Autonomous Identification of Vertical Lines in Video Images to Assist in UGV Cover and Concealment

摘要

The autonomous identification of appropriate locations that provide cover and concealment for a robotic vehicle is a difficult task. Fortunately, many appropriate locations are man-made and are characterized by a distinctive geometric characteristic, specifically, object edges that form vertical straight lines. The object of this report is to propose, implement, and assess an algorithm for the autonomous identification of vertical lines in video images that offer the most appropriate location for cover and concealment. Results obtained from application of the algorithm to a number of video sequences indicate that the algorithm is effective in identifying acceptable locations. Several areas for improvement of the algorithm are suggested and discussed.