A5080332748- Machine Fault Diagnosis Techniques
- Oil and Gas Production Techniques
- Fault Detection and Control Systems
- Brain Tumor Detection and Classification
- Robotics and Sensor-Based Localization
- Advanced Image and Video Retrieval Techniques
- Cavitation Phenomena in Pumps
- Robotic Path Planning Algorithms
- Dementia and Cognitive Impairment Research
- Aerodynamics and Acoustics in Jet Flows
- Industrial Vision Systems and Defect Detection
- Water Systems and Optimization
- Innovation in Digital Healthcare Systems
- Robotics and Automated Systems
- Machine Learning and ELM
- Cell Image Analysis Techniques
- Structural Health Monitoring Techniques
- Neurological Disease Mechanisms and Treatments
- Indoor and Outdoor Localization Technologies
- Image Processing Techniques and Applications
- Advanced Neural Network Applications
Ocean University of China
2023-2024
Dalian Maritime University
2023
Anhui Polytechnic University
2022
Abstract Simultaneous localization and mapping (SLAM) is considered to be an important way for some smart devices perform automatic path planning, many successful SLAM systems have been developed in the past few years. Most existing approaches rely heavily on static world assumptions, such strong assumptions limit application of most vSLAM (visual SLAM) complex dynamic reality environments, where objects often lead incorrect data association tracking, which reduces overall accuracy...
Based on data-driven and mixed models, this study proposes a fault detection method for autonomous underwater vehicle (AUV) rudder systems. The proposed can effectively detect faults in the absence of angle feedback from rudder. Considering parameter uncertainty AUV motion model resulting dynamics analysis method, we present identification based recurrent neural network (RNN). Prior to identification, singular value decomposition (SVD) was chosen denoise original sensor data as pretreatment...
With the rapid development of industry 4.0 and intelligent factory, safe cooperation between collaborative robot workers is a challenge for development. The use sensitive electronic skin to realize collision detection actively control more practical reliable solution. In this work, we propose an easily fabricated highly flexible, adaptable based on triboelectric nanogenerators, which constructed from BTO/Silicone nanocomposite with fully embedded conductive sponge, significantly enhances...
Aiming at the low accuracy of ship recognition sea and fact that traditional target detection algorithms are greatly affected by environment, this paper proposes a marine method based on an improved SSD (Single shot multibox detector). On basis original algorithm, backbone network VGG16 was first replaced with resnet50, convolution in last 5 layers Resnet50 Conv4_x structure hole convolution, followed removal Conv5_x, subsequent prediction feature layer. A multiscale receptive field module...
According to surveys, in the past 20 years, Alzheimer's disease has become top ten cause of death globally. The volume measurement hippocampus is a prerequisite for early prevention and complementary diagnosis disease. For this reason, by adding Attention mechanism residual module original U-net network, it used segment region automatically. To demonstrate method's efficacy, experiment was carried out on EADC-ADNI Harmonized Protocol Project data set. And compared with other traditional methods.
Abstract Loop closure detection is an indispensable part of the SLAM system, which can reduce cumulative error mobile robots in positioning and mapping. This paper proposes a supervised learning method, combines Siamese network VGG16 deep to handle loop detection. The method traditional are contrasted New College City Centre data sets prove method’s usefulness this research. experiment’s findings reveal that ensures precision recall rate while reducing time necessary assess similarities...
This paper presents a method that can accurately diagnose Alzheimer's disease. Normalized whole brain volume, normalized gray matter and white volume were obtained by segmenting amyloid PET images. Register to unified MNI spatial template perform standardization, select the optimal two-dimensional slice retention level in all images as experimental data. Finally, extracted features of each region are classified EfficientNet-B8. The correct rate classification: 0.987017; 0.998665; matter:...