A5103083984- Advanced Neural Network Applications
- Water Treatment and Disinfection
- Non-Invasive Vital Sign Monitoring
- Remote Sensing and Land Use
- Video Surveillance and Tracking Methods
- Autonomous Vehicle Technology and Safety
- Healthcare Technology and Patient Monitoring
- Sparse and Compressive Sensing Techniques
- Face and Expression Recognition
- Advanced Numerical Analysis Techniques
- Industrial Technology and Control Systems
- Simulation and Modeling Applications
- Wastewater Treatment and Reuse
- Speech and Audio Processing
- Hydrogen's biological and therapeutic effects
- Reinforcement Learning in Robotics
- Urban Heat Island Mitigation
- Urban Agriculture and Sustainability
- Antenna Design and Analysis
- Text and Document Classification Technologies
- Full-Duplex Wireless Communications
- Advanced oxidation water treatment
- Gaze Tracking and Assistive Technology
- Manufacturing Process and Optimization
- Machine Learning and ELM
Tsinghua University
2009-2023
University Town of Shenzhen
2022-2023
China Academy of Space Technology
2019
Wuhan University of Technology
2018
Beijing University of Posts and Telecommunications
2014-2017
Hunan University of Technology
2016
Shenzhen Institutes of Advanced Technology
2014
Harbin Institute of Technology
2013-2014
University of Hong Kong
2012
Hong Kong University of Science and Technology
2012
Spatial multiple access holds the promise to boost capacity of wireless networks when an point has antennas. Due asynchronous and uncontrolled nature LANs, conventional MIMO technology does not work efficiently concurrent transmissions from stations are uncoordinated. In this paper, we present design implementation a crosslayer system, called SAM, that addresses challenges enabling spatial for devices in random network like WLAN. SAM uses chain-decoding technique reliably recover channel...
Metropolitan areas increasingly confront complex challenges related to food security, social inequality, environmental degradation, and resource scarcity, exacerbated by rapid urbanization, climate change, the reliance on extended, fragile supply chains. Urban peri-urban agriculture (UPA) is recognized as a promising approach mitigate these issues. For example, it enhances security nutrition strengthening local systems, improves livelihoods providing employment income for residents, promotes...
Understanding whole-brain circuitry development in mice has been hindered by the lack of tools for accurate and comprehensive analysis. No existing 3D brain atlas provides daily postnatal resolution, as constructing such atlases is highly labour-intensive. Axonal morphologies change dynamically, making reliable segmentation challenging, many 2D datasets sufficient Z-resolution cross-modality Here, we present D-LMBmapX, a deep-learning pipeline automated profiling across development....
For most sparse coding methods, data samples are first encoded as hand-crafted features, followed by another separate learning step that generates dictionary and codes. However, such feature representations may not be optimally compatible with the process, thus producing suboptimal results. In this paper, we propose a new architecture for nonlinear coding, in which mapped into codes via carefully designed stacked auto-encoder (SAE) networks. We jointly learn low-dimensional embedding of...
We propose a deep fine-grained multi-level fusion architecture for monocular 3D object detection, with an additionally designed anti-occlusion optimization process. Conventional detection methods usually leverage geometry constraints such as keypoints, shape relationships, and to 2D optimizations offset the lack of accurate depth information. However, these still struggle against directly extracting rich information from estimation. To solve problem, we integrate features pseudo-LiDAR filter...
Cameras embedded in consumer devices have previously been used as physiological information sensors. The waveform of the photoplethysmographic image (PPGi) signals may be significantly affected by light spectra and intensity. purpose this paper is to evaluate performance PPGi acquisition red, green, blue channels using a commercial camera different conditions. system, developed for paper, comprises sources with varied intensities. Signals were acquired from fingertips 12 healthy subjects....
This paper presents first several steps towards an FPGA-based electronic system for remote pulse rate (PR) measurement. The uses a low-cost digital camera as image sensor, which operates at up to 30 frames per second (fps) in WXGA (1200×800 pixels) resolution. A novel algorithm PR measurement was implemented using FPGA development board. commercially-available photoplethysmography module (TP-TSD200A from BIOPAC) used golden standard verify the performance of suggested system. Ten male...
Effect of satellite pointing accuracy on spaceborne video SAR imaging quality has been studied in this paper to solve the problem indicators decomposition. First, effect model error single frame image was established based rough and precise imaging. Then relationship between stability Video analysised. Finally, simulations prove validity analysis. The analysis simulation results show that attitude mainly affects resolution, sidelobe ratio, ambiguity radiation performance. For mode, shift...
The method of extracting characteristic parameters lip according to template was presented. A dynamic clustering algorithm classify the lip-shape based on criteria least square error sum proposed, and state sequence that described movement obtained by improved ant colony algorithm. lip-reading pattern recognition performed DTW Experiments show has satisfactory result in classifying isolated syllables.
There are many different algorithms for solving complex optimization problems. Each algorithm has been applied successfully in some problems, but not efficiently other In this paper the Cauchy mutation and multi-parent hybrid operator combined to propose a evolutionary based on communication (Mixed Evolutionary Algorithm Communication), hereinafter referred as CMEA. The basic idea of CMEA is that initial population divided into two subpopulations. operators multiple paternal crossover used...
For non-sample-spaced multipath channels, multi-path energy leakage leads to an increase in the channel sparsity and detection difficulties. In this paper, we propose adaptive matching pursuit (SAMP) algorithm for estimation of channels. Compared with other greedy algorithms, most innovative feature SAMP is its capability signal reconstruction without prior information sparsity. To further improve quality, a regularized backtracking step which can flexibly remove inappropriate atoms adapted...
Multi-class object detection always involves the tasks of accurate target localization which is mainly related to bounding box regression. Smooth L1 loss most popular regression used in current state-of-the-art systems. However, such for regressing parameters a can't accurately and consistently regress associated ground truth well. We instead propose near-duplicated loss, that better evaluate disparity between consistently. present an approximate algorithm with kernel function not only...
User interest model is the key technology for us to efficiently obtain personalized high quality resources in huge amounts of data. On basis traditional user model, combining with concept ontology, this paper designs a system construct based on ontology semantic tree and space vector its eigenvalue changing by using Ebbinghaus's theory forgetting curve. The method describes theme tree, it not only simply describe relations between concepts, but also can reflect drift character. It more...
Driver’s lane changing maneuver prediction is important to improve the function of Advanced Driver Assistance System (ADAS) and maintain driving safety. In this paper, we model driver’s gaze attention by using spatio-temporal features propose a high-precision with multi-source information as inputs. Firstly, naturalistic dataset fifty-two drivers established on real-vehicle apparatus. The contains three sources information: ego vehicle, traffic environment, eye movement. Secondly,...