Touchless fingerprint matching systems and methods
In order to authenticate a user of an electronic device, an image of the user's fingerprint is captured. Before feature information is extracted, the fingerprint image is enhanced via localized normalization thereby increasing contrast within the fingerprint image. Thereafter, feature information, such as key point data, is extracted from the image and compared to a predefined template to determine whether the feature information matches the template. If so, the user is authenticated. By enhancing the quality of the fingerprint image through localized normalization and other techniques, the reliability of the matching operation is significantly enhanced. In addition, using key point comparisons for assessing similarity between the feature information and the template helps to address inconsistencies relating to finger rotation, scale, and translation during capture.
1. A mobile communication device for performing touchless fingerprint authentication, comprising:
memory for storing a plurality of templates indicative of images of a fingerprint at different depths of field, each of the templates defining feature descriptors for a plurality of key points within one of the images of the fingerprint at a respective depth of field;
a camera; and
at least one processor programmed to control the camera to capture an image of a fingerprint, the at least one processor further programmed to extract key point data from the captured image by identifying a plurality of key points in the captured image and characterizing each of the plurality of key points in the captured image with a respective feature descriptor that includes a plurality of intensity values for adjacent pixels around the key point being characterized, the at least one processor further programmed to perform comparisons between the feature descriptors characterizing the plurality of key points in the captured image and feature descriptors for each of the templates and to authenticate a user for accessing a resource of the mobile communication device based on the comparisons.
2. The mobile communication device of claim 1 , wherein the at least one processor is further programmed to perform localized normalization on the captured image before extracting the key point data from the captured image.
3. The mobile communication device of claim 1 , wherein the at least one processor is configured to determine which of the feature descriptors characterizing the plurality of key points in the captured image match at least one of the feature descriptors for one of the templates.
4. The mobile communication device of claim 3 , wherein the at least one processor is configured to determine a correlation score based on a number of the feature descriptors characterizing the plurality of key points in the captured image that match at least one of the feature descriptors for the one template, and wherein the at least one processor is configured to compare the correlation score to a threshold.
5. A method for use with a mobile communication device to perform touchless fingerprint authentication, comprising:
storing, in memory, a plurality of templates indicative of images of a fingerprint at different depths of field, each of the templates defining feature descriptors for a plurality of key points within one of the images of the fingerprint at a respective depth of field;
capturing an image of a fingerprint with a camera;
identifying with at least one processor a plurality of key points in the captured image;
characterizing with the at least one processor each of the plurality of key points in the captured image with a feature descriptor that includes a plurality of intensity values for adjacent pixels around the key point being characterized;
comparing with the at least one processor the feature descriptors characterizing the plurality of key points in the captured image to feature descriptors for each of the templates; and
authenticating a user of the mobile communication device based on the comparing.
6. The method of claim 5 , further comprising performing localized normalization on the captured image prior to the defining.
7. The method of claim 5 , further comprising:
analyzing a plurality of frames of a video feed from the camera with the at least one processor;
for each of the frames, determining a value indicative of a focus of an image defined by the respective frame; and
selecting one of the plurality of frames from the video feed for comparison to the templates based on the value indicative of the focus for the selected frame, wherein the selected frame defines the captured image.
8. The method of claim 7 , further comprising:
displaying the video feed with an output interface of the mobile communication device; and
displaying within the video feed a graphical alignment element; and
aligning a user's finger with the camera based on the graphical alignment element.
9. The method of claim 8 , further comprising performing localized normalization on the captured image, the performing localized normalization comprising:
determining a first normalization parameter for a first window of the captured image based on intensity values within the first window;
determining a second normalization parameter for a second window of the captured image based on intensity values within the second window;
adjusting the intensity values within the first window based on the first normalization parameter; and
adjusting the intensity values within the second window based on the second normalization parameter.
10. The mobile communication device of claim 1 , wherein the at least one processor is configured to analyze a plurality of frames of a video feed from the camera and, for each of the plurality of frames, determine a value indicative of a focus of an image defined by the respective frame, wherein the at least one processor is configured to select one of the frames from the video feed for comparison to the templates based on the value indicative of the focus for the selected frame, and wherein the selected frame defines the captured image.
11. The mobile communication device of claim 10 , further comprising an output interface, wherein the at least one processor is configured to display the video feed via the output interface and to display within the video feed a graphical alignment element for aligning a user's finger with the camera.
12. The mobile communication device of claim 11 , wherein the at least one processor is configured to perform localized normalization on the captured image before extracting the key point data by: (1) segmenting the captured image into a plurality of windows, including at least a first window and a second window, (2) determining a first normalization parameter based on intensity values within the first window, (3) determining a second normalization parameter based on intensity values within the second window, (4) adjusting the intensity values within the first window based on the first normalization parameter, and (5) adjusting the intensity values within the second window based on the second normalization parameter.