Automated learning of model classifications

• US 7,889,914 B2
• Filed: 06/05/2009
• Issued: 02/15/2011
• Est. Priority Date: 06/16/2003
• Status: Expired due to Fees

• Current Assignee: Drexel University
• Original Assignee: Drexel University
• Inventors: Regli, William C., Ip, Cheuk Yiu, Sieger, Leonard
• Primary Examiner(s): Chawan; Sheela C
• Application Number:
  US12/479,414
• Publication Number:
  US 20090319454A1
• Time in Patent Office: 620 Days
• Field of Search: 382/159, 382/228, 382/224, 382/225, 382/118, 382/190, 706/25, 706/45, 706/15, 706/20, 706/934, 706/13, 706/47, 706/59, 706/12, 707/102, 707/3, 707/E17.003, 707/100, 707/101, 707/6, 702/19, 702/20, 705/3, 705/10, 705/17, 704/245, 704/9, 704/239, 704/246, 704/E15.014, 704/E17.003, 704/230, 704/240, 704/256, 704/232, 704/244, 704/243, 725/46, 725/9
• US Class Current: 382/159
• CPC Class Codes: G06F 18/00 Pattern recognition
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 20/10 using kernel methods, e.g. support vector machines [SVM]
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching

Abstract

A method of providing an automated classifier for 3D CAD models wherein the method provides an algorithm for learning new classifications. The method enables existing model comparison algorithms to adapt to different classifications that are relevant in many engineering applications. This ability to adapt to different classifications allows greater flexibility in data searching and data mining of engineering data.

Claims

1. A method of classifying models comprising the steps of:

providing a processor, wherein the processor performs the steps of;

determining a first set of values based on predetermined properties of training models;

defining classifications based upon said first set of values;

determining a second set of values based on predetermined properties of a query model;

comparing said second set of values to said classifications; and

determining a classification of said query model based on said comparing step.

2. The method of claim 1, wherein said predetermined properties of said training models or said query models comprise a first set of distances between points selected on said training models or said query model;

wherein said first set of distances comprises a set of predetermined shape interrogation measures.

3. The method of claim 2, wherein said predetermined shape interrogation measures include IN distances, OUT distances, and MIXED distances measured at point pairs placed on a surface of said training models or said query model to maximize discrimination.

4. The method of claim 3, wherein said point pairs are randomly selected.

5. The method of claim 3, wherein said point pairs are selected according to a predetermined grid function.

6. The method of claim 1, wherein said step of defining classifications further comprises weighting said first set of values;

wherein said weighting step further comprises normalizing said first set of values.

7. The method of claim 6, wherein said classifications for said training models comprise shapes and functionality.

8. The method of claim 6, wherein said classifications comprise manufacturing processes to build said training models.

9. The method of claim 1, wherein said query model and said training models are models of 3D objects.

10. A method of searching for a model comprising the steps of:

using a processor, wherein the processor performs the steps of;

determining a first set of values based on predetermined properties of a query model;

comparing said first set of values to definitions for classification, wherein said definitions for said classification are based on a second set of values that is determined by predetermined properties for a group of training models; and

determining a classification of said query model based on said comparing step.

11. The method of claim 10, wherein said predetermined properties of said training models or said query model comprise a first set of distances between points selected on said training models and said query model, wherein said first set of distances comprises a set of predetermined shape interrogation measures.

12. The method of claim 11, wherein said predetermined shape interrogation measures included IN distances, OUT distances, and MIXED distances;

wherein said IN distances, OUT distances, and MIXED distances are measured at point pairs placed on a surface of a training model or said query model to maximize discrimination.

13. The method of claim 12, wherein said point pairs are randomly selected.

14. The method of claim 12, wherein said point pairs are selected according to a predetermined grid function.

15. The method of claim 10, wherein the step of determining a classification comprises a step of weighting said second set of values, wherein said weighting step comprises normalizing said first set of values.

16. The method of claim 15, wherein said definitions for classification comprise shapes and functionality.

17. The method of claim 15, wherein said definitions for classification comprise manufacturing processes to build said training models.

18. The method of claim 10, wherein said query model and said training models are models of 3D objects.

19. A system for classifying models comprising:

a database for storing a plurality of training models and classification definitions, wherein said classification definitions are based upon a first set of values determined by predetermined properties of said training models; and

a host unit connected to said database for receiving a query model, and determining a second set of values based on said predetermined properties of said query model, said host determining a classification of said query model by comparing said second set of values to said classification definitions.

20. The system of claim 19, wherein said query model and said training models are models of 3D objects.