Chu Hai College of Higher Education
Hong Kong, China

Chu Hai College of Higher Education is a private degree-granting institute in Hong Kong. The college is best known for its journalism and Chinese history programmes. At present, Chu Hai College is recognised as an Approved Post Secondary College under the Post Secondary Colleges Ordinance . Wikipedia.

Time filter
Source Type

Liu C.,Tsinghua University | Sun Q.,Tsinghua University | Yang Y.,Chu Hai College of Higher Education
Procedia Engineering | Year: 2017

Granular debris flows are often observed in mountainous areas in Southwestern China. The process is accompanied with large deformation and the evident transitions between solid- and fluid-like states bring difficulties in proposing a unified phenomeno- logical constitutive model. In this study, a hierarchical multi-scale modelling scheme is developed, and is applied to simulate a granular pile collapse. The macroscopic behavior is modelled by using material point method (MPM), which is suitable for large deformation treatment, while the constitution relation at each material point is extracted from discrete element method (DEM) modelling. This MPM/DEM multi-scale modelling strategy abandons any constitutive assumptions as required in MPM, and facilitates effective cross-scale interpretation and understanding of granular flow behavior. It provides a potential approach to simulate large deformation of granular materials when their constitute relations are hard to be derived explicitly. © 2017 The Authors. Published by Elsevier Ltd.

Ko Y.-H.,Oriental Institute of Technology | Ko Y.-H.,Chu Hai College of Higher Education
Applied Ergonomics | Year: 2017

This study explored and identified the effects of luminance contrast, colour combinations, font, and search time on brand icon legibility. A total of 108 participants took part in the experiment. As designed, legibility was measured as a function of the following independent variables: four levels of luminance contrast, sixteen target/background colour combinations, two fonts, and three search times. The results showed that a luminance contrast of 18:1 provided readers with the best legibility. Yellow on black, yellow on blue, and white on blue were the three most legible colour combinations. One of this study's unique findings was that colour combinations may play an even more important role than luminance contrast in the overall legibility of brand icon design. The 12-s search time corresponded with the highest legibility. Arial font was more legible than Times New Roman. These results provide some guidance for brand icon and product advertisement design. © 2017 Elsevier Ltd

Cheung V.S.P.,City University of Hong Kong | YEUNG S.C.R.,City University of Hong Kong | Chung H.S.H.,City University of Hong Kong | Lo A.W.L.,Chu Hai College of Higher Education | Wu W.,Shanghai Maritime University
IEEE Transactions on Power Electronics | Year: 2017

A single-phase transformer-less unified power quality conditioner (TL-UPQC) is presented. Apart from having no isolation transformer, the proposed structure utilizes four switching devices only, forming two half-bridge voltage-source inverters - one connected in parallel with the load and another one connected in series with the AC mains. The two inverters share the same DC link. The parallel inverter, which is controlled by a hysteresis current controller, is used to shape the current drawn from the AC mains and regulate the DC-link voltage. The series inverter, which is controlled by a boundary controller with second-order switching surface, is used to regulate the steady-state load voltage and provide voltage sag / swell ride-through. A DC-link capacitor voltage balancing control that coordinates the operations of the hysteresis and boundary controllers is designed. Modeling, design, and analysis of the whole system will be given. A 1kVA, 110V, 60Hz prototype has been built and evaluated on a setup with a nonlinear load. The steady-state and transient responses under a voltage sag will be given. Experimental results are favorably compared with the theoretical predictions and the performance of other UPQCs. IEEE

Wan K.T.,Chu Hai College of Higher Education | Leung C.K.Y.,Hong Kong University of Science and Technology
Sensors | Year: 2012

Steel corrosion is a major cause of degradation in reinforced concrete structures, and there is a need to develop cost-effective methods to detect the initiation of corrosion in such structures. This paper presents a low cost, easy to use fiber optic corrosion sensor for practical application. Thin iron film is deposited on the end surface of a cleaved optical fiber by sputtering. When light is sent into the fiber, most of it is reflected by the coating. If the surrounding environment is corrosive, the film is corroded and the intensity of the reflected signal drops significantly. In previous work, the sensing principle was verified by various experiments in laboratory and a packaging method was introduced. In this paper, the method of multiplexing several sensors by optical time domain reflectometer (OTDR) and optical splitter is introduced, together with the interpretation of OTDR results. The practical applicability of the proposed sensors is demonstrated in a three-year field trial with the sensors installed in an aggressive marine environment. The durability of the sensor against chemical degradation and physical degradation is also verified by accelerated life test and freeze-thaw cycling test, respectively. © 2012 by the authors.

Cheung V.S.-P.,Smart Energy | Chung H.S.-H.,Smart Energy | Lo A.W.-L.,Chu Hai College of Higher Education
IEEE Transactions on Power Electronics | Year: 2016

A modular and scalable voltage-regulation structure for enhancing service continuity and flexibly changing the system power rating is proposed. The methodology is based on paralleling multiple series-voltage compensators, namely multiparallel-connected series-voltage compensator (MSVC), to regulate the load voltage. The output voltage of each compensator is controlled locally by adjusting the phase angle of the output voltage of the inverter in each compensator, while the output current of each compensator is coupled to two adjacent compensators via two coupling transformers. The coupling transformers form a daisy-chained structure. The load current can be shared near-equally among the compensators through the transformer structure. The operating principle, steady-state and transient current-sharing characteristics of the architecture will be discussed and illustrated. A simplified design procedure will be given. A 3-kVA MSVC test bed with three parallel-connected single-phase compensator units has been built and evaluated. The response of the MSVC system with each compensator unit engaged and disengaged momentarily will be investigated. Such structure is applicable for regulating and stabilizing the supply voltage for consumers at the distribution side. © 1986-2012 IEEE.

Cheung V.S.-P.,Smart Energy | Chung H.S.-H.,Smart Energy | Wang K.-W.,Smart Energy | Lo A.W.-L.,Chu Hai College of Higher Education
IEEE Transactions on Power Electronics | Year: 2014

A modular and scalable series compensation technique for enhancing controllability and flexibility in power transmission in distributed power systems is proposed. The concept is based on paralleling multiple static synchronous series compensators (SSSC) through daisy-chained transformers to perform reactive power compensation, and thus control the amount and direction of the power flow over the transmission link. Each compensator unit is under an autonomous control for regulating its output voltage. Its output current is coupled to another compensator unit through one of the daisy-chained transformers, so that the transmission current is shared among the parallel-connected compensators. Modeling, design, and analysis of a compensator unit and the proposed multiparallel-connected SSSC (MSSSC) architecture in an elementary two-machine system will be presented. A 3-kVA MSSSC test bed with three-parallel-connected compensator units for a single-phase system has been built and evaluated. The steady-state and transient current-sharing characteristics in forward and reverse power transmission over the transmission link will be studied. The response of the MSSSC system with each compensator unit engaged sequentially will also be investigated. © 2013 IEEE.

Liang Z.,Hong Kong Polytechnic University | Chi Z.,Hong Kong Polytechnic University | Fu H.,Hong Kong Polytechnic University | Fu H.,Chu Hai College of Higher Education | And 2 more authors.
Pattern Recognition | Year: 2012

As an important problem in image understanding, salient object detection is essential for image classification, object recognition, as well as image retrieval. In this paper, we propose a new approach to detect salient objects from an image by using content-sensitive hypergraph representation and partitioning. Firstly, a polygonal potential Region-Of-Interest (p-ROI) is extracted through analyzing the edge distribution in an image. Secondly, the image is represented by a content-sensitive hypergraph. Instead of using fixed features and parameters for all the images, we propose a new content-sensitive method for feature selection and hypergraph construction. In this method, the most discriminant color channel which maximizes the difference between p-ROI and the background is selected for each image. Also the number of neighbors in hyperedges is adjusted automatically according to the image content. Finally, an incremental hypergraph partitioning is utilized to generate the candidate regions for the final salient object detection, in which all the candidate regions are evaluated by p-ROI and the best match one will be the selected as final salient object. Our approach has been extensively evaluated on a large benchmark image database. Experimental results show that our approach can not only achieve considerable improvement in terms of commonly adopted performance measures in salient object detection, but also provide more precise object boundaries which is desirable for further image processing and understanding. © 2012 Elsevier Ltd. All rights reserved.

Xu X.,City University of Hong Kong | Po L.-M.,City University of Hong Kong | Cheung K.-W.,Chu Hai College of Higher Education | Ng K.-H.,City University of Hong Kong | And 2 more authors.
ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings | Year: 2012

The performance of view synthesis using depth image based rendering (DIBR) highly depends on the accuracy of depth map. Inaccurate boundary alignment between texture image and depth map especially for large depth discontinuities always cause annoying artifacts in disocclusion regions of the synthesized view. Pre-filtering approach and reliability-based approach have been proposed to tackle this problem. However, pre-filtering approach blurs the depth map with drawback of degradation of the depth map and may also cause distortion in non-hole region. Reliability-based approach uses reliable warping information from other views to fill up holes and is not suitable for the view synthesis with single texture video such as video-plus-depth based DIBR applications. This paper presents a simple and efficient depth map preprocessing method with use of texture edge information to refine depth pixels around the large depth discontinuities. The refined depth map can make the whole texture edge pixels assigned with foreground depth values. It can significantly improve the quality of the synthesized view by avoiding incorrect use of foreground texture information in hole filling. The experimental results show the proposed method achieves superior performance for view synthesis by DIBR especially for large baseline. © 2012 IEEE.

Po L.-M.,City University of Hong Kong | Wong K.-M.,City University of Hong Kong | Cheung K.-W.,Chu Hai College of Higher Education | Ng K.-H.,City University of Hong Kong
IEEE Transactions on Circuits and Systems for Video Technology | Year: 2010

Motion compensated prediction plays a vital role in achieving enormous video compression efficiency in advanced video coding standards. Most practical motion compensated prediction techniques implicitly assume pure translational motions in the video contents for effective operation. Some attempts aiming at more general motion models are usually too complex requiring parameter estimation in practical implementation. In this paper, zoom motion compensation is investigated to extend the assumed model to support both zoom and translation motions. To accomplish practical complexity, a novel and efficient subsampled block-matching zoom motion estimation technique is proposed which makes use of the interpolated reference frames for subpixel motion estimation in a conventional hybrid video coding structure. Specially designed subsampling patterns in block matching are used to realize the translation and zoom motion estimation and compensation. No zoom parameters estimation and additional frame buffers are required in the encoder implementation. The complexity of the decoder is similar to the conventional hybrid video codec that supports subpixel motion compensation. The overall increase in memory requirement and computational complexity is moderate. Experimental results show that the new technique can achieve up to 9.89% bitrate reduction using KTA2.2r1 reference software implementation. © 2006 IEEE.

Shen T.-W.,Chu Hai College of Higher Education | Cheung K.-W.,Chu Hai College of Higher Education
Proceedings - IEEE International Symposium on Circuits and Systems | Year: 2016

Non-rigid image registration is an important preprocessing step in many medical image applications. It is a very computation intensive process. Adaptive bases method with local optimization is one approach used to speed up the commonly used free form deformation with B-spline transformation model. In this paper, an improved local optimization for the adaptive bases nonrigid image registration is proposed. By using smaller support size for basis function in the misregistration region identification process, the registration speed is significantly increased. The registration accuracy is also improved by using higher grid point density for local optimization in the identified misregistration region. The proposed algorithm not only preserves the advantage of computing local deformation on disjoint regions which trying to solve the problem globally, but also further reduces the running time. It can achieve notable improvement over the traditional methods. Performance of the proposed algorithm is evaluated using simulated and clinical brain dataset with standard measures such as RMSE and mutual information based similarity measure (NMI). Experimental results show that significant improvements on speed and accuracy are achieved as compared to the conventional approaches. © 2016 IEEE.

Loading Chu Hai College of Higher Education collaborators
Loading Chu Hai College of Higher Education collaborators