260 likes | 1.16k Views
Final Year Project. Kalman Tracking for Image Processing Applications Progress Presentation Julius Oyeleke 06360289. Project Brief. To track one or more red balls over a frame of video
E N D
Final Year Project Kalman Tracking for Image Processing Applications Progress Presentation Julius Oyeleke 06360289
Project Brief • To track one or more red balls over a frame of video • Find the centre of each ball and associate them in multiple frames. • Use Kalman filtering to • track the red ball in the image. • predict the path of the ball in future as an aid of detection. • OpenCV (computer vision library) is being used in this project
Why OpenCV • Real time computer vision. • Provides powerful function to assist in object identification, motion tracking etc. • Virtually assist in any image processing application. • C-based program computer vision repository.
Milestones • Pass • commission the OpenCV system to load frames of video into memory. • convert the RGB frames to the HSV format. • threshold the HSV to identify the region of interest. • recombine the HSV image to return to RGB space. • Average • Find the centre point of the ball. • Good • Associate the detected centers.
Very Good • Implementation of Kalman Filter • Exceptional • Display of video, with overlays highlighting the position of the balls in each frames.
Progress! • Currently, I am at my Pass milestone • I’ve commissioned the OpenCV system to load frames of video into memory.(Completed) • Conversion from • RGB HSV • RGB HSV RGB (Completed) • RGB HSV Threshold RGB output to screen (Completed)
Things to do • Feb 5th – Feb 12th • Continuing with work on Finding the centre point of each ball • Feb 19th –Feb 26th • Complete the centre point association. • Feb 26th – Mar 5th • Introduction of the Kalman Filter.
References • http://opencv.willowgarage.com/wiki/ • http://opencv.jp/opencv-1.0.0_org/docs/ref/opencvref_highgui.htm • Image • http://farm4.static.flickr.com