A5051897815- Video Coding and Compression Technologies
- Advanced Vision and Imaging
- Traffic and Road Safety
- Image and Video Quality Assessment
- Advanced Image Processing Techniques
- Traffic control and management
- Cancer-related molecular mechanisms research
- Acupuncture Treatment Research Studies
- Face and Expression Recognition
- Photonic and Optical Devices
- Meat and Animal Product Quality
- Biochemical Analysis and Sensing Techniques
- Human Pose and Action Recognition
- Mechanical and Optical Resonators
- Diet and metabolism studies
- Genetic Syndromes and Imprinting
- Pituitary Gland Disorders and Treatments
- Advanced MEMS and NEMS Technologies
- Growth Hormone and Insulin-like Growth Factors
- Bipolar Disorder and Treatment
- Transportation Planning and Optimization
- Human-Automation Interaction and Safety
- Advanced Data Compression Techniques
- Nanoplatforms for cancer theranostics
- Video Surveillance and Tracking Methods
Nanjing University of Posts and Telecommunications
2017-2025
Chengdu University of Traditional Chinese Medicine
2021-2025
Anhui Xinhua University
2025
Guangzhou Xinhua University
2025
University of Malaya
2025
Hubei Normal University
2024
Jiangxi Mental Health Center
2016-2024
Nanchang University
2016-2024
Weatherford College
2020-2024
Huashan Hospital
2022-2024
The extensive computational burden limits the usage of CNNs in mobile devices for dense estimation tasks. In this paper, we present a lightweight network to address problem, namely LEDNet, which employs an asymmetric encoder-decoder architecture task real-time semantic segmentation. More specifically, encoder adopts ResNet as backbone network, where two new operations, channel split and shuffle, are utilized each residual block greatly reduce computation cost while maintaining higher...
The newly developed HEVC video coding standard can achieve higher compression performance than the previous standards, such as MPEG-4, H.263 and H.264/AVC. However, HEVC's high computational complexity raises concerns about burden on real-time application. In this paper, a fast pyramid motion divergence (PMD) based CU selection algorithm is presented for inter prediction. PMD features are calculated with estimated optical flow of downsampled frames. Theoretical analysis shows that be used to...
In the downlink transmission scenario, power allocation and beamforming design at transmitter are essential when using multiple antenna arrays. This paper considers a input–multiple output broadcast channel to maximize weighted sum-rate under total constraint. The classical minimum mean-square error (WMMSE) algorithm can obtain suboptimal solutions but involves high computational complexity. To reduce this complexity, we propose fast method unsupervised learning, which trains deep neural...
Spectral imaging paves the way for various fields, particularly in biomedical research. However, spectral imaging, mainly depending on spatial or temporal scanning, cannot achieve high temporal, spatial, and resolution simultaneously. In this study, we demonstrated a silicon real-time ultraspectral chip based reconfigurable metasurfaces, comprising 155,216 ( <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> <mml:mrow class="MJX-TeXAtom-ORD"> <mml:mn>356</mml:mn>...
Abstract Metasurfaces have an exceptional capacity to manipulate the phase, amplitude, polarization, or spectrum of light. However, unit cells, meta‐atoms, metasurfaces are conventionally designed using regular shapes, limiting performance improvement. The utilization with freeform shaped meta‐atoms for on‐chip ultraspectral imaging is proposed, where patterns generated controllable feature sizes and boundary curvatures feasible fabrication. These broaden design diversity enrich...
Human emotions recognition (HERO) is considered as one of the important techniques for realizing intelligent Internet Things. The demand a robust and precise facial expression algorithm urgent HERO. In this paper, we propose deep based on ensemble learning model. proposed consists three sub-networks with different depths. Each sub-network comprised convolutional neutral networks trained independently. more layers recognizes by extracting local details such features eyes mouth, while less...
Yeast extract can impart thickness, complexity and long-lasting taste impression, coined kokumi taste, to blank chicken broth. In this research, the kokumi-active peptide in yeast was discovered by ultrafiltration, liquid chromatographic quadrupole-time-of-flight-tandem mass spectrometric technologies. Furthermore, sensory characters of these peptides were evaluated a sensomics approach.A total 10 identified from extract. They γ-Glu-Cys-Gly, γ-Glu-Leu, γ-Glu-Val, γ-Glu-Tyr, Leu-Lys, Leu-Gln,...
Traditional surveillance systems for observing water levels are often complex, costly, and time-consuming. In this paper, we developed a low-cost unmanned system consisting of remote measuring stations monitoring center. The uses map-based Web service, as well video cameras, level analyzers, wireless communication routers necessary to display real-time measurements rivers reservoirs on platform. With the aid router, information is transmitted server connected Internet via cellular network....
The emerging high efficiency video coding (HEVC) standard has improved compression performance significantly in comparison with H.264/AVC. However, more intensive computational complexity been introduced by adopting a number of new tools. In this paper, fast inter CU decision is proposed based on the latent sum absolute differences (SAD) estimation. Firstly, two-layer motion estimation (ME) method designed to take advantage SAD cost. ME can obtain costs for both upper and its sub-CUs....
This paper investigates deep learning for risk detection and trajectory tracking at construction sites. Typically, safety officers are responsible inspecting verifying site due to many potential risks. Traditional target algorithms depend heavily on hand-crafted features. However, these features difficult design, accuracy is poor. To solve problems, this proposes a deep-learning-based algorithm that uses pedestrian wearable devices (e.g., helmets colored vests) identify pedestrians. We train...
China has experienced unprecedented economic growth and changes in urban form the past decades. Increased urbanisation motorisation puts pedestrians automobiles at greater conflict. Because of China's long blocks (superblocks), many conflicts occur mid-block informal or illegal crossings. This study focuses on factors influencing crossing gap acceptance. We remotely observed a six-lane arterial Kunming, China, tracking 522 accepted gaps 152 rejected two-stage (roadside to median roadside)....
The newly developed High Efficiency Video Coding (HEVC) Standard has improved video coding performance significantly in comparison to its predecessors. However, more intensive computation complexity is introduced by implementing a number of new tools. In this paper, fast unit (CU) decision based on Markov random field (MRF) proposed for HEVC inter frames. First, it observed that the variance absolute difference (VAD) proportional with rate-distortion (R-D) cost. VAD feature designed CU...
With the rapid development in smart vehicles, security and privacy issues of Vehicular Ad-hoc Network (VANET) have drawn significant attention. Devices an On-Board Unit (OBU) access to internet through Communication Module (VCM), hence a real-time accurate intrusion detection method is favored be applied VCM. In this paper, we present Deep Learning (DL) based end-to-end automatically detect malware traffic for OBUs. Different from previous methods, our proposed only requires raw instead...
Abstract Non‐small‐cell lung cancer (NSCLC) is the most common malignancy along with high mortality rate worldwide. Recently, nucleolar and spindle‐associated protein 1 (NUSAP1) has been reported to be involved in malignant progression of several cancers. However, NSCLC, biological function NUSAP1 its molecular mechanism have not reported. Here, our findings indicated that messenger RNA expression level was remarkably upregulated NSCLC tissues compared adjacent normal tissues. We also found...