AIRCC's International Journal of Computer Science and Information Technology

Chao-Ju Chen ¹, Wernhuar Tarng ², Kuo-Hua Lo ³

Affiliations
1 Graduate Institute of Computer Science, National Hsinchu University of Education, TW
2 Graduate Institute of e-Learning Technology, National Hsinchu University of Education, TW
3 Industrial Technology Research Institute, TW

Abstract

A colour-based particle filter can achieve the goal of effective target tracking, but it has some drawbacks when applied in the situations such as: the target and its background with similar colours, occlusion in complex backgrounds, and deformation of the target. To deal with these problems, an improved particle filter tracking system based on colour and moving-edge information is proposed in this study to provide more accurate results in long-term tracking. In this system, the moving-edge information is used to ensure that the target can be enclosed by the bounding box when encountering the problems mentioned above to maintain the correctness of the target model. Using 100 targets in 10 video clips captured indoor and outdoor as the test data, the experimental results show that the proposed system can track the targets effectively to achieve an accuracy rate of 94.6%, higher than that of the colour-based particle filter tracking system proposed by Nummiaro et al. (78.3%) [10]. For the case of occlusion, the former can also achieve an accuracy rate of 91.8%, much higher than that of the latter (67.6%). The experimental results reveal that using the target's moving-edge information can enhance the accuracy and robustness of a particle filter tracking system.

Keywords

Particle Filter, Computer Vision, Target Tracking, Moving Edges, Robustness.

Full Text

   

Wernhuar Tarng ¹, Nien-Yin Lu ², Yi-Syuan Shih ², Hsin-Hun Liou ³

Affiliations
1 Graduate Institute of e-Learning Technology, National Hsinchu University of Education, TW
2 Graduate Institute of E-Learning Technology, National Hsinchu University of Education, IN
3 Department of Computer Science and Information Engineering, National Central University, TW

Abstract

The campus ecological pond is an effective tool to assist science teaching since it allows students to obtain knowledge of aquatic biology in freshwater environments by practical observation. In this study, a virtual campus ecological pond was developed for applications in science education in elementary schools. The system integrates real ecological situations of aquatic environments into learning activities to enhance the learning interest and motivation of students. They can observe the features of aquatic plants and aquatic animals on mobile devices and understand the relation between food chain and ecological balance in aquatic ecosystems by role playing and game missions. The virtual ecological pond can save the cost and manpower needed for building and maintaining a real ecological pond, and it can also solve the problems of insufficient species and difficulty to observe under water. Thus, it is a useful assistant tool for teaching aquatic ecology in elementary schools.

Keywords

Virtual Reality, Campus Ecological Ponds, Role Playing, Food Chain, Ecological Balance.

Full Text

   

Wernhuar Tarng ¹, Shih-Hsien Tsai ¹, Chih Ming Lin ², Chi-Young Lee ³, Hsin-Hun Liou ⁴

Affiliations
1 Graduate Institute of Computer Science, National Hsinchu University of Education, TW
2 Department of Applied Science, National Hsinchu University of Education, TW
3 Department of Materials Science and Engineering, National Tsing Hua University, TW
4 Department of Computer Science and Information Engineering, National Central University, TW

Abstract

The lotus effect is a learning topic of nanoscience commonly seen in elementary schools. However, nanostructures cannot be observed with naked eyes or using general microscopes. Lacking interactive operating environments, it is not easy to inspire students to explore nanophenomena. This study used virtual reality (VR) technology to develop two physical games, namely "Water Drop in the Maze" and "Cleaning the Dust", to show the macroscopic and microscopic structures of lotus leaves respectively. Students can control the rolling direction of water drop by the 3-axis accelerometers in smart phones or tablet PCs to complete the missions, and observe the hydrophobic effect and self-cleaning effect of the lotus as well the as the scales of water drops, dust and the nanostructure on the surface of lotus leaves. A teaching experiment was conducted and the results showed the learning achievement of using physical games to learn the lotus effect was significantly higher than that of using instructional video. The questionnaire results also revealed that most students thought the physical games could enhance their motivation and interest in learning the lotus effect.

Full Text

   

Wernhuar Tarng ¹, Cheng-Kang Chou ¹, Kuo-Liang Ou ¹

Affiliations
1 Graduate Institute of Computer Science, National Hsinchu University of Education, TW

Abstract

A P2P botnet virus detection system based on data-mining algorithms is proposed in this study to detect the infected computers quickly using Bayes Classifier and Neural Network (NN) Classifier. The system can detect P2P botnet viruses in the early stage of infection and report to network managers to avoid further infection. The system adopts real-time flow identification techniques to detect traffic flows produced by P2P application programs and botnet viruses by comparing with the known flow patterns in the database. After trained by adjusting the system parameters using test samples, the experimental results show that the accuracy of Bayes Classifier is 95.78% and that of NN Classifier is 98.71% in detecting P2P botnet viruses and suspected flows to achieve the goal of infection control in a short time.

Keywords

Data Mining, Bayes Classifier, Neural Network, P2P Botnet, Virus Detection Systems.

Full Text

   

Wernhuar Tarng ¹, Chia-Hwei Lin ¹, Hsin-Hun Liou ²

Affiliations
1 Graduate Institute of Computer Science, National Hsinchu University of Education, TW
2 Department of Computer Science and Information Engineering, National Central University, TW

Abstract

Due to the advance of modern medical technology, the average age of global population increases continuously in recent years. In 1993, the society in Taiwan had already reached the criterion of aging population, bringing the country an important issue of looking after old people which could not be ignored by its society. In this study, a wireless sensor device is developed to monitor the activities of old people at home and report fall events in real time. It can decrease the mental and physical injuries caused by fall events, and reduce the cost of looking after the injured old people for a long time. Another objective of this study is to enable the society to realize the problem of looking after old people and pay more attention to the senior citizens living alone as well.

Keywords

Wireless Sensor Networks, Activity Monitoring, Machine Learning, Fall Detection.

Full Text

   

Wernhuar Tarng ¹, Hao-Wei Lin ¹, Kuo-Liang Ou ¹

Affiliations
1 Graduate Institute of Computer Science, National Hsinchu University of Education, Hsinchu, TW

Abstract

The electricity of sensor nodes in wireless sensor networks is very limited, so it is an important research topic to deploy the sensor nodes and cooperate with an efficient routing algorithm for extending the network lifetime. In the related research, LEACH routing algorithm randomly selects cluster heads in each round to form a cluster network, which may cause additional power consumption and inability to maintain the optimal routes for data transmission. The cluster allocation and routing algorithm proposed in this study is based on the cluster architecture of LEACH, and the objective is to produce clusters with more sensor nodes to balance energy consumption of cluster head. For indirect-transmission routing algorithms, the sensor nodes near the base station may consume more energy due to a larger amount of data transmission. Therefore, this study proposed to increase the node density near the base station during deployment to compensate for the requirement of high energy consumption. The experimental results show that the proposed algorithm based on node density distribution can efficiently increase the lifetime of wireless sensor networks.

Keywords

Wireless Sensor Networks, Routing Algorithms, Cluster Networks, Node Density, Network Lifetime.

Full Text