Removing ambiguous image areas for improved certainty in object silhouette identification

摘要

A method for identifying ambiguous areas of an object image through analysis of an object silhouette comprises examining and comparing plural image pixels to corresponding plural pixels in an image selected from a library of images, and using such comparisons to remove ambiguous areas from consideration when determining the certainty level of object identification. Also independently claimed is a system for performing such a method, specifically applied to identifying object attitude, and additionally referring to sequentially comparing object image pixels to a library mask image representing an external structure, e.g. a refueling boom (Figures 4A-C) or building, not associated with the object and which partially obstructs the object. Ambiguous pixels may be eliminated from consideration in determining a fit score, where the fit score represents a percentage value indicative of a certainty of the matching process in identifying the attitude of the aircraft. Library mask images may be scaled for comparison.