Method of recognizing patterns

• US 4,901,362 A
• Filed: 08/08/1988
• Issued: 02/13/1990
• Est. Priority Date: 08/08/1988
• Status: Expired due to Fees

• Current Assignee: Raytheon Company Lexington Massachusetts 02173
• Original Assignee: Raytheon Company (Rtx Corporation)
• Inventors: Terzian, John
• Primary Examiner(s): Boudreau, Leo H.
• Assistant Examiner(s): Cammarata, Michael
• Application Number:
  US07/229,395
• Time in Patent Office: 554 Days
• Field of Search: 382/19, 382/21, 382/22, 382/16, 382/28, 382/20, 382/23, 382/26, 382/1, 382/34
• US Class Current: 382/197
• CPC Class Codes: G06V 10/421 by analysing segments intersecting the pattern

Abstract

A method of identifying an object from a digitized image is disclosed. One embodiment of the method is particularly applicable to the classification of ships and a second embodiment is particularly applicable to the classification of aircraft. The digitized image of an unknown object is represented as a set of vertical vectors or, in the second embodiment, a set of radial vectors. The vectors are normalized and then compared to a library containing normalized digitized images of known objects. The unknown object is thereby identified as being similar to the known object corresponding to the library image most closely matching the image of the unknown object.

Claims

1. A method of identifying an unknown object from a digitized image of the silhouette of the object by comparing the digitized image to each one of a plurality of images of the silhouettes of known objects, comprising the steps of:

(a) representing the digitized image of the silhouette of the unknown object as a set of radial vectors;

(b) forming a normalized set of radial vectors by dividing each radial vector in the set by a scale factor; and

(c) rotating the set of radial vectors until the largest radial vector is the first vector in the set.

2. A method of identifying an unknown object as in claim 1 further comprising the steps of:

(a) comparing the rotated, normalized set of radial vectors to each one of a plurality of normalized, rotated sets of radial vectors representing digitized images of known objects to compute a score for each one of the comparisons, such score equaling the summation of the absolute values of the differences between each radial vector in the unknown normalize rotated silhouettes and the corresponding radial vector in each one of the known normalized, rotated silhouettes.

3. A method of identifying an unknown object as in claim 2 furthering the steps of:

(a) identifying a peak radial vector in the set of normalized radial vectors;

(b) rotating set of normalized until the peak radial vector is the first radial vector in the set; and

(c) comparing the rotated, normalized set of radial vectors to each one of a plurality of normalized, rotated sets of radial vectors representing digitized images of known objects to compute a score for each one of the comparisons, such score equalizing the summation of the absolute values of the differences between each radial vector in the unknown normalized rotated silhouettes and the corresponding radial vector in each one of the known normalized rotated silhouettes.

4. A method of identifying an unknown object as in claim 3 wherein plurality of digitized images representing known objects includes silhouettes formed by airplanes at various aspect angles.