Systems and methods for biometric image alignment

• US 9,785,819 B1
• Filed: 06/30/2016
• Issued: 10/10/2017
• Est. Priority Date: 06/30/2016
• Status: Active Grant

• Current Assignee: Synaptics Incorporated
• Original Assignee: Synaptics Incorporated
• Inventors: Oreifej, Omar
• Primary Examiner(s): Seth, Manav
• Assistant Examiner(s): Perlman, David
• Application Number:
  US15/199,819
• Time in Patent Office: 467 Days
• Field of Search: None
• CPC Class Codes: G06F 18/22 Matching criteria, e.g. proximity measures
• CPC Class Codes: G06T 7/37 using transform domain methods
• CPC Class Codes: G06V 40/1376 Matching features related to ridge properties or fingerprint texture

Abstract

Systems and methods for aligning images are disclosed. A method includes receiving an input image of a biometric object; identifying a plurality of sets of candidate transformations, wherein each set of candidate transformations included in the plurality of sets of candidate transformations aligns the input image to a different enrollment image included in a plurality of enrollment images; grouping the plurality of sets of candidate transformations into a combined set of candidate transformations; selecting a subset of candidate transformations from the combined set of candidate transformations; and, identifying a first transformation based on the selected subset of candidate transformations, wherein the first transformation aligns the input image to a first enrollment image included in the plurality of enrollment images.

Claims

1. A method for biometric image alignment, comprising:

receiving an input image of a biometric object;

identifying a plurality of sets of candidate transformations, wherein each set of candidate transformations included in the plurality of sets of candidate transformations aligns the input image to a different enrollment image included in a plurality of enrollment images;

grouping the plurality of sets of candidate transformations into a combined set of candidate transformations;

selecting a subset of candidate transformations from the combined set of candidate transformations; and

identifying a first transformation based on the selected subset of candidate transformations, wherein the first transformation aligns the input image to a first enrollment image included in the plurality of enrollment images.

2. The method of claim 1, further comprising:

sorting the combined set of candidate transformations,

wherein selecting the subset of candidate transformations comprises selecting a top number of candidate transformations from the sorted combined set of candidate transformations.

3. The method of claim 2, further comprising:

for each given candidate transformation included in the plurality of sets of candidate transformations, computing a difference metric between the input image and a given enrollment image included in the plurality of enrollment images that corresponds to the given candidate transformation,

wherein sorting the combined set of candidate transformations is based on the difference metric for each given candidate transformation.

4. The method of claim 1, wherein identifying the plurality of sets of candidate transformations comprises:

for a given enrollment image included in the plurality of enrollment images, computing a distance metric between the input image and the given enrollment image at a plurality of possible transformations that align the input image with the given enrollment image; and

identifying a smaller set of candidate transformations out of the plurality of possible transformations, wherein the smaller set of candidate transformations minimize the difference metric between the input image and the given enrollment image.

5. The method of claim 1,

wherein, before the grouping, the plurality of sets of candidate transformations are identified using a first stage of alignment;

wherein, after the grouping, the first transformation is identified using a second stage of alignment;

wherein the first stage of alignment is coarser than the second stage of alignment.

6. The method of claim 1, further comprising:

generating a first representation of the input image; and

generating a second representation of the input image,

wherein, before the grouping, the plurality of sets of candidate transformations are identified based on a difference metric calculated using the first representation of the input image,

wherein, after the grouping, the first transformation is identified based on a difference metric calculated using the second representation of the input image.

7. The method of claim 6,

wherein the first representation of the input image has a lower resolution than the second representation of the input image.

8. The method of claim 6,

wherein the first representation of the input image is an orientation map of the input image;

and

wherein the second representation of the input image is a skeletonized representation of the input image.

9. The method of claim 6,

wherein the first representation of the input image is a representation of a set of points of interest in the input image.

10. The method of claim 1, further comprising:

applying the first transformation to align the input image with of the first enrollment image; and

computing a matching score between the input image and the first enrollment image based on a difference metric between the input image and the first enrollment image with the first transformation applied.

11. The method of claim 10, further comprising:

determining whether the computed matching score is above or below an authentication threshold; and

either allowing access or denying access to a computing device based on whether the computed matching score is above or below the authentication threshold.

12. The method of claim 1, further comprising:

capturing the input image of the biometric object with a biometric sensor.

13. The method of claim 1, wherein each enrollment image in the plurality of enrollment images corresponds to a single enrolled fingerprint.

14. The method of claim 1, wherein each enrollment image in the plurality of enrollment images corresponds to a different enrolled fingerprint.

15. A device, comprising:

a biometric sensor configured to capture an input image of a biometric object; and

a processing system configured to;

receive the input image of the biometric object;

identify a plurality of sets of candidate transformations, wherein each set of candidate transformations included in the plurality of sets of candidate transformations aligns the input image to a different enrollment image included in a plurality of enrollment images;

group the plurality of sets of candidate transformations into a combined set of candidate transformations;

select a subset of candidate transformations from the combined set of candidate transformations; and

identify a first transformation based on the selected subset of candidate transformations, wherein the first transformation aligns the input image to a first enrollment image included in the plurality of enrollment images.

16. The device of claim 15, wherein the processing system is further configured to:

sort the combined set of candidate transformations,

wherein selecting the subset of candidate transformations comprises selecting a top number of candidate transformations from the sorted combined set of candidate transformations.

17. The device of claim 16, wherein the processing system is further configured to:

for each given candidate transformation included in the plurality of sets of candidate transformations, compute a difference metric between the input image and a given enrollment image included in the plurality of enrollment images that corresponds to the given candidate transformation,

wherein sorting the combined set of candidate transformations is based on the difference metric for each given candidate transformation.

18. The device of claim 15, wherein identifying the plurality of sets of candidate transformations comprises:

for a given enrollment image included in the plurality of enrollment images, computing a difference metric between the input image and the given enrollment image at a plurality of possible transformations that align the input image with the given enrollment image; and

identifying a smaller set of candidate transformations out of the plurality of possible transformations, wherein the smaller set of candidate transformations minimize the difference metric between the input image and the given enrollment image.

19. The device of claim 15, wherein the processing system is further configured to:

generate a first representation of the input image; and

generate a second representation of the input image,

wherein, before the grouping, the plurality of sets of candidate transformations are identified based on a difference metric calculated using the first representation of the input image,

wherein, after the grouping, the first transformation is identified based on a difference metric calculated using the second representation of the input image.

20. The device of claim 19,

wherein the first representation of the input image is an orientation map of the input image;

and

wherein the second representation of the input image is a skeletonized representation of the input image.

21. The device of claim 19,

wherein the first representation of the input image is a representation of a set of points of interest in the input image.

22. A non-transitory computer-readable storage medium storing instructions that, when executed by a processor, cause a computing system to perform biometric image alignment, by performing the steps of:

receiving an input image of a biometric object;

identifying a plurality of sets of candidate transformations, wherein each set of candidate transformations included in the plurality of sets of candidate transformations aligns the input image to a different enrollment image included in a plurality of enrollment images;

grouping the plurality of sets of candidate transformations into a combined set of candidate transformations;

selecting a subset of candidate transformations from the combined set of candidate transformations; and

identifying a first transformation based on the selected subset of candidate transformations, wherein the first transformation aligns the input image to a first enrollment image included in the plurality of enrollment images.

23. A device, comprising:

a biometric sensor configured to capture an input image of a biometric object; and

a processor configured to;

receive the input image of the biometric object;

identify a plurality of sets of candidate transformations, wherein each set of candidate transformations included in the plurality of sets of candidate transformations aligns the input image to a different enrollment image included in a plurality of enrollment images;

group the plurality of sets of candidate transformations into a combined set of candidate transformations;

select a subset of candidate transformations from the combined set of candidate transformations; and

identify a first transformation based on the selected subset of candidate transformations, wherein the first transformation aligns the input image to a first enrollment image included in the plurality of enrollment images.

24. The device of claim 23, wherein the processor is further configured to:

sort the combined set of candidate transformations,

wherein selecting the subset of candidate transformations comprises selecting a top number of candidate transformations from the sorted combined set of candidate transformations.

25. The device of claim 24, wherein the processor is further configured to:

for each given candidate transformation included in the plurality of sets of candidate transformations, compute a difference metric between the input image and a given enrollment image included in the plurality of enrollment images that corresponds to the given candidate transformation,

wherein sorting the combined set of candidate transformations is based on the difference metric for each given candidate transformation.

26. The device of claim 23, wherein identifying the plurality of sets of candidate transformations comprises:

for a given enrollment image included in the plurality of enrollment images, computing a difference metric between the input image and the given enrollment image at a plurality of possible transformations that align the input image with the given enrollment image; and

identifying a smaller set of candidate transformations out of the plurality of possible transformations, wherein the smaller set of candidate transformations minimize the difference metric between the input image and the given enrollment image.

27. The device of claim 23, wherein the processor is further configured to:

generate a first representation of the input image; and

generate a second representation of the input image,

wherein, before the grouping, the plurality of sets of candidate transformations are identified based on a difference metric calculated using the first representation of the input image,

wherein, after the grouping, the first transformation is identified based on a difference metric calculated using the second representation of the input image.

28. The device of claim 27,

wherein the first representation of the input image is an orientation map of the input image;

and

wherein the second representation of the input image is a skeletonized representation of the input image.

29. The device of claim 27,

wherein the first representation of the input image is a representation of a set of points of interest in the input image.