Vertical vector pattern recognition algorithm

• US 5,040,231 A
• Filed: 10/17/1988
• Issued: 08/13/1991
• Est. Priority Date: 09/30/1987
• Status: Expired due to Fees

• Current Assignee: Raytheon Company (Rtx Corporation)
• Original Assignee: Raytheon Company (Rtx Corporation)
• Inventors: Terzian, John
• Primary Examiner(s): Moore, David K.
• Assistant Examiner(s): COUSO, YON JUNG
• Application Number:
  US07/258,915
• Time in Patent Office: 1,030 Days
• Field of Search: 382/21, 382/22, 382/25, 382/30, 382/34, 382/28, 382/47
• 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. The method is particularly applicable to the classification of ships. The digitized image of an unknown ship is represented as a set of vertical vectors. The vertical vectors are normalized and then compared to a library containing normalized digitized images of known ships. The unknown ship is thereby identified as being similar to the known ship corresponding to the library image most closely matching the image of the unknown ship.

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) selecting pixels in the digitized image representing the outline of the unknown object;

(b) representing the digitized image of the unknown object as a set of vertical vectors, each vertical vector being a digital word with a predetermined number of bits, said digital word representing a numeric value proportional to the distance in one direction between two pixels selected to represent the outline; and

(c) forming a normalized silhouette by normalizing the width and the area of the silhouette as represented by the set of vertical vectors.

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

(a) comparing the normalized silhouette to each one of a plurality of known normalized silhouettes formed from sets of vertical vectors representing digitized images of silhouettes of known objects to compute a score for each one of the comparisons, such score by summing the absolute values of the differences between the value of each vertical vector in the unknown normalized silhouette and the value of the corresponding vector in each one of the known normalized silhouettes.

3. A method of identifying an unknown object as in claim 2 further comprising the step of identifying the unknown object as being the same class as the known object from which the known normalized silhouette corresponding to the lowest score computed was formed.

4. A method of identifying an unknown ship as in claim 3 wherein the plurality of known normalized silhouettes includes silhouettes formed by ships at various aspect angles.

5. The method of identifying an unknown object as in claim 2 wherein each vertical vector is represented by a digital word having at least two bits.

6. The method of identifying an unknown object of claim 1 wherein the step of forming a normalized silhouette comprises normalizing the width of the silhouette by selecting a predetermined number of vertical vectors.