A5100656708- Industrial Technology and Control Systems
- Muscle activation and electromyography studies
- Advanced Vision and Imaging
- Hand Gesture Recognition Systems
- 3D Shape Modeling and Analysis
- Heat Transfer and Optimization
- Structural Load-Bearing Analysis
- Advanced Machining and Optimization Techniques
- Medical Image Segmentation Techniques
- Machine Fault Diagnosis Techniques
- Robotics and Sensor-Based Localization
- Advanced machining processes and optimization
- Tactile and Sensory Interactions
- Structural Behavior of Reinforced Concrete
- Advanced Sensor and Control Systems
- Advanced Manufacturing and Logistics Optimization
- Seismic Performance and Analysis
- Robot Manipulation and Learning
- Fault Detection and Control Systems
- Human Pose and Action Recognition
- Advanced Algorithms and Applications
- Nanofluid Flow and Heat Transfer
- Robotic Path Planning Algorithms
- Inertial Sensor and Navigation
- EEG and Brain-Computer Interfaces
University of Victoria
2024
Xidian University
2023
Wuhan University of Science and Technology
2020-2022
Advanced Technology & Materials (China)
2022
China University of Petroleum, East China
2022
Affiliated Hospital of Southwest Medical University
2022
Hefei University of Technology
2021
New York University
2021
China Iron and Steel Research Institute Group
2021
Chongqing University of Posts and Telecommunications
2020
Summary For the problem of surface electromyography (sEMG) gesture recognition, considering fact that traditional machine learning model is susceptible to sEMG feature extraction method, it difficult distinguish subtle differences between similar gestures. The NinaPro DB1 dataset used as research object, and image Convolutional Neural Network (CNN) are combined recognize 52 movements. CNN effectively solves limitations in combines 1‐dim convolution kernel extract deep abstract features...
Summary With the continuous development of sensor technology, acquisition cost RGB‐D images is getting lower and lower, gesture recognition based on depth Red‐Green‐Blue (RGB) has gradually become a research direction in field pattern recognition. However, most current processing methods for are relatively simple, ignoring relationship influence between its two modes, unable to make full use correlation factors different modes. In view above problems, this paper optimizes effect information...
Summary Gesture recognition has always been a research hotspot in the field of human‐computer interaction. Its purpose is to realize natural interaction with machine by recognizing semantics expressed gesture. In process gesture recognition, occlusion an inevitable problem. some or even all features will be lost due gesture, resulting wrong unrecognizability Therefore, it great significance study under occlusion. The single shot multibox detector (SSD) algorithm analyzed, and front‐end...
Point cloud registration (PCR) is a popular research topic in computer vision. Recently, the method an evolutionary way has received continuous attention because of its robustness to initial pose and flexibility objective function design. However, most evolving methods cannot tackle local optimum well they have rarely investigated success ratio, which implies probability not falling into optima closely related practicality algorithm. Evolutionary multi-task optimization (EMTO) widely used...
Abstract The manipulator workspace is an essential element in the field of research and great significance for motion planning. However, little has been conducted on dividing into working subspaces. No precise division method proposed; inverse kinematics multiple solutions trajectory planning may also cause abrupt joint changes, thus affecting planned trajectory. article proposes a subspace all ball‐wrist 6DOF(degree‐of‐freedom) manipulators that satisfy Piper criterion to address above...
Abstract In sEMG (surface electromyography) pattern recognition, most of the research focuses on static recognition different limbs, ignoring importance changing load intensity, and joint angle movement information. Traditional qualitative cannot adjust motion amplitude so it is great significance to study continuous prediction wrist under intensities. Based correlation between surface EMG signal signal, article based neural network identify predict loads continuously quantitatively. The in...
It is common for people to create different types of charts explore a multi-dimensional dataset (table). However, recommend commonly composed in real world, one should take the challenges efficiency, imbalanced data and table context into consideration. In this paper, we propose Table2Charts framework which learns patterns from large corpus (table, charts) pairs. Based on deep Q-learning with copying mechanism heuristic searching, does table-to-sequence generation, where each sequence...
A new mixed nanofluid (Cu/diamond–gallium [Cu/diamond–Ga] nanofluid) is proposed, and the mass ratio of Cu nanoparticles diamond in 10:1. The natural convection heat transfer Cu/diamond–Ga nanofluid, Cu–gallium (Cu–Ga) liquid metal gallium with different volume fractions a rectangular enclosure investigated by single‐phase model this paper. effects temperature difference, nanoparticle fraction kinds on are discussed. three fluids compared. It found that Nusselt numbers along X direction...
This paper proposes a novel earthquake-resilient steel joint with replaceable buckling-restrained link (RBRL). To study its seismic behavior and post-earthquake resilience performance, five full-scale specimens are tested under cyclic loads, during which three parameters investigated: the web connection type, connecting plate thickness core stiffeners. The experimental results presented in terms of test phenomena, hysteresis curves (corresponding skeleton curves, energy dissipation capacity...
Abstract The basis of multi-camera system measurement is the calibration system. Although has achieved certain results, there are still problems such as tedious process and high cost. In this paper, we propose a practical method for systems without overlapping fields view. Firstly, global 3D model scene reconstructed using Simultaneous Localization Mapping (SLAM). Then, each camera calibrated 2D-3D correspondence between images acquired by model. Finally, external parameters calibrated. SLAM...
In this article, we have compared and analyzed the clinical effects of temozolomide single agent combined doxorubicin in treatment glioma. To evaluate this, a total 70 patients diagnosed with glioma our hospital from July 2019 to 2020 were randomly divided into two groups, observation group control group, 35 each group. The received capsules orally. was treated doxorubicin. After treatment, efficacy, adverse reactions, KPS score groups observed. response rate 31.43%, 62.86%. difference...
Deep learning has been successfully applied to the field of fault diagnosis in recent years. Due advantages deep belief network (DBN) fitting nonlinear complex systems and ability wavelet analysis time-frequency analysis, this paper, an improved method based on a neural (DWNN), which combines DBN with morlet activation functions, is proposed for reciprocating compressor. A five-layer using sigmoid, tanh, rectified linear unit (ReLU) functions as hidden layers separately As contrast,...
This paper reported a double layer cantilever harvester powered for vibration energy.A space between the PZT beam and support silicon is existed, leading to structure.Such structure can allow more strain generated by piezoelectric due existing of layer, output voltage further increased.The formula then derived from effect equation model.And verifies relationship parameter.And simulation in micrometer lever used discuss relationship.Finally, prototype assembled measured verify feature...
This paper mainly presents that the influence from high-frequency wave disturbances to motion of AUV when it operates in shallow water.To solve issue above, a nonlinear observer is proposed derive disturbance information by estimating water wave's velocities and AUV's relative position attitude measurement.With estimates, errors X, Y, Z directions are obtained velocity surge, sway heave.Meanwhile, sliding mode controller designed make keep stability while moving three dimensional...
Rolling bearings are one of the most commonly used components in rotating machinery which is mainly operated complex working environment. Therefore, it great theoretical value and practical significance to study state monitoring fault diagnosis technology rolling bearing avoid sudden accidents make a better system maintenance. In this paper, we propose one-dimensional convolutional neural network identify fault. Furthermore, adopt novel activation function: exponential linear units task...
Abstract In this paper, the fluid flow around a cylinder is taken as an example to study different characteristics of flow. Gambit used build 2D model watershed and its cylinders, divide mesh define boundaries. The was constructed with Fluent, steady-state in state simulated calculated. And draw velocity vector steady fluid, use clearly show double vortex behind cylinder. Finally, through Carmen street simulation other parameters, single frequency calculated compared by Fluent verify...
With the trend of world trade entering a period rapid growth, logistics industry is developing very rapidly. At same time, it also has higher requirements for handling and work efficiency. Therefore, this article designs robot with mechanical arm claw crawler tracking unmanned warehouse production workshop. The functions detecting signals, identifying autonomously grasping objects, transporting objects. Intelligent an industrial that can be automated operations such as flexible materials....
In view of the influence external environment interference, parameter changing and random noise in process manipulator trajectory tracking, as well problems fuzzy PID control, such rules are not easy to set, quantisation interval parameters affected by subjective consciousness, which leads problem insufficient performance system. this paper, firstly, MATLAB simulation environment, based on D-H model, linkage model four degree freedom is established, homogeneous transformation matrix...