A5035843722- Topic Modeling
- Target Tracking and Data Fusion in Sensor Networks
- Domain Adaptation and Few-Shot Learning
- Microgrid Control and Optimization
- Natural Language Processing Techniques
- Multimodal Machine Learning Applications
- Multilevel Inverters and Converters
- Power Systems and Renewable Energy
- Reinforcement Learning in Robotics
- Advanced Sensor and Control Systems
- Video Surveillance and Tracking Methods
- Orbital Angular Momentum in Optics
- Advanced Computational Techniques and Applications
- Advanced Graph Neural Networks
- Prosthetics and Rehabilitation Robotics
- Muscle activation and electromyography studies
- Industrial Technology and Control Systems
- Advanced Image and Video Retrieval Techniques
- Distributed Sensor Networks and Detection Algorithms
- Advanced DC-DC Converters
- Human Pose and Action Recognition
- COVID-19 diagnosis using AI
- Environmental and Agricultural Sciences
- Planetary Science and Exploration
- Granular flow and fluidized beds
Xinjiang Agricultural University
2011-2025
Shanghai Electric (China)
2022-2023
Yanbian University of Science and Technology
2023
Dataminr (United States)
2023
Nanjing University of Posts and Telecommunications
2022
Xuzhou Medical College
2022
Nantong University
2022
Harbin Institute of Technology
2007-2022
Southern University of Science and Technology
2020-2022
Qilu Hospital of Shandong University
2022
The Visual Object Tracking challenge 2015, VOT2015, aims at comparing short-term single-object visual trackers that do not apply pre-learned models of object appearance. Results 62 are presented. number tested makes VOT 2015 the largest benchmark on tracking to date. For each participating tracker, a short description is provided in appendix. Features VOT2015 go beyond its VOT2014 predecessor are: (i) new dataset twice as large with full annotation targets by rotated bounding boxes and...
Novel classes frequently arise in our dynamically changing world, e.g., new users the authentication system, and a machine learning model should recognize without forgetting old ones. This scenario becomes more challenging when class instances are insufficient, which is called few-shot class-incremental (FSCIL). Cur-rent methods handle incremental retrospectively by making updated similar to one. By contrast, we suggest prospectively prepare for future updates, propose ForwArd Compatible...
The Thermal Infrared Visual Object Tracking challenge 2015, VOT-TIR2015, aims at comparing short-term single-object visual trackers that work on thermal infrared (TIR) sequences and do not apply pre-learned models of object appearance. VOT-TIR2015 is the first benchmark tracking in TIR sequences. Results 24 are presented. For each participating tracker, a short description provided appendix. based VOT2013 challenge, but introduces following novelties: (i) newly collected LTIR (Link -- ping...
New classes arise frequently in our ever-changing world, e.g., emerging topics social media and new types of products e-commerce. A model should recognize meanwhile maintain discriminability over old classes. Under severe circumstances, only limited novel instances are available to incrementally update the model. The task recognizing few-shot without forgetting is called class-incremental learning (FSCIL). In this work, we propose a paradigm for FSCIL based on meta-learning by LearnIng...
Person re-identification (ReID) aims to match people across multiple non-overlapping video cameras deployed at different locations. To address this challenging problem, many metric learning approaches have been proposed, among which triplet loss is one of the state-of-the-arts. In work, we explore margin between positive and negative pairs triplets prove that large beneficial. particular, propose a novel multi-stage training strategy learns incremental improves effectively. Multiple levels...
Benchmark contamination has become a significant concern in the LLM evaluation community. Previous Agents-as-an-Evaluator address this issue by involving agents generation of questions. Despite their success, biases methods remain largely unexplored. In paper, we present theoretical formulation bias, providing valuable insights into designing unbiased protocols. Furthermore, identify two type bias through carefully designed probing tasks on minimal setup. To these issues, propose Unbiased...
Agricultural production faces critical challenges in arid regions due to global climate change and water scarcity. Exploring optimal nitrogen irrigation combinations is essential enhancing use efficiency crop yields. This study employs the logistic growth model analyze impact of varying treatments on summer maize southern Xinjiang. The goal identify an strategy enhance productivity, optimize use, ensure precise management. Field experiments included three levels (W1: 80% ETc, W2: 100% W3:...
In this letter, we present an underactuated soft exoskeleton which can sequentially provide knee extension and ankle plantarflexion assistance for each leg with only one motor. The aim of work is to assist the motions that have chronological moments at lower-limb joints minimize mass carried by wearer. consists a novel actuation mechanism: Continuum-Cable System (CCS), allows actuators transmit assistive torques joint when flexed fully extended. Design concepts component descriptions were...
This paper presents mathematical modeling procedure of three-phase grid-connected photovoltaic inverter. Presents synchronous PI current control strategy and the method for adjuster design. The typical waveforms grid voltage, harmonics are carried out on a 100kW inverter, which can provide some guidelines engineers to analyze, design implement.
In recent years, more and social media users have complained of information overload. this paper, we obtained data from surveys the multi-purpose Chinese website, WeChat, to develop an integrated model The was used determine antecedents consequences WeChat users' perceived A field survey 244 conducted test research hypotheses. results showed following key findings: 1) amount received length content increased perceptions overload, while number subscriptions followed not a significant factor;...
This paper addresses the problem of multi-sensor multi-target tracking. The main contribution is an efficient implementation δ-Generalized labeled Multi-Bernoulli (δ-GLMB) update. To truncate weighted sums exponentials, ranked assignment algorithm used in update to determine most important terms without computing all terms. Simulation experiments via linear Gaussian mixture models confirm effectiveness proposed algorithm.
In traditional practices of transportation system's constructions, traffic-related information is collected based on dedicated sensor networks, which are not only coverage-limited but also cost-consuming. With the enrichment concepts concerning "social sensors" and transportation", Sparse Mobile Crowdsensing (MCS) proposed to collect data from a few subareas by recruiting participants with portable devices infer in unsensed acceptable errors. However, real-world sensing campaigns, MCS...
This paper presents mathematical modeling procedure of three-phase grid-connected photovoltaic inverter. It synchronous PI current control strategy and the method for adjuster design. The typical waveforms grid voltage, harmonics are carried out on a 100 kW inverter, which can provide some guidelines engineers to analyze, design implement.
Fault ride-through specifications listed in modern transmission and distribution grid codes specify that photovoltaic system have to operate under faults. In the classic strategies used systems power delivered remains constant when a fault occurs. As consequence, should with increasing inverter current, which may result damage. Noticing PV array's special characteristics,this paper presents new strategy for ensures current levels remaining within designed limits. addition,a nonlinear...
This paper presents mathematical modeling procedure of three-phase grid-connected photovoltaic inverter. synchronous PI current control strategy and the method for adjuster design. The typical waveforms grid voltage, harmonics are carried out in a 100kW inverter, which can provide some guidelines engineers to analyze, design implement.
Step refractive index axicon for generating bottle beam is proposed the first time, and two kinds of models-refractive step-increase or decrease in radial direction are discussed. The principle generation analyzed by geometrical optics. optical field distribution 2D light spot image at different propagation distance behind axicons numerically simulated using diffraction integral theory. Research result shows that step-decrease generates single beam, periodic beams. Bottle has a wild...
In view of the uncertainty evaluation indicators for water resources carrying capacity theory set pair analysis is adopted. The samples are preliminarily classified by calculating connection degree between and indicators. Then further ranked through identity, difference antagonism analysis. weights factors in model obtained from avail value data reflecting information entropy, which weight allocation more reasonable. application result this to an example compared with those integrated...
Stair ascending is laborious, which imposes very high load on human knee, leading to risks of knee injuries. Therefore, in this paper, a biologically-inspired soft exosuit for extension assistance during stair ascent was designed and the assistant effect proposed evaluated. We placed Bowden cables front thigh shank mimic quadriceps muscle, generating moments by length contraction. A force-based position control implemented deliver moment profile, based biological normal ascent. Biomechanical...
Objectives: This study aimed to identify novel biomarkers for osteoarthritis (OA) and explore potential pathological immune cell infiltration. Methods: We identified differentially expressed genes (DEGs) between OA normal synovial tissues using the limma package in R, performed enrichment analyses understand functions enriched pathways of DEGs. Weighted gene co-expression network analysis (WGCNA) distinct machine-learning algorithms were then used hub modules candidate biomarkers. assessed...
Preclinical trials of immunotherapy in ovarian cancer (OC) have shown promising results. This makes it meaningful to prospectively examine the biological mechanisms explaining differences response performances among OC patients. Open-accessed data was obtained from Cancer Genome Atlas and Gene Expression Omnibus database. All analysis conducted using R software. We firstly performed TIDE evaluate rate The machine learning algorithm LASSO logistic regression SVM-RFE were used identify...
Humans have a remarkable ability to learn continuously from th e environment and inner experience. One of the grand goals robots is build an artificial "lifelong learning" agent that can shape cultivated understanding world current scene previous knowledge via autonomous lifelong development. It challenging for robot learning process retain earlier when encounter new tasks or information. Recent advances in computer vision deep -learning methods been impressive due large-scale data sets,...