Methods for digitally noise averaging and illumination equalizing fingerprint images

• US 4,811,414 A
• Filed: 02/27/1987
• Issued: 03/07/1989
• Est. Priority Date: 02/27/1987
• Status: Expired due to Fees

• Current Assignee: Digital Biometrics, Inc. (Advent International Corporation)
• Original Assignee: CFA Technologies Incorporated
• Inventors: Fishbine, Glenn M., Ransom, Mark S., Germann, Daniel E.
• Primary Examiner(s): Boudreau, Leo H.
• Assistant Examiner(s): Daley, Donald J.
• Application Number:
  US07/020,331
• Time in Patent Office: 739 Days
• Field of Search: 382/4, 382/5, 382/52, 382/54, 358/163, 358/166
• US Class Current: 382/272
• CPC Class Codes: A61B 5/1172 using fingerprinting
• CPC Class Codes: G06T 5/20 using local operators
• CPC Class Codes: G06T 5/94 based on local image properties, e.g. for local contrast enhancement
• CPC Class Codes: G06V 40/12 Fingerprints or palmprints

Abstract

Methods for operating a programmable digital computer to enhance images of fingerprints represented by an array of pixel values. Methods for noise averaging, illumination equalizing, directional filtering, unhairing, curvature correcting, and scale correcting a fingerprint image are disclosed.

Claims

1. A method for operating programmable computing means to illumination equalize input pixel values of an array of input pixel values characteristic of a fingerprint image so as to produce an illumination equalized array of pixel values characteristic of an illumination equalized image, including:

receiving an array of input pixel values characteristic of a fingerprint image;

selecting equalizing subarrays of input pixel values, each including an input pixel value to be illumination equalized;

generating subarray average values as a function of the pixel values within the equalizing subarrays;

subtracting the subarray average values from the corresponding pixel values being equalized to generate pixel difference values;

adding a predetermined constant to the pixel difference values to generate intermediate illumination equalized pixel values;

setting the illumination equalized pixel values equal to a predetermined minimum pixel value if the corresponding intermediate illumination equalized pixel values are less than the minimum pixel value;

setting the illumination equalized pixel values equal to the corresponding intermediate illumination equalized pixel values if the corresponding intermediate illumination equalized pixel values are greater than or equal to the minimum pixel value, and less than or equal to a predetermined maximum pixel value;

setting the illumination equalized pixel values equal to the maximum pixel value if the corresponding intermediate illumination equalized pixel values are greater than the maximum pixel value; and

storing the illumination equalized pixel values as an array of illumination equalized pixel values characteristic of the fingerprint image.

2. The method of claim 1 wherein selecting equalizing subarrays comprises selecting equalizing subarrays, each including an input pixel value to be illumination equalized and a plurality of input pixel values adjacent to and surrounding the pixel value to be illumination equalized.

3. The method of claim 2 wherein selecting equalizing subarrays comprises selecting eight-by-eight subarrays of input pixel values.

4. The method of claim 1 wherein generating subarray average values includes, for each subarray, summing the pixel values of the subarray and dividing the sum by the number of pixel values summed.

5. The method of claim 1 wherein adding a predetermined constant to the pixel difference values includes adding a constant determined as a function of noise in the fingerprint image.

6. The method of claim 1 wherein adding a predetermined constant includes adding a constant characteristic of an expected average illumination of the image.

7. The method of claim 1 wherein adding a predetermined constant includes adding a constant representative of a pixel value halfway between the maximum pixel value and the minimum pixel value.