METHOD, SYSTEM, AND NON-TRANSITORY COMPUTER READABLE MEDIUM FOR VIDEO-BASED CIRCULAR OBJECT LOCALIZATION

摘要

A method and a system for localizing at least one circular object in a video frame captured by an image capture unit are disclosed. The method includes the steps of: A. performing edge detection to obtain a plurality of image edge points in the video frame; B. extracting a contour set of image contours by linking neighboring edge points; C. generating a plurality of circle samples from the extracted contour set by a structural sampling process; and D. localizing the at least one circular object from the generated circle samples by calculating a respective fitness score for each of the circle samples to measure the supporting evidence of the image edge points on a neighborhood of each of the circle samples, and by choosing the circle sample(s) whose fitness score(s) is/are greater than a threshold value as the circular object(s) found in the video frame.

主权项

1. A method executed on a computer system for localizing at least one circular object in a video frame captured by an image capture unit, the method comprising the steps of:
A. performing edge detection to obtain a plurality of image edge points in the video frame; B. extracting a contour set Φ of image contours φ by linking neighboring image edge points as said image contours Φ, each of which has a contour structure represented by a list of said neighboring image edge points arranged in a sequential order; C. generating a plurality of circle samples from the extracted contour set Φ by a structural sampling process that subsamples the image contours φ in the extracted contour set Φ, with respect to their contour structures, to obtain a plurality of point sets and to generate the circle samples with center coordinates and radii for the point sets, wherein each of the point sets contains at least three said image edge points; and D. localizing the at least one circular object from the generated circle samples by calculating a respective fitness score for each of the circle samples to measure the supporting evidence of the image edge points on a neighborhood of the circumference of each of the circle samples, and by choosing at least one of the circle sample(s) whose fitness score(s) is/are greater than a threshold value as the at least one circular object(s) found in the video frame.