Vision based mouse manipulation on flat surfaces

By: Alvarez, Pocholo C
Publisher: Cebu City ; CIT-U ; 2012DDC classification: T Al863 2012 Summary: Alvarez, Pocholo C.;Cebu Institute of Technology-Unversity; March 14 2012;Vision- Based Mouse Manipulation on Flat Surfaces. Adviser: Dr. Elmer Maravillas Human-Computer Interaction (HCI) has come a long way from mouse and keyboard to touch screens, voice recognition, and etc. Unfortunately, most of these applications are costly. I present a HCI that makes use of two-cost web cameras to manipulate mouse like touch screen system. The system uses background subtraction for the extraction of objects. The extracted object are then refined using morphological algorithm like erosion and dilation for erasing noise and enhancing the object shape respectively. The system then searches the biggest extracted object then gets its shape or contour. BAsed on the points give by the contour, the system will search for the lowest point; this lowest point is the location of the mouse. Scan-line based algorithm was used to detect if an object has touched the surfaced. Experimental result show that the positions of the cameras effects the accurancy and the stability of the mouse. While background subtraction is not an effective object extaction since it is sensitive to illumination. Since the system uses background subtraction, the system quickly fails if there is a change in the lighting. I recommend future researchers of this study to use efficient color segmentation for the extraction objects. Next issue is the stability and accurancy of the mouse location. I recommend to use a special formula to convert (x,y) location from 640x480 to (x,y) location from 1024x724.
Tags from this library: No tags from this library for this title. Log in to add tags.
    Average rating: 0.0 (0 votes)

Alvarez, Pocholo C.;Cebu Institute of Technology-Unversity; March 14 2012;Vision- Based Mouse Manipulation on Flat Surfaces.

Adviser: Dr. Elmer Maravillas

Human-Computer Interaction (HCI) has come a long way from mouse and keyboard to touch screens, voice recognition, and etc. Unfortunately, most of these applications are costly. I present a HCI that makes use of two-cost web cameras to manipulate mouse like touch screen system.

The system uses background subtraction for the extraction of objects. The extracted object are then refined using morphological algorithm like erosion and dilation for erasing noise and enhancing the object shape respectively. The system then searches the biggest extracted object then gets its shape or contour. BAsed on the points give by the contour, the system will search for the lowest point; this lowest point is the location of the mouse. Scan-line based algorithm was used to detect if an object has touched the surfaced.

Experimental result show that the positions of the cameras effects the accurancy and the stability of the mouse. While background subtraction is not an effective object extaction since it is sensitive to illumination.

Since the system uses background subtraction, the system quickly fails if there is a change in the lighting. I recommend future researchers of this study to use efficient color segmentation for the extraction objects. Next issue is the stability and accurancy of the mouse location. I recommend to use a special formula to convert (x,y) location from 640x480 to (x,y) location from 1024x724.

000-099

There are no comments for this item.

to post a comment.