Location:Home > Engineering science > Detection Technology and Automation > 360 degree rotation based on single- camera tracking system

360 degree rotation based on single- camera tracking system

Downloads: []
Tutor: JinFan
School: Kunming University of Science and Technology
Course: Detection Technology and Automation
Keywords: Robot Vision,Image Tracking,SIFT feature matching,Kalman filter algorithm
CLC: TP391.41
Type: Master's thesis
Year:  2009
Facebook Google+ Email Gmail Evernote LinkedIn Twitter Addthis

not access Image Error Other errors

Computer vision technology has been more widely used in mobile robot , video surveillance and tracking , automatic navigation . Movement tracking of target objects is a key technology for robot vision addition require the system to image processing to identify the target location , it also requires the robot motion control system according to the target location , in order to achieve its goal tracking . The work of this study is to use the camera capture an image , and gradation processing is performed and then the image , and select the features in the image area , and to create the image coordinate system . Continuous acquisition process using the image processing technology to keep the two images for image matching , and correcting the image coordinates are consistent , if the offset of the offset vector to the processor , and then according to the direction of the offset of the rotating electrical machine Reversible and speed control . Making tracking target image is always in the center of the camera monitor as the eye to achieve tracking effect . In the area of image processing , the first color image gray pretreatment , and then using SIFT feature matching method for feature matching two images before and after Lowe et al , in order to achieve the purpose of image tracking . The characteristics of rotation , scale scaling , brightness change to maintain the invariance of the advantages and uniqueness of rich amount of information that can provide excellent matching feature points . To adopt tracking control dieless to tracking control can be implemented on any target tracking, eliminating specific targets modeled then tracking this step , and greatly improve the diversity of the tracking system . Algorithm using Kalman filter algorithm , it is about the concept of measuring signal is extracted by the algorithm to estimate the required signal a filtering algorithm , the advantage is not only to be able to estimate the steady one-dimensional random process , non-stationary , multi-dimensional stochastic process can also be estimated .
Related Dissertations
Last updated
Sponsored Links
Home |About Us| Contact Us| Feedback| Privacy | copyright | Back to top