A5036387111- Video Surveillance and Tracking Methods
- Infrared Target Detection Methodologies
- Advanced Vision and Imaging
- Robotics and Sensor-Based Localization
- Gaze Tracking and Assistive Technology
- Human Pose and Action Recognition
- Advanced Image and Video Retrieval Techniques
- Hand Gesture Recognition Systems
- Advanced Neural Network Applications
- Advanced Measurement and Detection Methods
- Impact of Light on Environment and Health
- Image and Object Detection Techniques
- Advanced Computing and Algorithms
- 3D Surveying and Cultural Heritage
- Face recognition and analysis
- Industrial Vision Systems and Defect Detection
- Image Enhancement Techniques
- Face and Expression Recognition
- Catalytic Processes in Materials Science
- Target Tracking and Data Fusion in Sensor Networks
- Fire Detection and Safety Systems
- Soft Robotics and Applications
- Power Line Inspection Robots
- Glaucoma and retinal disorders
- Advanced Measurement and Metrology Techniques
East China University of Science and Technology
2020-2024
University of Science and Technology of China
2024
Beihua University
2021-2024
University of Electronic Science and Technology of China
2024
Quzhou University
2019-2024
Qingdao Agricultural University
2022-2024
Inner Mongolia University of Technology
2024
Southwest Jiaotong University
2024
Anhui University
2024
Second Hospital of Liaohe Oilfield
2023
Deep learning has been proved effective in multiple object tracking, which confronts the difficulties of frequent occlusions, confusing appearance, in‐and‐out objects, and lack enough labelled data. Recently, deep based multi‐object tracking methods make a rapid progress from representation to network modelling due development theory benchmark setup. In this study, authors summarise analyse are top‐ranked public test. First, they investigate functionality networks these methods, classify...
Grain size and filling are two key determinants of grain thousand-kernel weight (TKW) crop yield, therefore they have undergone strong selection since cereal was domesticated. Genetic dissection the traits will improve yield potential in crops. A quantitative trait locus significantly associated with wheat TKW detected on chromosome 7AS flanked by a simple sequence repeat marker Wmc17 Chinese 262 mini-core collection genome-wide association study. Combined bulked segregant RNA-sequencing...
Abstract Nasopharyngeal carcinoma (NPC) is a common head and neck malignant with high incidence in Southern China. Genetic aberrations play vital role the pathogenesis, progression prognosis of NPC. In present study, we elucidated underlying mechanism FAS-AS1 its genetic variation rs6586163 We demonstrated that variant genotype carriers were associated lower risk NPC (CC vs. AA, OR = 0.645, P 0.006) better overall survival (AC + CC HR 0.667, 0.030). Mechanically, increased transcriptional...
19©2019IEEER obot-assisted therapy (RAT) offers potential advantages for improving the social skills of children with autism spectrum disorders (ASDs).This article provides an overview developed technology and clinical results EC-FP7funded Development Robot-Enhanced AutisM (DREAM) project, which aims to develop next level RAT in both technological perspectives, commonly referred as robot-enhanced (RET).Within this a supervised autonomous robotic system is collaboratively by interdisciplinary...
Recently, siamese-based trackers have achieved significant successes. However, those are restricted by the difficulty of learning consistent feature representation with object. To address above challenge, this paper proposes a novel siamese implicit region proposal network compound attention for visual tracking. First, an (IRP) module is designed combining pixel-wise correlation method. This can aggregate information different regions that similar to pre-defined anchor boxes in Region...
Selecting reliable beacon nodes plays a significant role in relocalizing unknown wireless sensor network. When the position of node is drifted or spoofed, it becomes an unreliable node, which would lead to large relocalization deviation its neighbor. However, when selecting nodes, most algorithms only screen either drifting malicious whose spoofed. This article proposes algorithm that can simultaneously and nodes. The divided into four steps. First, three indicators are introduced, where two...
The need to mitigate greenhouse effect and decrease environmental pollution has become an increasingly essential driving force for carbon dioxide (CO2) capture from natural gas or flue gas, especially under...
In this paper, we present a novel appearance model using sparse representation and online dictionary learning techniques for visual tracking. our approach, the is represented by representation, strategy used to adapt variations during We unify defining sparsity consistency constraint that facilitates generative discriminative capabilities of model. An elastic-net enforced stage capture characteristics local appearances are insensitive partial occlusions. Hence, target effectively recovered...
In the manufacturing process of industrial robots, defect detection raw materials includes two types tasks, which makes guarantee its accuracy. It also task challenging in practical work. analyzing disadvantages existing methods, such as low precision and generalization ability, a method on basis attention mechanism dilated convolution module is proposed. order to effectively extract features, two-stage framework chosen by applying Resnet50 pre-training network our model. With this...
In this paper, we present an effective and accurate gaze estimation method based on two-eye model of a subject with the tolerance free head movement from Kinect sensor. To accurately efficiently determine point gaze, i) employ to improve accuracy; ii) propose improved convolution-based means gradients localize iris center in 3D space; iii) new personal calibration that only needs one point. The approximates visual axis as line eyeball centers Kappa angles. final can be calculated by using...
Depth image based human action recognition has attracted many attentions due to the popularity of depth sensors. However, accurate still remains a challenge because various object appearances, poses and video sequences. In this paper, novel skeleton joints descriptor on 3D Moving Trend Geometry (3DMTG) property is proposed for recognition. Specifically, histogram moving directions between consecutive frames each joint constructed represent trend feature in spatial domain. The geometry...
Kerogen types exert a decisive effect on the onset and capacity of hydrocarbon generation source rocks. Lacustrine rocks in Liaohe Western Depression are characterized by thick deposition, high total organic carbon (TOC) content, various kerogen types, wide range thermal maturity. Consequently, their potential resource estimation can be misinterpreted. In this study, geochemical tests, numerical analysis, kinetics, basin modeling were integrated to investigate differential effects lacustrine...