Bee Theng Lau

Portable real time emotion detection system for the disabled

This paper discusses the development of a portable real time emotion detection system for the use of disabled. This system allows the user to train the system with his/her profile comprising of expressions shown on the faces when different emotions occur. With a trained profile that can be updated flexibly, a user can detect his/her behaviour on real time basis. It utilizes the state of the art of face detection and recognition algorithms. The combination of Viola-Jones is used to detect frontal face from video which integrates Haar like features, integral image and AdaBoost learning rule. Principal Component Analysis that uses Eigenfaces to detect emotion shown on the faces is applied. For portability, it requires only a laptop or palmtop with built in camera and speaker that can be mounted on wheelchair easily. It works well in both indoor and outdoor environment day and night.

A human orientation tracking system using Template Matching and active Infrared marker

Human tracking research has been reinforced with the introduction of vision-based motion tracking sensors such as Microsoft Kinect and Intel RealSense since 2009. No longer limited to just wearable sensors, embedded environments and 2D camera feed, stereoscopic digital cameras and Infrared imaging has enabled perceivable depth and distance to contribute to geo-location, activity tracking, and automated detection of abnormal events. However, current motion tracking systems have a limited zone of detection and tracking, subjected to environmental lighting, lens occlusion and hardware factors. This study proposes and demonstrates the possibility of complementing these existing sensors with a simple human tracking system that can be used to direct their reorientation in order to maintain the tracked person within their optimal zone of detection. The prototype in this study utilizes an Infrared Camera and an active Infrared marker to track the pan orientation of a person by means of Template Matching - finding a template image match that indicates the angle of marker pan. By knowing the targets angle of pan, a Microsoft Kinect sensor can be autonomously relocated to the front of the person via the shortest path around. This implementation may also be retooled for other systems that have a smaller zone of detection.

An Improved Indoor Robot Human-Following Navigation Model Using Depth Camera, Active IR Marker and Proximity Sensors Fusion

Creating a navigation system for autonomous companion robots has always been a difficult process, which must contend with a dynamically changing environment, which is populated by a myriad of obstructions and an unspecific number of people, other than the intended person, to follow. This study documents the implementation of an indoor autonomous robot navigation model, based on multi-sensor fusion, using Microsoft Robotics Developer Studio 4 (MRDS). The model relies on a depth camera, a limited array of proximity sensors and an active IR marker tracking system. This allows the robot to lock onto the correct target for human-following, while approximating the best starting direction to begin maneuvering around obstacles for minimum required motion. The system is implemented according to a navigation algorithm that transforms the data from all three types of sensors into tendency arrays and fuses them to determine whether to take a leftward or rightward route around an encountered obstacle. The decision process considers visible short, medium and long-range obstructions and the current position of the target person. The system is implemented using MRDS and its functional test performance is presented over a series of Virtual Simulation Environment scenarios, greenlighting further extensive benchmark simulations.

Optimising technology to measure functional vision, mobility and service outcomes for people with low vision or blindness: protocol for a prospective cohort study in Australia and Malaysia

Introduction Orientation and mobility (O&M) specialists assess the functional vision and O&M skills of people with mobility problems, usually relating to low vision or blindness. There are numerous O&M assessment checklists but no measures that reduce qualitative assessment data to a single comparable score suitable for assessing any O&M client, of any age or ability, in any location. Functional measures are needed internationally to align O&M assessment practices, guide referrals, profile O&M clients, plan appropriate services and evaluate outcomes from O&M programmes (eg, long cane training), assistive technology (eg, hazard sensors) and medical interventions (eg, retinal implants). This study aims to validate two new measures of functional performance vision-related outcomes in orientation and mobility (VROOM) and orientation and mobility outcomes (OMO) in the context of ordinary O&M assessments in Australia, with cultural comparisons in Malaysia, also developing phone apps and online training to streamline professional assessment practices. Methods and analysis This multiphase observational study will employ embedded mixed methods with a qualitative/quantitative priority: corating functional vision and O&M during social inquiry. Australian O&M agencies (n=15) provide the sampling frame. O&M specialists will use quota sampling to generate cross-sectional assessment data (n=400) before investigating selected cohorts in outcome studies. Cultural relevance of the VROOM and OMO tools will be investigated in Malaysia, where the tools will inform the design of assistive devices and evaluate prototypes. Exploratory and confirmatory factor analysis, Rasch modelling, cluster analysis and analysis of variance will be undertaken along with descriptive analysis of measurement data. Qualitative findings will be used to interpret VROOM and OMO scores, filter statistically significant results, warrant their generalisability and identify additional relevant constructs that could also be measured. Ethics and dissemination Ethical approval has been granted by the Human Research Ethics Committee at Swinburne University (SHR Project 2016/316). Dissemination of results will be via agency reports, journal articles and conference presentations.

Undergraduate Students’ Perceptions as Producer of Screencast Videos in Learning Mathematics

In this digital age, universities shift their teaching and learning from conventional to blended learning utilizing various state-of-the-art digital technologies. Thus, online video-based teaching and learning have also gained its popularity among educators to facilitate the students’ learning. However, little research has been conducted on learner-produced-videos-based approach. Therefore, this research investigates and designs an innovative student-centered learning approach through collaborative production of mathematics screencast videos. A survey was also conducted to understand the learners’ perceptions in terms of benefits, challenges, motivation, sharing platform, and effectiveness of learning. This paper discusses the design, methodology, evaluation, and findings of the study conducted.

Parental Perceptions, Attitudes and Involvement in Interventions for Autism Spectrum Disorders in Sarawak, Malaysia

Purpose: This study explores and compares perspectives of educators and parents regarding interventions used in managing Autism Spectrum Disorder (ASD) in Sarawak, Malaysia. Information on parental desires and limitations when selecting and maintaining management will aid in the development of strategies for ASD educators to work effectively with parents and caregivers, and vice versa. Method: This qualitative research employed traditional question and answer interviews with 7 ASD educators and 30 parents. Interviews were semistructured and questions were open-ended to allow for additional details to be relayed within the scope of the subject matter. Thematic analysis revealed overarching perceptions concerning parental attitudes towards involvement in their children’s interventions, and implications of cultural context. Results: Perspectives were similar regarding the importance of confidentiality from educators and cultural factors playing a major role in content of intervention chosen by parents. Perspectives differed across four themes. Of significance was the way in which both samples viewed parental self-reliance in supplementing interventions and parental attitudes in effort and perseverance. Conclusion: Parental resources and culture influence ability and attitudes towards involvement. While educators may not agree on certain aspects, mutual appreciation of differing perspectives would benefit the children. Limitations: Due to the lack of ASD service providers in the region as well as cultural reservations, sample size of educators and parents was small. The sample of parents is not typical of parents coping with ASD in the region. Vol. 29, No.1, 2018; doi 10.5463/DCID.v29i1.664

Learn with me: collaborative virtual learning for the special children

Collaborative learning environment is regarded as stimulating and engaging for normal learners. The main aim of our research is to investigate its effectiveness in assisting the learning of children with disabilities. We developed a prototype, Learn with Me and conducted a testing on 6 children who have been diagnosed with cerebral palsy and 7 children who have been diagnosed with autism spectrum disorders. Participants were invited to take part in two tests. Result showed participants learn better with responsive virtual tutor as compared to non-responsive virtual learning.

An Intelligent Real-Time Communication Assistant for the Disabled

This research focuses on developing a real-time communication assistant for the disabled utilizing the biometric information of their facial features. Our targeted community is that with communication disabilities. While communicating effectively is a problem for this community, they have facial expressions and incomprehensible speech that can be interpreted to associate with their needs or requests. Thus, we utilize the real-time face detection technique to capture their facial expressions. This paper focuses mainly on discussion of the performance of real-time biometric information detection in our prototype, the EmoCom. EmoCom can also serve as a monitoring system to track the activities of the disabled with limited physical mobility.

Template Design Using Extremal Optimization with Multiple Search Operators

The template design problem is a constrained optimization problem originated from the printing industry. It involves printing several variations of a design onto one or more stencil sheets, where the aims are to minimize the number of stencils as well as the overproduction of prints of a particular design. Over the years, exact solution methods have been used to solve the problem. These methods could be useful for small to moderate-sized problem instances. However, when the problem instances are huge, the search space may easily grow too large for the systematic approaches. To date, no meta-heuristic or soft computing techniques have been used for this problem. In this paper, we propose the use of Extremal Optimization (EO) with multiple search operators for solving the template design problem. Different combinations of the search operators are tested via extensive numerical experiments. The results show that EO is indeed a feasible approach for template design optimization. The hybridization of EO with a deterministic local search has proven to be particularly effective.