry analysis, compressed sensing, attention mechanism, reinforcement...
Seamless Internet of Things (IoT) connections expose many vulnerabilities in wireless networks, and IoT devices inevitably face malicious active attacks. automatic modulation recognition (AMR) is an effective way to combat physical layer threats. In the field noncollaborative communication, feature representation learning for unlabeled signals important task AMR. However, due unavailability a priori knowledge influence interference during signal transmission, intercepted are difficult...
With the development of in-vivo free-space fluorescence molecular imaging and multi-modality for small animals, there is a need new reconstruction methods real animal-shape models with large dataset. In this paper we are reporting novel hybrid adaptive finite element algorithm tomography reconstruction, based on linear scheme. Two different inversion strategies (Conjugate Gradient Landweber iterations) separately applied to first mesh level succeeding levels. The was validated by numerical...
In ultrasound elastography, tissue axial strains are calculated from the gradient of estimated displacements. However, common differentiation operation amplifies noises in displacement estimation, especially at high frequencies. this paper, a low-pass digital differentiator (LPDD) is proposed to calculate strain displacement. Several LPDDs that have been well developed field signal processing presented. The corresponding performances compared qualitatively and quantitatively computer...
The Bohai Sea area, offshore of the Bay Basin, is one most petroliferous regions in China, with proven original oil place approximately 2.4 × 109 m3 (150.94 108 bbl) and gas over 5 1012 (1.76 1013 ft3). Cumulative production 50 million tons (3.5 bbl). In this study, using limited data on source rock thickness, core samples, Rock-Eval pyrolysis along sedimentary facies analysis, characteristics different depositional settings were identified, richness, organic matter type, thermal evolution...
Convolutional neural networks (CNNs) have achieved remarkable performance in driver drowsiness detection based on the extraction of deep features drivers' faces. However, methods decreases sharply when complications, such as illumination changes cab, occlusions and shadows driver's face, variations head pose, occur. In addition, current are not capable distinguishing between states, talking versus yawning or blinking closing eyes. Therefore, technical challenges remain detection. this...
With the development of deep learning (DL) technology, feature extraction capability networks has gradually been enhanced, and high accuracy achieved in signal modulation classification (SMC) tasks. DL requires numerous training samples to achieve accuracy. However, non-cooperative cases, only a few labeled data are usually available. To solve this problem, we propose modular few-shot framework for SMC, called MsmcNet. MsmcNet comprises an IQ fusion (IQF) module, 1D processing (1D-SFP)...
Transformer has shown excellent performance in remote sensing field with long-range modeling capabilities. Remote video (RSV) moving object detection and tracking play indispensable roles military activities as well urban monitoring. However, transformers these fields are still at the exploratory stage. In this survey, we comprehensively summarize research prospects of RSV tracking. The core designs advanced first analyzed. It mainly includes attention mechanism evolution for specific tasks,...
With the advancement of Internet Things technology, need for sophisticated signal modulation classification has intensified, ensuring seamless communication and bolstering security among interconnected devices. In contemporary complex channel environment, difficult lies in dealing with a multitude schemes that exhibit subtle distinctions. Prior knowledge-guided deep learning methods have complementary strengths current context classification. To synthesize advantages these two methods, we...
Fluorescence molecular tomography (FMT) plays an important role in studying physiological and pathological processes of small animals vivo at level. However, this technique suffers from relatively low spatial resolution. To complement the problem, there has been a strong demand for providing functional morphological analysis same time. In paper, we proposed hybrid full-angle free-space FMT X-ray micro-cone-beam computed (CT) (micro-CBCT) prototype system, both anatomical images. During whole...
X-ray luminescence computed tomography (XLCT), which aims to achieve molecular and functional imaging by X-rays, has recently been proposed as a new modality. Combining the principles of excitation luminescence-based probes optical signal detection, XLCT naturally fuses anatomical images provides complementary information for wide range applications in biomedical research. In order improve data acquisition efficiency previously developed narrow-beam XLCT, cone beam (CB-XLCT) mode is adopted...
Hyperspectral imaging (HSI) classification has drawn tremendous attention in the field of Earth observation. In big data era, explosive growth occurred amount obtained by advanced remote sensors. Inevitably, new classes and refined categories appear continuously, such are limited terms timeliness application. These characteristics motivate us to build an HSI model that learns classifying capability rapidly within a few shots while maintaining good performance on original classes. To achieve...