Converting any surfaces to touch interface

• Main Author: Sun, Wen Bin
• Other Authors: Andy Khong Wai Hoong
• Format: Final Year Project
• Language: English
• Published: 2015
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/63571
• Institution: Nanyang Technological University

This project employs the use of computer stereo vision to deliver a touch interface experience. The realization is through the use of two web cameras placed on top of the targeted surface with the field of vision covering and overlapping it. Image Threshold is performed to detect human fingers using HSV or difference detection. Adaptive background model recapturing is implemented to allow the system work in changing lighting environment. A multi-point calibration method is developed to achieve better touch accuracy. This approach can be used for a wide variety of scalable surfaces. The written software utilizes on the availability of Windows 8 Application Program Interface (API) to inject touch inputs. The system and graphical user interface (GUI) are redesigned to work on Windows 8 Operating System (OS) using Microsoft Visual Studio 2013 and QT Creator Integrated Development Environment (IDE). It provides software future-proofing and more user-friendly interface. Due to occlusion of the touch points, it is challenging to track the actual position of the touch points. This report will present the approach for resolving these problems and discuss the remaining problems.