Abstract: This paper describes the principles and implementation of scene matching used in position location systems in brief. Several essential problems in scene matching technique are reviewedt measures of similarity （MAD, MSD, PROD, NPROD, Pair Functions etc. ）are outlined; fast search me-thods（resolution from coarse to fine, quantization from coarse to fine, feature extraction, modified SSDA etc. ）are introduced; probability of acquisition for max. or min. algorithm is given; preprocessing of images （filtering, restoration, whitening etc. ）is considered; geometrical distortion（sync-hronization error, rotation, scale factor, perspective distortion）and its effects are analyzed, and the methods for overcoming such effects are presented.After then, two noticeable important techniques for homing seeker, the feature matching and the correlation tracking, are remarked. Several methods of feature extraction are introduced for such features in scene matching as invariant moments, edge features, line features, planar features, vertex features etc. . The emphasis is put on feature selection, update and replacement in terminal target homing.Finally, the block diagrams of a combined optimum aided navigation system and a homing seeker using correlation tracking are given.