Illumination compensation system for industrial inspection

• US 6,445,812 B1
• Filed: 01/22/1999
• Issued: 09/03/2002
• Est. Priority Date: 01/22/1999
• Status: Expired due to Fees

• Current Assignee: Siemens Corp. (Siemens AG)
• Original Assignee: Siemens Corporate Research Incorporated (Siemens AG)
• Inventors: Fang, Ming, Lai, Shang-Hong
• Primary Examiner(s): Johns, Andrew W.
• Assistant Examiner(s): AZARIAN, SEYED H
• Application Number:
  US09/235,997
• Time in Patent Office: 1,320 Days
• Field of Search: 382/141, 382/142, 382/143, 382/144, 382/145, 382/146, 382/147, 382/190, 382/110, 382/164
• US Class Current: 382/141
• CPC Class Codes: G06T 2207/10152 Varying illumination
• CPC Class Codes: G06T 2207/20064 Wavelet transform [DWT]
• CPC Class Codes: G06T 2207/30164 Workpiece; Machine component
• CPC Class Codes: G06T 5/94 based on local image properties, e.g. for local contrast enhancement
• CPC Class Codes: G06T 7/001 using an image reference approach
• CPC Class Codes: G06V 10/145 Illumination specially adapted for pattern recognition, e.g. using gratings

Abstract

An illumination compensation system for correcting smooth intensity variations due to illumination changes is based on an assumption that an underlining image reflectance function is approximately a piecewise constant and that an image irradiance function is spatially smooth. The system first takes the logarithm of an image brightness function. Gradient constraints are then computed using a finite difference. Reliable gradient constraints are selected based on a local uniformity test. A process is subsequently applied to estimate the logarithmic irradiance function. A logarithmic irradiance function is subtracted from the logarithmic image brightness function and an exponential operation of the above subtracted image function is taken and an illumination compensated image is outputted from the system.

Claims

1. A method for compensating for variations of illumination in an automatic visual inspection system, the method comprising the steps of:

receiving image data;

estimating an illumination component based on the image data;

removing the estimated illumination component from the image data to provide illumination-compensated image data; and

outputting the illumination-compensated image data to the visual inspection system for further processing,

wherein the step of estimating an illumination component comprises the steps of;

computing an image brightness function representing the image data, the image brightness function being the product of an irradiance function and a reflectance function; and

estimating the irradiance function from the image data; and

wherein the step of removing the estimated illumination component comprises the steps of;

computing the reflectance function using the estimated irradiance function and the computed image brightness function, the computed reflectance function representing the illumination-compensated image data.

2. The method of claim 1, wherein the step of estimating the irradiance function comprises the steps of:

performing a logarithmic operation on the image brightness function to provide a logarithmic image brightness function;

deriving gradient constraints at locations inside regions of uniform reflectance values of the reflectance function using the logarithmic image brightness function, the reflectance function being approximated as a piecewise constant; and

estimating a logarithmic irradiance function from selected ones of the gradient constraints using a surface fitting process; and

wherein the step of computing the reflectance function comprises the steps of;

subtracting the logarithmic irradiance function from the logarithmic image brightness function to provide a logarithmic reflectance function; and

performing an exponential operation on the logarithmic reflectance function to obtain the illumination-compensated image data.

3. The method of claim 2, further comprising the step of selecting reliable gradient constraints at the locations inside regions of uniform reflectance.

4. The method of claim 3, wherein the step of selecting reliable gradient constraints comprises the step of using a local uniformity test that determines if pixels in a local neighborhood belong to a same region with a uniform reflectance value.

5. The method of claim 3, wherein the surface fitting process estimates the logarithmic irradiance function by utilizing the selected reliable gradient constraints and considering the logarithmic irradiance function as a spatially smooth function.

6. The method of claim 5, wherein the surface fitting method is one of a spline surface fitting method and a polynomial surface fitting method in which the irradiance function is approximated using a low order polynomial.

7. A system for compensating for variations of illumination in an automatic visual inspection system, comprising:

an illumination estimator adapted to estimate an illumination component of received image data; and

a compensator adapted to remove the estimated illumination component from the image data to provide illumination-compensated image data, wherein the illumination-compensated image data is output to the visual inspection system for further processing,

wherein the illumination estimator comprises;

means for computing an image brightness function representing the image data, the image brightness function being the product of an irradiance function and a reflectance function; and

means for estimating the irradiance function of the image brightness function from the image data; and

wherein the compensator comprises;

means for computing the reflectance function using the estimated irradiance function and the computed image brightness function, the computed reflectance function representing the illumination-compensated image data.

8. The system of claim 7, wherein the automatic visual inspection system is configured as an image reference inspection system.

9. The system of claim 7, wherein the means for estimating the irradiance function comprises:

means for performing a logarithmic operation on the image brightness function to provide a logarithmic image brightness function;

means for deriving gradient constraints at locations inside regions of uniform reflectance values of the reflectance function using the logarithmic image brightness function, the reflectance function being approximated as a piecewise constant; and

means for estimating a logarithmic irradiance function from selected ones of the gradient constraints using a surface fitting process; and

wherein the means for computing the reflectance function comprises;

means for subtracting the logarithmic irradiance function from the logarithmic image brightness function to provide a logarithmic reflectance function; and

means for performing an exponential operation on the logarithmic reflectance function to obtain the illumination-compensated image data.

10. The system of claim 9, wherein the means for estimating the irradiance function further comprises means for selecting reliable gradient constraints at the locations inside regions of uniform reflectance.

11. The system of claim 10, wherein the means for selecting reliable gradient constraints implements a local uniformity test to determine if pixels in a local neighborhood belong to a same region with a uniform reflectance value.

12. The system of claim 10, wherein the means for estimating the logarithmic irradiance function implements the surface fitting method using the reliable gradient constraints, and considers the logarithmic irradiance function as a spatially smooth function for the estimation process.

13. The system of claim 12, wherein the surface fitting method is one of a spline surface fitting method and a polynomial surface fitting method in which the irradiance function is approximated using a low order polynomial.

14. A method for improving automatic visual inspection performance under varying illumination conditions, comprising the steps of:

preprocessing an inspection image to provide an illumination-compensated inspection image, wherein the step of preprocessing the inspection image comprises the steps of estimating an illumination component of the inspection image, and removing the estimated illumination component from the inspection image to provide the illumination-compensated inspection image;

aligning the illumination-compensated image with a corresponding reference image; and

taking the difference between the illumination-compensated image and the reference image to provide a difference image.

15. The method of claim 14, further including the step of analyzing the difference image to determine if there are defects in an object of the inspection image.

16. The method of claim 14, wherein the step of estimating an illumination component comprises the steps of:

computing an image brightness function representing the inspection image, the image brightness function being the product of an irradiance function and a reflectance function; and

estimating the irradiance function from the inspection image; and

wherein the step of removing the estimated illumination component comprises the steps of;

computing the reflectance function using the estimated irradiance function and the computed image brightness function, the computed reflectance function representing the illumination-compensated inspection image.

17. A program storage device readable by a machine, tangibly embodying a program of instructions executable by the machine to perform method steps for compensating for variations of illumination in an automatic visual inspection system, the method steps comprising:

receiving image data;

estimating an illumination component based on the image data;

removing the estimated illumination component from the image data to provide illumination-compensated image data; and

outputting the illumination-compensated image data to the visual inspection system for further processing,

wherein the step of estimating an illumination component comprises the steps of;

computing an image brightness function representing the image data, the image brightness function being the product of an irradiance function and a reflectance function; and

estimating the irradiance function from the image data; and

wherein the step of removing the estimated illumination component comprises the steps of;

computing the reflectance function using the estimated irradiance function and the computed image brightness function, the computed reflectance function representing the illumination-compensated image data.

18. The program storage device of claim 17, wherein the instructions for performing the step of estimating the irradiance function comprise instructions for performing the steps of:

performing a logarithmic operation on the image brightness function to provide a logarithmic image brightness function;

deriving gradient constraints at locations inside regions of uniform reflectance values of the reflectance function using the logarithmic image brightness function, the reflectance function being approximated as a piecewise constant; and

estimating a logarithmic irradiance function from selected ones of the gradient constraints using a surface fitting process; and

wherein the step of computing the reflectance function comprises the steps of;

subtracting the logarithmic irradiance function from the logarithmic image brightness function to provide a logarithmic reflectance function; and

performing an exponential operation on the logarithmic reflectance function to obtain the illumination-compensated image data.

19. A program storage device readable by a machine, tangibly embodying a program of instructions executable by the machine to perform method steps for improving automatic visual inspection performance under varying illumination conditions, the method steps comprising:

preprocessing an inspection image to provide an illumination-compensated inspection image, wherein the step of preprocessing the inspection image comprises the steps of estimating an illumination component of the inspection image, and removing the estimated illumination component from the inspection image to provide the illumination-compensated inspection image;

aligning the illumination-compensated image with a corresponding reference image; and

taking the difference between the illumination-compensated image and the reference image to provide a difference image.

20. The program storage device of claim 19, wherein the instructions for performing the step of estimating an illumination component comprise instructions for performing the steps of:

computing an image brightness function representing the inspection image, the image brightness function being the product of an irradiance function and a reflectance function; and

estimating the irradiance function from the inspection image; and

wherein the step of removing the estimated illumination component comprises the steps of;

computing the reflectance function using the estimated irradiance function and the computed image brightness function, the computed reflectance function representing the illumination-compensated inspection image.

21. The program storage device of claim 20, wherein the instructions for performing the step of estimating the irradiance function comprise instructions for performing the steps of:

performing a logarithmic operation on the image brightness function to provide a logarithmic image brightness function;

deriving gradient constraints at locations inside regions of uniform reflectance values of the reflectance function using the logarithmic image brightness function, the reflectance function being approximated as a piecewise constant; and

estimating a logarithmic irradiance function from selected ones of the gradient constraints using a surface fitting process; and

wherein the step of computing the reflectance function comprises the steps of;

subtracting the logarithmic irradiance function from the logarithmic image brightness function to provide a logarithmic reflectance function; and

performing an exponential operation on the logarithmic reflectance function to obtain the illumination-compensated image data.