Method and device for recognizing a target

• US 5,068,664 A
• Filed: 10/24/1990
• Issued: 11/26/1991
• Est. Priority Date: 10/24/1989
• Status: Expired due to Fees

• Current Assignee: Aerospatiale Societe Nationale Industrielle
• Original Assignee: Aerospatiale Societe Nationale Industrielle
• Inventors: Coulombeix, Colette, Barthelemy, Regis, Appriou, Alain
• Primary Examiner(s): Barron, Jr., Gilberto
• Application Number:
  US07/602,488
• Time in Patent Office: 398 Days
• Field of Search: 342/90, 342/195, 342/192, 342/13, 382/17, 382/39
• US Class Current: 342/90
• CPC Class Codes: G01S 7/2923 based on data belonging to a number of consecutive radar periods
• CPC Class Codes: G01S 7/295 Means for transforming co-ordinates or for evaluating data, e.g. using computers
• CPC Class Codes: G01S 7/412 based on a comparison between measured values and known or stored values
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes

Abstract

A method and device are disclosed for recognizing a target (T) among a plurality of n known targets Ti (with i=1, 2, 3, . . . , n).

According to the invention, by direct examination of said target (T) to be recognized and from first values (Cki) representative of a plurality of p recognition criteria Ck (with k=1, 2, 3, . . . , p), at least p third values (pCk(Ti)) are established representative of the probabilities that, for each recognition criterion (Ck), the target (T) to be recognized is each of said known targets (Ti); and said third values associated with the same known target (Ti) are merged while taking into account second values (PCk(Ti)) representative of the probabilities that a target (Ti) is recognized, in so far as criterion Ck is concerned, when it is examined so as to obtain n fourth values (p(Ti)), each of which is representative of the probability that the target to be recognized is one of the known targets (Ti).

Claims

1. Method for recognizing a target (T) among a plurality of n known targets Ti (with i=1,3,4, . . . , n), wherein:

in a first preparatory step, for each of said known targets (Ti) first values (CKi) are determined representative of a plurality of p recognition criteria Ck (with k=1,2,3, . . . , p);

in a second preparatory step, from direct examination of said n known targets (Ti), using said first values (CKi) and for each recognition criterion (Ck), nxp second values (PCk(Ti)) are established each representative of the probability for a target (Ti) to be recognized when it is examined whereas it is the target (Tj) which is examined (j=1,2,3, . . . , n); and

in a step for recognizing said target;

by direct examination of said target (T) to be recognized and from said first values (CKi) determined by said first preparatory step, at least p third values (pCk(Ti)) are established representative of the probabilities, for each recognition criterion (Ck), that the target (T) to be recognized is each of said n known target (Ti) insofar as each of said p criteria is concerned;

said nxp third values associated with the same known target (Ti) are merged together by taking into account said nxp second values, so as to obatin n fourth values (p(Ti)), each of which is representative of the probability that the target to be recognized is one of the known targets (Ti); and

the identity of the target to be recognized (T) is determined from the largest of said fourth values.

2. The method as claimed in claim 1, wherein merging of said third values is obtained by using mathematical laws given by any one of the theories which are BAYES inference, entropy maximum, the theory of evidence or the theory of fuzzy sets.

3. The method as claimed in claim 1, wherein said third values are certain values.

4. The method as claimed in claim 1, wherein said third values are uncertain values.

5. The method as claimed in claim 1, wherein, in said first preparatory step, said first values are recorded criterion by criterion for all said known targets (Ti), so that all said first values of all said targets corresponding to a criterion are grouped together.

6. The method as claimed in claim 1, wherein determination of said first values is achieved by direct examination of said known targets (Ti).

7. The method as claimed in claim 1, wherein determination of said first values is achieved from models or plans of said known targets.

8. The method as claimed in claim 1, wherein, in said second preparatory step, said second values are obtained by comparison of said first values with measurement of each of said targets in accordance with each of said criteria, said measurements being repeated numerous times.

9. The method as claimed in claim 1, wherein said second values are recorded in matrix memories (MC1 to MCp), each matrix memory being associated with a criterion and in each recording position (10.ji) of such a matrix memory the probability is recorded, for said criterion, that the target (Ti) is recognized whereas the target (Tj) is subjected to direct examination.