A5101468862- Advanced Computational Techniques and Applications
- Advanced Image and Video Retrieval Techniques
- Advanced Neural Network Applications
- Robotics and Sensor-Based Localization
- Visual Attention and Saliency Detection
- 3D Surveying and Cultural Heritage
- Conservation Techniques and Studies
- Video Surveillance and Tracking Methods
- Infrared Target Detection Methodologies
- Advanced Decision-Making Techniques
- Optical Systems and Laser Technology
- Domain Adaptation and Few-Shot Learning
- Natural Language Processing Techniques
- Rough Sets and Fuzzy Logic
- Infrastructure Maintenance and Monitoring
- Advanced Image Processing Techniques
- Topic Modeling
- COVID-19 diagnosis using AI
- Advanced Measurement and Detection Methods
- Advanced Numerical Analysis Techniques
- Handwritten Text Recognition Techniques
- Remote-Sensing Image Classification
- Advanced Text Analysis Techniques
- Advanced Clustering Algorithms Research
- Control and Dynamics of Mobile Robots
Beihang University
2023-2025
Anhui University
2006-2024
Qinghai Normal University
2023-2024
Yantai University
2023-2024
Taiyuan Normal University
2010-2024
Southeast University
2023
Beijing Normal University
2023
Changzhou University
2019-2023
Luoyang Normal University
2020-2023
Beijing Jiaotong University
2021
Abstract Manual inspection (i.e., visual and/or with professional equipment) is the most predominant approach for identifying and assessing superficial damage of masonry historic structures at present. However, this method costly times difficult to apply remote or components. Existing convolutional neural network (CNN)‐based detection methods have not been specifically designed multiple identification structures. To overcome these limits, a deep architecture CNN classification techniques...
Abstract Detecting and measuring the damage on historic glazed tiles plays an important role in maintenance protection of buildings. However, current visual inspection method for identifying assessing superficial buildings is time labor intensive. In this article, a novel two‐level object detection, segmentation, measurement strategy large‐scale structures based deep‐learning technique proposed. The data study are from roof images Palace Museum China. first level model, which Faster...
This paper delves into the practical applications and effectiveness of two prominent text representation methods, Bag-of-Words (BoW) model Term Frequency-Inverse Document Frequency (TF-IDF), in realm Natural Language Processing (NLP). It commences with an introductory overview NLP its pivotal role broader field Artificial Intelligence (AI), elucidating importance enabling computers to comprehend manipulate human language. Subsequently, a comprehensive elucidation underlying principles...
In this letter, we propose a hybrid maximum likelihood (HML) classifier for continuous phase modulation (CPM). To the best of our knowledge, proposed function is first one CPM signals that based on two its main features: nonlinear waveform, which represented with principal components, and signal memory, modeled as Markov mapping symbol sequence. Unknown channel parameters are estimated through expectation-maximization (EM) algorithm. An approximation method further to ensure improves...
At present, computer-aided spoken English learning is becoming increasingly popular among learners. The comprehensive analysis tech-nology can evaluate and correct learner's pronunciation, thereby im-proving their pronunciation. Based on analy-sis, this paper aims to explore the automatic recognition scoring methods of in education. For this, it studies effective matching feedback information with known pronunciation results, then develops a computer evaluation plug-in consisting dif-ferent...
Ancient architectures have a lot of distinctive patterns on their component surfaces. But due to the wind and Sun, will present extensive deficiencies, which are not conducive routine inspections subsequent repair work. To overcome these limits, an automatic deep learning-based method segmentation, inpainting, classification for defective ancient architecture is proposed. First, You Only Look At CoefficienTs, real-time instance segmentation network, applied obtain mask parts. Then Generative...
This paper presents a collision-free path planning approach based on Bezier curve for mobile manipulator with the endpoints restricted by manipulator. Based these candidate and initial posture of manipulator, series feasible paths are obtained constraints from velocity, acceleration environment. And then optimal is determined according to related information as well obstacles. The has ability adapt environment will be updated once new detected obstacles block this path. verified simulations.
Due to the complicacy of communication control system (CCS) structure and variations in operating conditions, occurrence a fault inside CCS is uncertain random. Aiming at limitations current diagnosis CCS, paper presents an approach fuzzy fusion based on tree. Elaborated method considers characteristic diagnostic object establish tree, convert index rate into number rate, perform analysis for determine confidence interval probability top event, achieve reasoning result. The details design...