Extraction of imperfection features through spectral analysis

• US 7,403,871 B2
• Filed: 05/02/2007
• Issued: 07/22/2008
• Est. Priority Date: 11/22/2004
• Status: Active Grant

• Current Assignee: Stylwan IP Holding LLC
• Original Assignee: Stylianos Papadimitriou, Wanda G. Papadimitriou
• Inventors: Papadimitriou, Stylianos, Papadimitriou, Wanda G.
• Primary Examiner(s): Wachsman; Hal D
• Application Number:
  US11/743,550
• Publication Number:
  US 20070219757A1
• Time in Patent Office: 447 Days
• Field of Search: 702/185, 702/34, 702/35, 702/109
• US Class Current: 702/185
• CPC Class Codes: G01N 27/9046 by analysing electrical signals
• CPC Class Codes: G01R 33/072 Constructional adaptation of the sensor to specific applications
• CPC Class Codes: G05B 19/41875 characterised by quality surveillance of production
• CPC Class Codes: G05B 2219/32196 Store audit, history of inspection, control and workpiece data into database
• CPC Class Codes: Y02P 80/40 Minimising material used in manufacturing processes
• CPC Class Codes: Y02P 90/02 Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Abstract

Autonomous non-destructive inspection equipment provides automatic and/or continuous inspection and evaluation of a material under inspection. The inspection equipment preferably comprises at least one detection sensor and at least one detection sensor interface for a computer. The autonomous non-destructive inspection system may also be retrofitted into conventional inspection systems by extracting pertinent features through spectral frequency analysis and sensor compensation and utilizing those features in the autonomous non-destructive inspection system.

Claims

1. An inspection system to detect imperfections in materials comprising:

at least one imperfection detection sensor for producing an imperfection signal;

a bank of filters comprising a plurality of programmable filters arranged in parallel such that said imperfection signal is an input for each of said plurality of programmable filters, said plurality of programmable filters comprising a plurality of filter outputs;

at least one processor; and

at least one memory storage associated with said at least one processor, said at least one memory storage storing a plurality of processing coefficients, said at least one memory storage also storing at least one program comprising instructions for said at least one processor to utilize said plurality of filter outputs to provide a plurality of variables for use with said plurality of processing coefficients within at least one equation for detecting said imperfections.

2. The inspection system of claim 1, wherein at least one of said plurality of programmable filters is a low pass filter wherein at least one of said plurality of filter outputs comprises a DC frequency component of said imperfection signal.

3. The inspection system of claim 1, wherein said bank of filters comprise at least one digital filter.

4. A method to detect imperfections in materials being inspected comprising:

inducing an excitation into a material;

producing an imperfection signal responsively to said step of inducing said excitation;

applying said imperfection signal to a plurality of programmable filters, said plurality of programmable filters being arranged in parallel such that said imperfection signal is an input for each of said plurality of programmable filters;

storing a plurality of processing coefficients; and

utilizing a plurality of filter outputs from said plurality of programmable filters to provide a plurality of variables in combination with said plurality of processing coefficients within at least one equation for detecting said imperfections.

5. The method of claim 4, further comprising varying a bandpass of at least one of said plurality of programmable filters responsively to a scanning speed.

6. The method of claim 4, further comprising utilizing digital filters for at least one of said plurality of programmable filters.

7. The method of claim 4, further comprising programming a gain of at least one of said plurality of programmable filters.

8. The method of claim 4, further comprising the step of at least partially aligning in time said plurality of filter outputs.