Thesis
Long-term large-scale vision health monitoring with cyber glasses
Washington State University
Master of Science (MS), Washington State University
2014
Handle:
https://hdl.handle.net/2376/103492
Abstract
The shortage of wearable vision health monitoring systems hinders the ability to monitor and collect vision health information in a large population over a long period of time. This makes diagnosis, treatment, and prevention of vision health problems (e.g., myopia) ineffective. This thesis introduces Cyber Glasses as a low-cost wearable vision health monitoring system, that is designed for long-term and large-scale vision health monitoring. Based on requirements of vision health monitoring, this thesis presents the design, development, and evaluation of Cyber Glasses and its supporting mechanisms. Cyber Glasses is built entirely from low-cost and commercial-off-the-shelf components. Our Cyber Glasses' vision health monitoring framework includes five main layers: sensing, networking, sensor data collection and processing, vision health parameter extraction, and application. This thesis describes, analyzes, and evaluates each of these layers to identify the most efficient and appropriate supporting mechanisms while carefully considering trade-offs between two substantial requirements: low cost and functionality. Especially, this thesis introduces two novel algorithms for extracting two important vision health parameters: eye blink and eyelid squint. These two algorithms both use a data-driven approach to obtain high accuracy and efficiency, and thus are well-suited for the design of resource-limited Cyber Glasses. The overall evaluation proves that Cyber Glasses, with a suite of efficient supporting mechanisms, can enable long-term and large-scale vision health monitoring.
Metrics
4 File views/ downloads
35 Record Views
Details
- Title
- Long-term large-scale vision health monitoring with cyber glasses
- Creators
- Hoang Congminh Le
- Contributors
- Thanh Xuan Dang (Degree Supervisor)
- Awarding Institution
- Washington State University
- Academic Unit
- Electrical Engineering and Computer Science, School of
- Theses and Dissertations
- Master of Science (MS), Washington State University
- Publisher
- Washington State University; [Pullman, Washington] :
- Identifiers
- 99900525060801842
- Language
- English
- Resource Type
- Thesis