A5014397444- Heat Transfer and Boiling Studies
- Heat Transfer and Optimization
- Heat Transfer Mechanisms
- Statistical Methods and Inference
- Technology and Data Analysis
- Advanced Clustering Algorithms Research
- Advanced Causal Inference Techniques
- Face and Expression Recognition
- Advanced Combustion Engine Technologies
- Statistical Methods in Clinical Trials
- Heat transfer and supercritical fluids
- Sparse and Compressive Sensing Techniques
- Neural Networks and Applications
- Image and Signal Denoising Methods
- Robotic Path Planning Algorithms
- Combustion and flame dynamics
- Gene expression and cancer classification
- Hand Gesture Recognition Systems
- Gaze Tracking and Assistive Technology
- Advanced Vision and Imaging
- Robotics and Sensor-Based Localization
- AI in cancer detection
- Advanced Image Processing Techniques
- IoT and Edge/Fog Computing
- Cloud Computing and Resource Management
University of Toronto
2024
Zhejiang University of Science and Technology
2023
Hangzhou Dianzi University
2022
China Aerodynamics Research and Development Center
2022
China Telecom (China)
2022
China Telecom
2022
Zhejiang University
2015-2021
University of North Carolina at Chapel Hill
2016-2020
Qingdao University
2020
Seattle University
2020
This paper reviews the NTIRE 2023 challenge on image denoising (σ = 50) with a focus proposed solutions and results. The aim is to obtain network design capable produce high-quality results best performance measured by PSNR for denoising. Independent additive white Gaussian noise (AWGN) assumed level 50. had 225 registered participants, 16 teams made valid submissions. They gauge state-of-the-art
This paper summarizes the 2nd NTIRE challenge on stereo image super-resolution (SR) with a focus new solutions and results. The task of is to super-resolve low-resolution pair high-resolution one magnification factor ×4. Compared single SR, major this lies in how exploit additional information another viewpoint maintain consistency has 3 tracks, including track distortion (e.g., PSNR) bicubic degradation, perceptual quality LPIPS) as well real degradations. In total, 175, 93, 103...
Summary Precision medicine is an emerging scientific topic for disease treatment and prevention that takes into account individual patient characteristics. It important direction clinical research, many statistical methods have been proposed recently. One of the primary goals precision to obtain optimal rule (ITR), which can help make decisions on selection according each patient's specific Recently, outcome weighted learning (OWL) has estimate such ITR in a binary setting by maximizing...
Due to heterogeneity for many chronic diseases, precise personalized medicine, also known as precision has drawn increasing attentions in the scientific community. One main goal of medicine is develop most effective tailored therapy each individual patient. To that end, one needs incorporate characteristics detect a proper treatment rule (ITR), by which suitable decisions on assignments can be made optimize patients' clinical outcome. For binary settings, outcome weighted learning (OWL) and...
Precision medicine is an important area of research with the goal identifying optimal treatment for each individual patient. In literature, various methods are proposed to divide population into subgroups according heterogeneous effects individuals. this paper, a new exploratory machine learning tool, named latent supervised clustering, identify subpopulations. particular, we formulate problem as regression subject specific coefficients, and use adaptive fusion cluster coefficients This...
Decision tree algorithms are very popular in the field of data mining. This paper proposes a distributed decision algorithm and shows examples its implementation on big platforms. The major contribution this is novel KS-Tree which builds environment. applied to some real world mining problems compared with state-of-the-art techniques that implemented R Apache Spark. results show can achieve better results, especially large sets. Furthermore, we demonstrate be various tasks, such as variable...
Abstract When describing an image, people can rapidly extract the topic from image and find main object, generating sentences that match idea of image. However, most scene graph generation methods do not emphasise importance Consequently, captions generated by graph‐based captioning models cannot reflect in then expressing central In this paper, we propose a method for based on graphs (TSG). Firstly, structure express images' topics relationships between objects. Then, combined with graph,...
In this paper, by using the combination of Leap Motion and Myo armband, two methods for hand tracking online gesture identification are proposed. With proposed methods, We have improved measurement accuracy palm direction solved problem insufficient when is at limit range. use Kalman filter algorithm neural network classification method to process data measured Myo, so that operator's more accurate robust even positions close one single sensor. The which improve accuracy, can be used robotic...
Learning in the Reproducing Kernel Hilbert Space (RKHS) has been widely used many scientific disciplines. Because a RKHS can be very flexible, it is common to impose regularization term optimization prevent overfitting. Standard learning employs squared norm penalty of function. Despite its success, challenges remain. In particular, one cannot directly use for variable selection or data extraction. Therefore, when there exists noise predictors, underlying function sparse representation dual...
Paper-based friction material is a typical paper-based composite that usually cured via hot-pressing. This curing method does not account for the effect of pressure on matrix resin, resulting in uneven distribution resin and reducing mechanical properties materials. To overcome above shortcomings, pre-curing was introduced before hot-pressing, effects different degrees surface morphology materials were studied. The degree significantly affected interfacial bonding strength material. When at...
This paper reports on the mode-locked operation of a 980-nm external-cavity passively laser with extremely narrow RF linewidth. Optical pulses 10-ps pulse duration were generated at repetition rate 955 MHz, an average output power 39.3 mW – which corresponds to peak 4.1 W, directly from oscillator. The spectrum displays -3dB linewidth only ~40 Hz, as well 60-dB dynamic contrast, revealing exceptionally low-noise fundamental this laser. At ~1 GHz, highest 5.26 W was achieved, albeit increased...
In drone racing, the time-minimum trajectory is affected by drone's capabilities, layout of race track, and configurations gates (e.g., their shapes sizes). However, previous studies neglect configuration gates, simply rendering racing a waypoint-passing task. This formulation often leads to conservative choice paths through as spatial potential not fully utilized. To address this issue, we present time-optimal planner that can faithfully model gate constraints with various thereby generate...
Time-critical tasks such as drone racing typically cover large operation areas. However, it is difficult and computationally intensive for current time-optimal motion planners to accommodate long flight distances since a yet unknown number of knot points required represent the trajectory. We present polynomial-based automatic optimal synthesis (AOS) approach that can address this challenge. Our method not only achieves superior time optimality but also maintains consistently low...
Learning with Reproducing Kernel Hilbert Spaces (RKHS) has been widely used in many scientific disciplines. Because a RKHS can be very flexible, it is common to impose regularization term the optimization prevent overfitting. Standard learning employs squared norm penalty of function. Despite its success, challenges remain. In particular, one cannot directly use for variable selection or data extraction. Therefore, when there exists noise predictors, underlying function sparse representation...
An experimental study on local flow boiling heat transfer performance in two longitudinal dimple-grooved tubes and an equivalent smooth tube was performed. All three test cooper have the same inner diameter of 11.5 mm; working fluid is near-azeotropic mixture, R410A; all runs are conducted a 2 m long horizontal tube-in-tube exchanger. Constant evaporation temperature at 10 °C maintained when flux ranging from 32 kW/m2 to 37 refrigerant quality varied inlet 0.1 outlet 0.9 mass 150 kg/(m2 s)....
Precision medicine is an emerging scientific topic for disease treatment and prevention that takes into account individual patient characteristics. It important direction clinical research, many statistical methods have been recently proposed. One of the primary goals precision to obtain optimal rule (ITR), which can help make decisions on selection according each patient's specific Recently, outcome weighted learning (OWL) has proposed estimate such ITR in a binary setting by maximizing...
The main purpose of our research is to apply the Arduino platform design an anti-theft system that incorporates GPS satellite positioning, and use sensors control peripheral devices in car reduce risk theft. tracking allows user's mobile phone locate current position on Google map immediately, reed switch door device determine whether there improper external force open door, activates front windshield be completely black.
As the enterprise IT systems continuously migrate to cloud, conventional waterfall development process model undermines system flexibility and brings some unacceptable disadvantages, such as slow delivery of business requirements, long release cycles high risk manual deployment. However, adopting DevOps methodology can solve these problems it helps develop deliver software products efficiently. Based on widely adopted processes toolchains, we have developed a novel ultra-large-scale...
With the physical property changing dramatically, supercritical aviation kerosene obtains unique heat transfer characteristics. In this way, it is difficult to investigate characteristics by normal experiment and therefore we resort numerical analysis address scientific questions in study. The project proposed demystify of with CFD 4mm inside diameter vertical circular tubes. Under conditions different pressures (3.5MPa-5MPa), properties fluid are expressed linear poly-nominal fitting...