A5100766916- Dental Radiography and Imaging
- Advanced Neural Network Applications
- Advanced X-ray and CT Imaging
- Video Surveillance and Tracking Methods
- Advanced MIMO Systems Optimization
- Network Security and Intrusion Detection
- Drilling and Well Engineering
- Gait Recognition and Analysis
- Anomaly Detection Techniques and Applications
- Photovoltaic System Optimization Techniques
- Solar Radiation and Photovoltaics
- Visual Attention and Saliency Detection
- Face recognition and analysis
- Endodontics and Root Canal Treatments
- Advanced Algorithms and Applications
- Medical Image Segmentation Techniques
- Sensor Technology and Measurement Systems
- Indoor and Outdoor Localization Technologies
- Oral microbiology and periodontitis research
- Internet Traffic Analysis and Secure E-voting
- Advanced Image and Video Retrieval Techniques
- AI in cancer detection
- Energy Load and Power Forecasting
- Advanced Sensor and Control Systems
- Brain Tumor Detection and Classification
University of Science and Technology Liaoning
2024
Gansu Institute of Political Science and Law
2023
Chang Gung Memorial Hospital
2021-2023
Tatung University
2019
System Equipment (China)
2016
Software (Spain)
2012
Caries is a dental disease caused by bacterial infection. If the cause of caries detected early, treatment will be relatively easy, which in turn prevents from spreading. The current common procedure dentists to first perform radiographic examination on patient and mark lesions manually. However, work judging markings requires professional experience very time-consuming repetitive. Taking advantage rapid development artificial intelligence imaging research technical methods help make...
Apical lesions, the general term for chronic infectious diseases, are very common dental diseases in modern life, and caused by various factors. The current prevailing endodontic treatment makes use of X-ray photography taken from patients where lesion area is marked manually, which therefore time consuming. Additionally, some images significant details might not be recognizable due to different shooting angles or doses. To make diagnosis process shorter efficient, repetitive tasks should...
Furcation defects pose a significant challenge in the diagnosis and treatment planning of periodontal diseases. The accurate detection furcation involvements (FI) on periapical radiographs (PAs) is crucial for success therapy. This research proposes deep learning-based approach to defect using convolutional neural networks (CNN) with an accuracy rate 95%. has undergone rigorous review by Institutional Review Board (IRB) received accreditation under number 202002030B0C505. A dataset 300 teeth...
In this study, we propose a novel lightweight detection model for rebar counting, which is rectified mobilenet feature pyramid network based on YOLO (RM-LFPN-YOLO). The incorporates backbone that integrates the coordinate attention (CA) mechanism, (LFPN), and loss function combines focal efficient intersection over union (EIOU) loss, all meticulously designed to enhance model's performance. Experimental results demonstrate our improved algorithm, with mere 25.08M parameters, computes...
This paper presents a novel method for automatic segmentation of dental X-ray images into single tooth sections and placing every segmented onto precise corresponding position table. Moreover, the proposed automatically determines tooth’s in panoramic film. The image-processing step incorporates variety image-enhancement techniques, including sharpening, histogram equalization, flat-field correction. image processing was implemented iteratively to achieve higher pixel value contrast between...
Person reidentification matches person observations captured by nonoverlapping cameras at different times and locations. Due to the appearance variations caused view angle, pose, lighting, background, occlusion, is a challenging task. To improve accuracy of reidentification, this paper proposes metric learning method. The main idea our method learn low-dimensional space, in which features extracted from same are pulled, persons but neighborhood original feature space pushed. Then all...
In the fast-evolving landscape of digital networks, incidence network intrusions has escalated alarmingly.Simultaneously, crucial role time series data in intrusion detection remains largely underappreciated, with most systems failing to capture time-bound nuances traffic.This leads compromised accuracy and overlooked temporal patterns.Addressing this gap, we introduce a novel SSAE-TCN-BiLSTM (STL) model that integrates analysis, significantly enhancing capabilities.Our approach reduces...
The number of time steps used for machine learning methods on the hourly prediction solar photovoltaic (PV) generation is typically set to one. In this paper, we investigate larger-than-one long short-term memory (LSTM)-based method improve accuracy with a limited dataset. Besides, analyze contribution all available features toward different and select proper better training. By using real-world dataset, experimental results show that proposed can effectively outperform conventional one whilst same
In this study, energy-efficient deterministic adaptive beamforming algorithms are proposed for distributed sensor/relay networks. Specifically, DBSA, D-QESA, D-QESA-E, and a hybrid algorithm, hybrid-QESA, that combines the benefits of both random algorithms, proposed. Rigorous convergence analyses provided all our to global optimal solution is shown algorithms. Through extensive numerical simulations, we demonstrate superior performance achieved by DBSA D-QESA over static channels....
To solve the problem that positioning strategy with sliding window approaches requires exhaustive search in feature pyramids, paper proposes an object detection algorithm based on deformable part models Bing features to help detection.First of all, input images are preprocessed objectness and a set potential windows may contain target objects obtained, then model is regarded as class-specific detector match windows, at last Non-Maximum Suppression used merge reduce areas results obtain final...