Apply FPGA Implementation in Interactive System Which Adopts Real-Time Image Processing and Multi Light-Spot Recognition

• Main Authors: Yuan-Kong Liao, 廖元功
• Other Authors: Yih-Ran Sheu
• Format: Others
• Language: zh-TW
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/43423299116061670978

碩士 === 南台科技大學 === 電機工程系 === 98 === Human-Machine Interaction (HMI) has in the past referred to the transmission of commands to a machine through a keyboard and a mouse. Humans, on the other hand, are most accustomed to communicating through body movement and language. For the detection of body movements, such as hand gestures and finger movements, data gloves that integrate 3D virtual reality technology can be employed, with movement recognition achieved through image processing. The popular Wii gaming console is an example of the utilization of body movement detection technology. The instinctive reaction of humans toward objects of interest is the physical act of pointing; therefore an integration of pointing with a graphical user interface on a touch screen system was thus devised. Touch panels feature user-friendly input interface characteristics that are straightforward and almost require no instruction before use. Fingers or stylus pens are utilized to follow directions on the screen to access desired function or content. For this reason, touch panels have been widely applied for use in daily life, such as for guidance systems, PDAs, and electronic dictionaries. Most mainstream touch panels sold in the early stages featured small-sized resistive touch screens. Resistive touch screens detect variables in impedance to sense pointing device location on the screen. Conventional touch screen systems provide detection of only one point of contact at a time, restricting the development of touch screen technology. With the advent of multi-touch panels, the development of advanced functions, such as zooming and rotation, among others, has become possible. There is a diverse array of technologies available for implementation on multi-touch devices. Mainstream technologies seen on the market include Resistive, Projective Capacitive, Capacitive-Magnetic, Optical-Infrared and On Cell Capacitive. Contemporary presentations, meeting reports, and product marketing often requires the use of projectors as a medium. In addition, price decreases have further driven on their application for educational and entertainment purposes. Laser pens act as a basic tool to enhance speaker-audience interaction, whereby the movement of the laser beam leads listeners along with the speaker’s points on the screen. This research paper attempts to utilize laser beams as a functional cursor for further integration with multi-touch technologies. This can result in enhanced presentation performance, and can be further developed for many educational and entertainment applications. Nevertheless, the implementation of multi-touch technology over a long distance has encountered two main problems: One is the issue of producing a multi-touch control device that simple, user-friendly, and yet affordable. The other is the production of a high-speed, high definition and low-cost option that enables multi-touch detection capability for a large screen. Without changing the basic structure of the laser pen, a sleeve was added to our prototype to enable multiple light source functionality. This achieved the desired goal of a low-cost and user-friendly multi-touch mechanism, allowing for its application in long distance multi-touch technology. In addition, FPGA hardware technology was utilized with a low-cost camera as a multiple light source detection solution for high-speed and high-definition requirements of a large screen frame. Furthermore, This research is not confined for use only on Windows system platforms. They can flexibly be put into use on other platforms as well, even being widely applied to uses in digital media, education, and other fields.