A5100386965- Impact of Light on Environment and Health
- Remote Sensing and Land Use
- Engineering Applied Research
- Advanced Image Fusion Techniques
- Remote-Sensing Image Classification
- Aerodynamics and Fluid Dynamics Research
- Land Use and Ecosystem Services
- Geochemistry and Geologic Mapping
- Gold and Silver Nanoparticles Synthesis and Applications
- Cooperative Communication and Network Coding
- Remote Sensing in Agriculture
- Advanced Wireless Network Optimization
- Wireless Networks and Protocols
- Robotic Path Planning Algorithms
- Mobile Ad Hoc Networks
- Infrared Target Detection Methodologies
- Data Management and Algorithms
- Air Quality and Health Impacts
- Advanced Numerical Analysis Techniques
- Innovation in Digital Healthcare Systems
- Nanoplatforms for cancer theranostics
- Wind and Air Flow Studies
- Context-Aware Activity Recognition Systems
- Computer Graphics and Visualization Techniques
- Urban Heat Island Mitigation
Seoul National University of Science and Technology
2023-2025
Korea Development Institute
2019-2024
University of California, Berkeley
2023
Seoul National University
2013-2022
Samsung (South Korea)
2015-2022
Dongguk University
2009-2021
Sejong University
2019-2021
Baylor University
2021
Armed Forces Capital Hospital
2021
Hanyang University
2003-2020
Hot electron chemistry has drawn tremendous attention from applications related to materials, energy, sensing, and catalysis. The plasmon‐induced generation of hot electrons their transfer behavior are very important for understanding plasmonic‐enhanced achieving practically useful efficiency. From a plasmonic perspective, well‐designed structures that can manipulate surface plasmons able enhance the efficiencies electron‐based processes. This progress report summarizes recent experimental...
This study used geographic object-based image analysis (GEOBIA) with very high spatial resolution (VHR) aerial imagery (0.3 m resolution) to classify vegetation, channel and bare mud classes in a salt marsh. Three classification issues were investigated the context of segmentation scale: (1) comparison single- multi-scale GEOBIA using spectral bands, (2) relative benefit incorporating texture derived from grey-level co-occurrence matrix (GLCM) classifying marsh features (3) effect...
This study investigated the use of a geographic object-based image analysis (GEOBIA) approach with incorporation object-specific grey-level co-occurrence matrix (GLCM) texture measures from multispectral Ikonos for delineation deciduous, evergreen, and mixed forest types in Guilford Courthouse National Military Park, North Carolina. A series automated segmentations was produced at range scales, each resulting an associated number size objects (or segments). Prior to classification, spatial...
For biomedical applications, the NIR-II window provides several advantages over conventional NIR-I window, including deeper penetration depth, low autofluorescence, and higher value of maximum permissible exposure to laser power. An overview recently reported NIR-II-window-responsive plasmonic gold nanostructures is presented, opportunities, challenges, future directions for these in research fields are discussed. Colloidal nanoparticles (PNPs) have attracted enormous attention because their...
Uniformly controlling a large number of metal nanostructures with plasmonically enhanced signal to generate quantitative optical signals and the widespread use these structures for surface-enhanced Raman scattering (SERS)-based biosensing bioimaging applications are paramount importance but extremely challenging. Here, we report highly controllable, facile selective-interdiffusive dealloying chemistry synthesizing dealloyed intra-nanogap particles (DIPs) ∼2 nm intragap in high yield (∼95%)...
We describe a fast VLSI architecture for full-search motion estimation the blocks with 7 different sizes in MPEG-4 AVC/H.264. The proposed variable block size (VBSME) consists of 16x16 PE array, an adder tree and comparators to find all 41 vectors their minimum SADs 16x16, 16x8, 8x16, 8x8, 8x4, 4x8 4x4. It employs 2-D datapath its control search area data is simple regular. VBSME can achieve 100% utilization by employing preload register buffer inside each allow real-time processing...
Connected vehicles are at risk of exposing their location history when using potentially untrusted location-based services (LBSs) in the driving process. We propose a method called mutually obfuscating paths (MOP) that enables to provide highly accurate realtime updates LBS while preventing from tracking vehicles. The instrument is leverage connected vehicles' two network interfaces: in-car LTE Internet (for accessing LBS) and car-to-car Dedicated short-range communications...
Porous nanostructures exhibit excellent catalytic properties due to high surface-to-volume ratio, good surface reactivity and various structural features, but controlling the distribution, size, shape density of pores features these particles is highly challenging. Herein, we report a tunable dealloying-based facile synthetic strategy form porous Au core/porous shell nanoparticles (CPS NPs) in yield by selectively dissolving Ag atoms from Au/Au-Ag core/alloy NPs. The CPS NPs very short...
Surface-enhanced Raman scattering (SERS)-based sensing is promising in that it has potential to allow for highly sensitive, selective, and multiplexed detection imaging. However, the controlled assembly gap formation between plasmonic particles generating strong SERS signals a quantitative manner challenging, especially on biodetection platforms, particle-to-particle variation signal enhancement can vary by several orders of magnitude single batch, largely limiting reliable use practical...
Local Climate Zones (LCZ) offer a climate-aware and standardized classification scheme composed of 17 urban natural landscape classes. Recent deep learning-based LCZ studies have adopted scene approach with computer vision-inspired models. In light these advancements, this study introduces multi-scale, multi-level attention network (MSMLA-Net) for classification. MSMLA-Net integrates multi-scale (MS) module to generate features from the input data novel (MLA) as branch unit model's main...
As the elderly people living alone are enormously increasing recently, we need system inferring activities of daily (ADL) for maintaining healthy life and recognizing emergency. The should be constructed with sensors, which used to associate people's while remaining as non intrusive views possible. To do this, proposed use a triaxial accelerometer sensor environment sensors indicating contact subject in home. Particularly, order robustly infer ADLs, present component ADL, is decided...
Vegetation mapping was performed using geographic object-based image analysis (GEOS1A) and very high spatial resolution (VHR) imagery for two study areas in Great Smoky Mountains National Park. This investigated how accurately GEOB1A with ancillary data emulates manual interpretation rugged mountain multi-level vegetation classes of the Classification System (NVCS). It discovered that incorporation texture topographic variables spectral from scanned color infrared aerial photographs...
The demand for wireless local area networks (WLANs) has drastically increased due to the prevalence of mobile devices, such as smartphones and tablet personal computers. In an environment, a large number basic service sets (BSSs) can overlap with each other, causing well-known hidden-node problem occur more frequently. Solutions alleviate problem, request-to-send/clear-to-send (RTS/CTS) exchange, often contain protocol overhead that may lead throughput degradation when there is no hidden...
In this work, we use recurrent neural network (RNN) to predict the medical examination data with missing parts. There often exist parts in due various human factors, for instance, because subjects occasionally miss their annual examinations. Such make it hard future by machines. Thus, imputation of information is needed accurate prediction data. Among types RNNs, choose simple (SRN) and long short-term memory (LSTM) as well data, they show good performance many relevant applications. our...
This study examines best image fusion approaches for generating pansharpened very high resolution (VHR) multispectral images to be utilized monitoring coastal barrier island development. Selected techniques assessed in this research come from the three categories of spectral substitution (e.g., Brovey transform and multiplicative merging), arithmetic merging modified intensity-hue-saturation principal component analysis), spatial domain high-pass filter, subtractive merge). The methods...
The recent economic crisis in United States has led to an increase home foreclosures and subsequent abandonments. A by-product of this trend been associated rise the number neglected swimming pools, which provide new habitats for larval stages Culex mosquito vectors West Nile Virus (WNV) urban suburban environments. WNV major concern related pools California. Our research focused on using very high spatial resolution (VHR) satellite imagery processing techniques, including image...
We compared global-scale pansharpening of full- or large-scene QuickBird satellite imagery with local-scale a subset these scenes by using modified intensity-hue-saturation (MIHS), principal component analysis (PCA), multiplicative (MP), and high pass filter ( HPF ) methods. The spectral properties all pansharpened images were evaluated quantitative indices such as erreur relative globale adimensionnelle de synthese (ERGAS), correlation coefficient (CC), difference to mean (RDM), standard...
In this study, we investigated the use of composite patches to repair aluminum alloy double-lap joints. The carbon were cured using induction curing or oven curing. Joints repaired precured cocured patches. bond strengths different joints compared. We also whether incorporation nanotubes (CNTs) in adhesive bondline affected strength. found that induction-cured samples exhibited similar those corresponding oven-cured samples; was true for both baseline and CNT-reinforced samples. Further,...
We describe a fast VLSI architecture for full-search motion estimation the blocks with 7 different sizes in MPEG-4 AVC/H.264. The proposed variable block size (VBSME) consists of 16/spl times/16 PE array, an adder tree and comparators to find all 41 vectors their minimum SADs times/16, times/8, 8/spl times/4, 4/spl times/8 times/4. It employs 2D datapath its control search area data is simple regular. VBSME can achieve 100% utilization by employing preload register buffer inside each allow...