Improved line detection algorithm for locating road lane markings

An improved line detection algorithm for locating road lane markings is described. After scanning images to find mid-points of potential lane markings, they are passed to the line detection algorithm. This makes multiple hypotheses of lines based on random pairs of potential line features and selects the one having the greatest support. After finding each line the algorithm deletes not only the inliers within a `fit tolerance` distance d_f but also the innermost outliers lying within a `delete tolerance` distance d_d. It is found that as d_d - d_f is increased the final result becomes more robust because elimination of remanent points near each line prevents them from confusing the algorithm when searching for further lines.