Time filter

Source Type

Haiphong, Vietnam

Vietnam Maritime University is a university in Haiphong operated by the Ministry of Transport. The university was established on 1 April 1956 as Haiphong Maritime University. As of 2007, this is the only maritime university in Vietnam. The university has over 1000 staff, of which, 450 are faculty staff. The university provides undergraduate and graduate education of shipbuilding, maritime navigation, nautical technology. It has a second campus in Vung Tau city. Wikipedia.

Nguyen D.T.,Vietnam Maritime University | Jung J.E.,Chung - Ang University
Future Generation Computer Systems | Year: 2016

Social networking services are becoming increasingly popular during the daily lives of Internet citizens, especially since the advent of smart mobile devices with integrated utility modules such as 4G/WIFI connectivity, global positioning services, cameras, and heart beat sensors. Many devices are available for sharing information at any time, which can be listed by posting a photo, sharing a status, or narrating an event. The behavior of users means that the flow of data (or a social data stream) has real-time characteristics, which actually comprise notifications about your friends' posts after a short delay for diffusion over the network. The data stream contains news pieces related to real social facts as well as unfocused information. In addition, important information (or events) attracts more public attention, which is demonstrated by the number of relevant messages or communication interactions between people interested in specific topics. From a technical perspective, the characteristics of data in the aforementioned scenario provide us with an opportunity to construct a model that can automatically determine the occurrence of events based on a social data stream. In this study, we propose an approach to solve the problem of early event identification, which requires appropriate approaches for processing incoming data in terms of the processing performance and number of data. © 2016 Elsevier B.V.

Nguyen M.Y.,Thai Nguyen University | Nguyen D.M.,Vietnam Maritime University
Electric Power Components and Systems | Year: 2016

This article presents a new approach to the energy efficiency of households based on smart grid technologies applied to the existing power grid. The idea is to make the electrical appliances capable of actively responding to the condition of the upstream grid, e.g., the electricity price, which, under market environments, is passed to end-use customers through retailers or load aggregators. This both saves on the electricity bill for customers and improves the energy efficiency of the system as a whole. The key element is advanced metering infrastructure, which performs two main functions: (1) computing the so-called control policy for electrical appliances with anticipated data and (2) real-time control based on the above control policy and actual measurements. By stochastic optimization, the proposed scheme is capable of monitoring the uncertainty related to the electricity consumption in households, e.g., temperature and/or time and quantity demanded, etc. In addition, with the physical model taken into account, the problem can be applied to various appliances, including what were considered critical loads. The proposed scheme was tested in heat, ventilation, and air-conditioning systems; the simulation result showed the effectiveness in comparison with the traditional (on/off) scheme based on thermostats. © 2016 Taylor & Francis Group, LLC.

Pham D.V.,Vietnam Maritime University
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2012

This paper presents a library written by C# language for the online handwriting recognition system using UNIPEN-online handwritten training set. The recognition engine based on convolution neural networks and yields recognition rates to 99% to MNIST training set, 97% to UNIPEN's digit training set (1a), 89% to a collection of 44022 capital letters and digits (1a,1b) and 89% to lower case letters (1c). These networks are combined to create a larger system which can recognize 62 English characters and digits. A proposed handwriting segmentation algorithm is carried out which can extract sentences, words and characters from handwritten text. The characters then are given as the input to the network. © 2012 Springer-Verlag.

Nguyen M.Y.,Thai Nguyen University | Nguyen D.M.,Vietnam Maritime University
International Journal of Emerging Electric Power Systems | Year: 2015

This paper presents a generalized formulation of Demand Response (DR) under deregulated electricity markets. The problem is scheduling and controls the consumption of electrical loads according to the market price to minimize the energy cost over a day. Taking into account the modeling of customers' comfort (i.e., preference), the formulation can be applied to various types of loads including what was traditionally classified as critical loads (e.g., air conditioning, lights). The proposed DR scheme is based on Dynamic Programming (DP) framework and solved by DP backward algorithm in which the stochastic optimization is used to treat the uncertainty, if any occurred in the problem. The proposed formulation is examined with the DR problem of different loads, including Heat Ventilation and Air Conditioning (HVAC), Electric Vehicles (EVs) and a newly DR on the water supply systems of commercial buildings. The result of simulation shows significant saving can be achieved in comparison with their traditional (On/Off) scheme. © 2015 by De Gruyter.

Nguyen H.-T.,Vietnam Maritime University | Caplier A.,CNRS GIPSA Laboratory
IEEE Transactions on Information Forensics and Security | Year: 2015

We present a novel feature extraction method named local patterns of gradients (LPOGs) for robust face recognition. LPOG uses block-wised elliptical local binary patterns (BELBP), a refined variant of ELBP, and local phase quantization (LPQ) operators directly on gradient images for capturing local texture patterns to build up a feature vector of a face image. From one input image, two directional gradient images are computed. A symmetric pair of BELBP and a LPQ operator are then separately applied upon each gradient image to generate local patterns images. Histogram sequences of local patterns images' nonoverlapped subregions are finally concatenated to form the LPOG vector for the given image. Based on LPOG descriptor, we propose a novel face recognition system which exploits whitened principal component analysis (WPCA) for dimension reduction and weighted angle-based distance for classification. Experimental results on three large public databases (FERET, AR, and SCface) prove that LPOG WPCA system is robust against a wide range of challenges, such as illumination, expression, occlusion, pose, time-lapse variations, and low resolution. In addition, comparison with other systems shows that LPOG WPCA significantly outperforms the state-of-the-art methods. Computationally, timing benchmarks also demonstrate that our LPOG method is faster than many advanced feature extraction algorithms and can be applied in real-world applications. © 2015 IEEE.

Discover hidden collaborations