A5038219098- Digital Image Processing Techniques
- Advanced Steganography and Watermarking Techniques
- Interconnection Networks and Systems
- Image Retrieval and Classification Techniques
- Chaos-based Image/Signal Encryption
- Face and Expression Recognition
- Rough Sets and Fuzzy Logic
- Digital Media Forensic Detection
- Data Mining Algorithms and Applications
- Medical Image Segmentation Techniques
- Parallel Computing and Optimization Techniques
- Text and Document Classification Technologies
- User Authentication and Security Systems
- Network Security and Intrusion Detection
- Data Management and Algorithms
- Biometric Identification and Security
- Neural Networks and Applications
- Anomaly Detection Techniques and Applications
- Advanced Image and Video Retrieval Techniques
- Metaheuristic Optimization Algorithms Research
- Stock Market Forecasting Methods
- Cryptography and Data Security
- Advanced Malware Detection Techniques
- Advanced Clustering Algorithms Research
- Distributed systems and fault tolerance
Asia University
2023-2024
China Medical University Hospital
2023-2024
China Medical University
2023-2024
National Taiwan University of Science and Technology
2014-2023
Thammasat University
2023
Sichuan University
2023
University of Technology Sydney
2019-2023
National Yang Ming Chiao Tung University
2018-2022
National Research and Innovation Agency
2022
Nipissing University
2022
Air quality forecasting has been regarded as the key problem of air pollution early warning and control management. In this article, we propose a novel deep learning model for (mainly PM2.5) forecasting, which learns spatial-temporal correlation features interdependence multivariate related time series data by hybrid architecture. Due to nonlinear dynamic characteristics data, base modules our include one-dimensional Convolutional Neural Networks (1D-CNNs) Bi-directional Long Short-term...
The security and privacy preservation issues are prerequisites for vehicular ad hoc networks. Recently, secure enhancing communication schemes (SPECS) was proposed focused on intervehicle communications. SPECS provided a software-based solution to satisfy the requirement gave lower message overhead higher successful rate than previous solutions in verification phase. also presented first group protocol allow vehicles authenticate securely communicate with others of known vehicles....
Vehicular ad hoc networks (VANETs) can significantly improve traffic safety and efficiency. The basic idea is to allow vehicles send information roadside units (RSUs) or other vehicles. Vehicles have be prevented from some attacks on their privacy misuse of private data. For this reason, security preservation issues are important prerequisites for VANETs. identity-based batch verification (IBV) scheme has been recently proposed make VANETs more secure efficient practical use. In paper, we...
The performance of multilabel learning depends heavily on the quality input features. A mass irrelevant and redundant features may seriously affect learning, feature selection is an effective technique to solve this problem. However, most methods mainly emphasize removing these useless features, exploration interaction ignored. Moreover, widespread existence real-world data with uncertainty, ambiguity, noise limits selection. To end, our work dedicated designing efficient robust scheme....
This paper proposes a blind watermarking algorithm based on the significant difference of wavelet coefficient quantization for copyright protection. Every seven nonoverlap coefficients host image are grouped into block. The largest two in block called this and their is difference. We quantized local maximum by comparing value with average all blocks. so that between watermark bit 0 1 exhibits large energy which can be used extraction. During extraction, an adaptive threshold designed to...
Uncertainty and fuzziness generally exist in real-life data. Approximations are employed to describe the uncertain information approximately rough set theory. Certain rules induced directly from different regions partitioned by approximations. Approximation can further be applied datamining-related task, e.g., attribute reduction. Nowadays, types of data collected applications evolve with time, especially new attributes may appear while objects added. This paper presents an approach for...
Traffic flow forecasting has been regarded as a key problem of intelligent transport systems. In this work, we propose hybrid multimodal deep learning method for short-term traffic forecasting, which can jointly and adaptively learn the spatial-temporal correlation features long temporal interdependence multi-modality data by an attention auxiliary architecture. According to highly nonlinear characteristics data, base module our consists one-dimensional Convolutional Neural Networks (1D CNN)...
Rule induction method based on rough set theory (RST) has received much attention recently since it may generate a minimal of rules from the decision system for real-life applications by using attribute reduction and approximations. The vary with time, e.g., variation objects, attributes values. approximations alter Attribute Values' Coarsening Refining (AVCR), kind values, which results in alteration simultaneously. This paper aims dynamic maintenance w.r.t. AVCR. definition discernibility...