A5012322206- Face and Expression Recognition
- Face recognition and analysis
- Biometric Identification and Security
- Image Retrieval and Classification Techniques
- Advanced Image and Video Retrieval Techniques
- Domain Adaptation and Few-Shot Learning
- Video Surveillance and Tracking Methods
- AI in cancer detection
- Anomaly Detection Techniques and Applications
- Remote-Sensing Image Classification
- Imbalanced Data Classification Techniques
- Advanced Computational Techniques and Applications
- Industrial Vision Systems and Defect Detection
- Multimodal Machine Learning Applications
- Cryptography and Data Security
- COVID-19 diagnosis using AI
- Advanced Text Analysis Techniques
- Underwater Acoustics Research
- Sentiment Analysis and Opinion Mining
- Privacy-Preserving Technologies in Data
- Text and Document Classification Technologies
- Image and Object Detection Techniques
- Artificial Intelligence in Healthcare
- Vehicle License Plate Recognition
- Advanced Malware Detection Techniques
University College Dublin
2025
Nanchang University
2004-2025
Northwest A&F University
2025
Changsha Central Hospital
2024
University of South China
2024
Donghua University
2013-2023
Beijing University of Technology
2023
University of Exeter
2023
Shanghai Normal University
2022
China Information Technology Security Evaluation Center
2021
Linear discriminant analysis (LDA) is a classical method for discriminative dimensionality reduction. The original LDA may degrade in its performance non-Gaussian data, and be unable to extract sufficient features satisfactorily explain the data when number of classes small. Two prominent extensions address these problems are subclass (SDA) mixture (MSDA). They divide every class into subclasses re-define within-class between-class scatter matrices on basis subclass. In this paper we study...
Rice is an important food crop plant in the world and also a model for genetics breeding research. The germination rate indicator that measures performance of rice seeds. Currently, solutions involving image processing techniques have substantial challenges identification seed germination. detection without human intervention involves because seeds are small densely distributed.In this article, we develop convolutional neural network (YOLO-r) can detect status automatically evaluate total...
The biodiversity components of ESG ratings are analysed to understand whether and how this disclosure mechanism can affect investment decisions, improve outcomes for nature or lead better management based risks. We analyse the relationship between stock returns firms’ related firm characteristics. conclude that largely uncorrelated characteristics other than via size, do not predict returns. Analysis operating performance sheds light on why: return asset profit margins affected by ratings....
The aim of this paper is to demonstrate the integration cultural awareness into AI language learning systems using Natural Language Processing (NLP) models. Recognising that traditional methods tend not capture variations, studies focus on how culturally relevant NLP models (GPT-3, BERT, RNNs) promote acquisition and cross-cultural integration. Data preprocessing, data augmentation model training was used encode variables such as greetings, politeness, contextually expressions datasets....
Accurate millet appearance quality assessment is critical for fair procurement pricing. Traditional manual inspection time-consuming and subjective, necessitating an automated solution. This study proposes a machine-vision-based approach using deep learning dense-scene detection evaluation. High-resolution images of standardized samples were collected via smartphone annotated into seven categories covering impurities, high-quality grains, various defects. To address the challenges with small...
Abstract Transition metal dichalcogenide (TMDC) monolayers provide an ideal platform for exciton and valley-spintronics exploration due to their unique properties. Integrating TMDC with conventional semiconductors allows harnessing the properties of both materials. This strategy holds great promise development advanced optoelectronics spintronic devices. In this work, we investigated valley dynamics in WSe 2 /GaAs heterostructure by employing femtosecond pump-probe ultrafast spectroscopy....
K-nearest neighbor (KNN) is a popular classification algorithm with good scalability, which has been widely used in many fields. When dealing imbalanced data, minority examples are given the same weight as majority existing KNN algorithm. In this paper, we pay more attention to class than class, and increase of according local characteristic distribution. addition, compare proposed Weighted Distance (WDKNN). Experimental results show that our performs better WDKNN data sets.
Deep learning networks have achieved great success in many areas, such as large-scale image processing. They usually need large computing resources and time process easy hard samples inefficiently the same way. Another undesirable problem is that network generally needs to be retrained learn new incoming data. Efforts been made reduce realize incremental by adjusting architectures, scalable effort classifiers, multi-grained cascade forest (gcForest), conditional deep (CDL), tree CNN,...
Few-shot learning has received increasing attention and witnessed significant advances in recent years. However, most of the few-shot methods focus on optimization training process, metric sample generating networks. They ignore importance ground-truth feature distributions classes. This paper proposes a direction-driven weighting method to make classes precisely fit distributions. The learned can generate an unlimited number samples for avoid overfitting. Specifically, proposed consists two...
To understand how diversity change with environmental gradients is a fundamental aim for clarifying biodiversity pattern and underlying mechanisms. Here, we studied the characteristics of beta its partitioning components woody plant communities along an elevation gradient in subtropical forests China, thus explored effects environment space on diversity. By using Classification Method, divided species Daiyun Mountain into four groups, namely generalists, high-elevation specialists,...
As the number of products being sold online increases, it is becoming increasingly difficult for customers to make purchasing decisions based on only pictures and short product descriptions. Thus, customer reviews, particularly text describing features, comparisons experiences using a particular provide rich source information compare decisions. Especially, all kinds reviews from various people have different degree impact buyer, that is, we tend believe our friends who always right than...
Linear discriminant analysis (LDA) is a powerful supervised dimensionality reduction method for analysing high-dimensional data. However, LDA cannot use locality information in data, which makes degrade dramatically performance on multimodal A number of variants have been proposed to exploit including subclass-based LDAs. We discover problem with these variants, that subclasses are selected within-class basis without considering other classes. This causes the loss important at class...