Parameter normalized features for classification procedures, systems and methods

• US 5,263,097 A
• Filed: 07/24/1991
• Issued: 11/16/1993
• Est. Priority Date: 07/24/1991
• Status: Expired due to Term

• Current Assignee: Texas Instruments, Inc.
• Original Assignee: Texas Instruments, Inc.
• Inventors: Gately, Michael T., Katz, Alan J.
• Primary Examiner(s): Mancuso, Joseph
• Application Number:
  US07/735,403
• Time in Patent Office: 846 Days
• Field of Search: 382/14, 382/39, 382/15, 382/28, 382/48, 382/36, 395/20, 395/21, 395/62
• US Class Current: 382/190
• CPC Class Codes: G01S 3/7865 using correlation of the live video image with a stored image
• CPC Class Codes: G01S 7/417 involving the use of neural networks
• CPC Class Codes: G06F 18/15 Statistical pre-processing, e.g. techniques for normalisation or restoring missing data
• CPC Class Codes: G06F 18/213 Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
• CPC Class Codes: G06V 10/32 Normalisation of the pattern dimensions
• CPC Class Codes: G10L 15/02 Feature extraction for speech recognition; Selection of recognition unit

Abstract

This application discloses a method for distinguishing targets from clutter, comprising the steps of inputting data, extracting feature information from said data, extracting parameter information from said data, normalizing said features with said parameter information to produce parameter normalized feature information, inputting said parameter normalized feature information into the classification procedure, operating said classification procedure on said parameter normalized feature information, and outputting target and clutter data. Classification systems are also disclosed.

Claims

1. A method for distinguishing targets from clutter, comprising the steps of

(a) inputting data containing target information and background information and preprocessing said data to highlight selected information to ultimately improve target recognition;

(b) extracting feature information from said data;

(c) extracting parameter information from said data;

(d) normalizing said feature information with said parameter information to produce parameter normalized feature information;

(e) inputting said parameter normalized feature information into an appropriate classification procedure;

(f) operating said appropriate classification procedure on said parameter normalized feature information; and

(g) outputting said target information and said background information.

2. The method of claim 1, wherein said data is a format selected from the group consisting of one-dimensional, two-dimensional, and multi-dimensional or a combination thereof.

3. The method of claim 1, wherein said data is a format selected from the group consisting of digitized imagery, speech signals, and radar signals or a combination thereof.

4. The method of claim 1, wherein parallel architectures are used to select the appropriate classification procedure.

5. The method of claim 4, wherein said parallel architectures comprise neural networks.

6. The method of claim 1, wherein parallel architectures are used to implement said appropriate classification procedure.

7. The method of claim 6, wherein said parallel architectures comprise neural networks.

8. The method of claim 1, wherein said feature information is a format selected from the group consisting of contrast-based and texture-based or a combination thereof.

9. The method of claim 1, wherein said appropriate classification procedure distinguishes a plurality of classes from one another.

10. The method of claim 1, further comprising the step of inputting parameter information after step (c).

11. The method for distinguishing targets from clutter of claim 1, wherein said selected information highlighted by said preprocessing is selected from the group consisting of feature information and parameter information.

12. The method for distinguishing targets from clutter of claim 1, wherein said selected information highlighted by said preprocessing is used to select said appropriate classification procedure.

13. A classification system, comprising

(a) a preprocessor to highlight selected information of a data input containing target information and background information to ultimately improve target recognition;

(b) a feature extractor to extract feature information from said data input;

(c) a parameter extractor procedures to extract parameter information from said data input;

(d) a normalizer that normalizes said feature information with said parameter information to produce parameter normalized feature information; and

(e) a classifier that uses said parameter normalized feature information to distinguish target information from said background information and output said target information and background information.

14. The classification system of claim 13, wherein said data input is a format selected from the group consisting of one-dimensional, two-dimensional, and multidimensional.

15. The classification system of claim 13, wherein said data input a format selected from the group consisting of is digitized imagery, speech signals, and radar signals.

16. The classification system of claim 13, wherein parallel architectures are used to implement said switching procedures.

17. The classification system of claim 13, wherein parallel architectures are used to implement said classification procedures.

18. The classification system of claim 13, wherein said feature information is a format selected from the group consisting of contrast-based and texture-based.

19. A classification hardware system, comprising

(a) a preprocessor to highlight selected information of a data input containing target information and background information to ultimately improve target recognition;

(b) a feature extractor to extract feature information from said data input;

(c) a parameter extractor to extract parameter information from said data input;

(d) a normalizer procedures that normalizes said feature information with said parameter information to produce parameter normalized feature information; and

(e) a classifier that use said parameter normalized feature information to distinguish said target information from said background information and output said target information and said background information.

20. A classification software system, comprising

(a) a preprocessor to highlight selected information of a data input containing target information and background information to ultimately improve target recognition;

(b) a feature extractor to extract feature information from said data input;

(c) a parameter extractor to extract parameter information from said data input;

(d) a normalizer procedures that normalizes said feature information with said parameter information to produce parameter normalized feature information; and

(e) a classifier that use said parameter normalized feature information to distinguish said target information from said background information and output said target information and said background information.

21. A classification system, comprising

(a) a preprocessor to diminish environmental effects of a data input containing target information and background information to ultimately improve target recognition;

(b) a feature extractor to extract feature information from said data input;

(c) a parameter extractor procedures to extract parameter information from said data input;

(d) a normalizer that normalizes said feature information with said parameter information to produce parameter normalized feature information; and

(e) a classifier that uses said parameter normalized feature information to distinguish target information from said background information and output said target information and background information.

22. The classification system of claim 21, wherein said environmental effects are selected from the group consisting of sensor effects and climatic effects.

23. The classification system of claim 21, wherein said preprocessor also detects environmental effects which are used to select said classifier.

24. A classification system, comprising

(a) a preprocessor to diminish environmental effects of a data input containing target information and background information to ultimately improve target recognition;

(b) a feature extractor to extract feature information from said data input;

(c) a parameter extractor procedures to extract parameter information from said data input; and

(d) a classifier that uses said parameter information and said feature information to distinguish target information from said background information and output said target information and background information.

25. The classification system of claim 24, wherein said preprocessor also detects environmental effects which are used to select said classifier.

26. A classification system, comprising

(a) a preprocessor to highlight selected information of a data input containing target information and background information to ultimately improve target recognition;

(b) a feature extractor to extract feature information from said data input;

(c) a parameter extractor procedures to extract parameter information from said data input; and

(e) a classifier that uses said parameter information and said feature information to distinguish target information from said background information and output said target information and background information.

27. The method for distinguishing targets from clutter of claim 26, wherein said selected information highlighted by said preprocessing is used to select said classifier.