Asian Science Citation Index is committed to provide an authoritative, trusted and significant information by the coverage of the most important and influential journals to meet the needs of the global scientific community.  
ASCI Database
308-Lasani Town,
Sargodha Road,
Faisalabad, Pakistan
Fax: +92-41-8815544
Contact Via Web
Suggest a Journal
 
Articles by Hu Ng
Total Records ( 7 ) for Hu Ng
  Hau-Lee Tong , Mohammad Faizal Ahmad Fauzi , Su-Cheng Haw , Hu Ng , Timothy Tzen-Vun Yap and Chiung Ching Ho
  This study reports the classification of subdural and extradural hematomas in brain CT images. The major difference between subdural and extradural hematomas lies in their shapes, therefore eight shape descriptors are proposed to describe the characteristics of the two types of hematoma. The images will first undergo the pre-processing step which consists of two-level contrast enhancement separated by parenchyma extraction processes. Next, k-means clustering is performed to garner all Regions of Interest (ROIs) into one cluster. Prior to classification, shape features are extracted from each ROI. Finally for classification, fuzzy k-Nearest Neighbor (fuzzy k-NN) and Linear Discriminant Analysis (LDA) are employed to classify the regions into subdural hematoma, extradural hematoma or normal regions. Experimental results suggest that fuzzy k-NN produces the optimum accuracy. It manages to achieve over 93% correct classification rate on a set of 109 subdural and 247 extradural hematoma regions, as well as 629 normal regions.
  Aruneswaran Selvam , Timothy Tzen-Vun Yap , Hu Ng , Hau-Lee Tong and Chiung-Ching Ho
  This research aims to develop an information retrieval application based on Augmented Reality (AR) technologies to enhance visitors’ experience in a museum exhibition. The purpose of developing this application is to give visitors of museums a customized interactive experience through a handheld smartphone. The application recognizes objects of interest and retrieve information of such objects for display through feeds from a smartphone’s camera in real time and overlays the information over the object. This is achieved with vision-based AR, utilizing 3D object tracking, thus eliminating the use of markers which could prove unreliable due to obfuscation or damage.
  Timothy Tzen Vun Yap , Hu Ng , Vik Tor Goh and Jeng Weng Seah
  The objective of this research is to collect and analyze road surface conditions in Malaysia and develop a classification model that can identify road surface conditions from the collected data. Data is collected through a mobile application that collects positional dynamics of vehicles on the road. Features considered include statistical measures such as minimum, maximum, standard deviation, median, average, skewness and kurtosis. Selection of the extracted features is performed using Ranker, Tabu search and Particle Swarm Optimization (PSO) followed by classification using k-Nearest Neighborhood (k-NN) Random Forest (RF) and Support Vector Machine (SVM) with linear, Radial Basis Function (RBF) and polynomial kernels. The classification model that gave the highest accuracy is SVM (RBF) with a Correct Classification Rate (CCR) of 91.71%. Trailing closely was RF at 91.17%. Although not as accurate as SVM, the difference was negligible and its computational time was much lower than the former. In the feature selection process, features which provide positive contribution to the classification process were chosen and the best performances were produced by PSO with an average CCR of 89.88%. Tabu selected 11 features while PSO selected 13 features where the extra two features made a difference in the results. Ranker selected every single feature but has the lowest average CCR. This is attributed to a subset of features that were selected were ineffectively impeding the classification. The features and classification model employed were able to effectively classify road surface conditions from vehicle positional dynamics. Using only 3D positional readings of the vehicle and standard statistical measures, road surface conditions can be effectively identified for the prioritisation and facilitation of road maintenance.
  Hu Ng , Timothy Tzen Vun Yap , Hau Lee Tong , Chiung Ching Ho , Lay Kun Tan , Wan Xin Eng , Seng Kuan Yap and Jia Hao Soh
  The objective of this study is to classify multimodal human actions of the Berkeley Multimodal Human Action Database (MHAD). Actions from accelerometer and motion capture modals are utilized in this study. Features extracted include statistical measures such as minimum, maximum, mean, median, standard deviation, kurtosis and skewness. Feature extraction level fusion is applied to form a feature vector comprising two modalities. Feature selection is implemented using Particle Swarm Optimization (PSO) Tabu and Ranker. Classification is performed with Support Vector Machine (SVM) Random Forest (RF) k-Nearest Neighbour (k-NN) and Best First Tree (BFT). The classification model that gave the highest accuracy is support vector machine with radial basis function kernel with a Correct Classification Rate (CCR) of 97.6 % for the Accelerometer modal (Acc) 99.8% for the Motion capture system modal (Mocap) and 99.8% for the Fusion Modal (FusioMA). In the feature selection process, ranker selected every single extracted feature (162 features for Acc and 1161 features for Mocap and 1323 features for FusioMA) and produced an average CCR of 97.4%. Comparing with PSO (68 features for Acc, 350 features for Mocap and 412 features for FusioMA) it produced an average CCR of 97.1% and Tabu (54 features for Acc, 199 features for Mocap and 323 features for FusionMA) produced an average CCR of 97.2%. Although, Ranker gave the best result, the difference in the average CCR is not significant. Thus, PSO and Tabu may be more suitable in this case as the reduced feature set can result in computational speedup and reduced complexity. The extracted statistical features are able to produce high accuracy in classification of multimodal human actions. The feature extraction level fusion to combine the two modalities performs better than single modality in the classification.
  Hau Lee Tong , Hu Ng , Tzen Vun Timothy Yap , Wan Siti Halimatul Munirah Wan Ahmad and Mohammad Faizal Ahmad Fauzi
  The main objective is to evaluate different feature extraction and selection techniques as well as classification performances for the wood defect images. This study presents a classification system to classify the defect images from a database provided by a wood factory. This database consists of 1498 defect images and they are classified using Support Vector Machine (SVM), J48, random forest and K-NN classifiers. The features for each defect image are extracted using six types of feature extraction techniques. Feature selection methods are used to choose the features according to their significance. From the findings, it can be observed that Ranker method produced the best performance for most of the feature extraction techniques and classifiers. This directly indicates that all the extracted features have significant contribution. For SVM, it is tested with three different settings: linear, RBF and polynomial. The highest classification rate is obtained by using Gray Level Co-occurrence Matrix (GLCM) with SVM polynomial. For J48 and random forest classifier, features computed using Colour Coherence Vector (CCV) yielded the best measure, whilst for K-NN, it is Gabor features which performed best. Besides 89.85% of case crack are correctly classified, 38.63% for fungus, 16.48% for knot, 88.06% for worm holes and 51.61% for watermark case. For defect cases other than crack, it is observed that the number of misclassification cases is biased on crack case. The proposed methodology can be applied to create an automated visual inspection system for detection of semi-finished wood defect in the wood industry.
  Hau Lee Tong , Mohammad Faizal Ahmad Fauzi , Su Cheng Haw , Tzen Vun Timothy Yap and Hu Ng
  The main objective is to annotate and classify different types of hemorrhagic slices such as intra-axial, subdural and extradural slices. A two-segregated annotation is proposed to classify hemorrhagic slices due to their different shapes and locations in the brain. The first annotation is to identify the intra-axial hemorrhage slice whereas the second annotation is to classify the subdural and extradural slices. All the extracted features from both annotations will be used as inputs to the Support Vector Machine (SVM) classifier. Experiments conducted on a set of 519 CT slices under the proposed method show significant results. From the findings, the proposed method yields 79.3, 85 and 89.2% correct classification rate for intra-axial, subdural and extradural. On overall, the CCR obtained for subdural and extradural slices is higher than intra-axial slices. This is contributed by more specific local shape features are employed for subdural and extradural which results in better recognition. Global features are adopted to classify the intra-axial slices due to their arbitrary shapes. The proposed approach can be used to create an automated retrieval system so that radiologists and medical students can use it to retrieve the hemorrhage images for further study and analysis.
  Wan Siti Halimatul Munirah Wan Ahmad , Hau-Lee Tong , Hu Ng , Timothy Tzen-Vun Yap and Mohammad Faizal Ahmad Fauzi
  This study introduces a analysis on the comparison of feature extraction techniques on the segmented wood defects. The method used is to classify four types of wood defects, namely knot, crack, holes and algae. The wood defect dataset used in this research consisted of 145 images were obtained from various sources. Fuzzy C-Means (FCM) is employed to segment wood defects into four clusters. Six types of feature extraction techniques namely Colour Histogram, Colour Coherence Vector, Local Binary Pattern, Gacfrdfbor Transform, Discrete Wavelet Frame and Gray Level Co-occurrence Matric are employed to describe the image’s feature. The performances of Support Vector Machines (SVM), Bayes Networks and tree-based classifiers are comparedon the different defects and the classifiers’ performances for each extraction technique are investigated. The experiment shows promising results with the highest classification accuracy of 94.5%, achieved by Random Forest classifier using Colour Histogram features. The proposed framework is useful in the automation of the detection of wood defects and is a superior alternative to manual selection and classification in the wood quality control.
 
 
 
Copyright   |   Desclaimer   |    Privacy Policy   |   Browsers   |   Accessibility