Multi-stage liveness determination
A method of determining whether a biometric object is part of a live individual is described. In one such method, image information is acquired from the biometric object by using a sensor, such as an ultrasonic sensor. The image information may be analyzed in at least two analysis stages. One of the analysis stages may be a temporal analysis stage that analyzes changes in the image information obtained during a time period throughout which the biometric object was continuously available to the sensor. For example, a dead/alive stage may analyze differences between image information taken at two different times in order to identify changes from one time to the next. Other stages may focus on aspects of a particular image information set, rather than seeking to assess changes over time. These other stages seek to determine whether an image information set exhibits characteristics similar to those of a live biometric object.
1. A method of determining whether a biometric object is part of a live individual, comprising:
acquiring image information from the biometric object using an ultrasonic sensor, wherein acquiring the image information involves a plurality of range gate delays and a plurality of ultrasonic frequencies; and
analyzing the image information in at least two analysis stages, wherein one of the analysis stages is a temporal analysis stage that analyzes changes in the image information obtained during a time period throughout which the biometric object was continuously available to the sensor and the other of the analysis stages involves analyzing the image information to determine whether the biometric object includes at least one manufactured layer.
2. The method of claim 1 , wherein the temporal analysis stage determines whether changes in the image information are similar to changes consistent with a live object.
3. The method of claim 1 , wherein the temporal analysis stage involves determining whether a feature of the biometric object changed during the time period.
4. The method of claim 3 , wherein the feature is selected from the group consisting of a pore opening size, a pore depth, a ridge pattern, and a texture.
5. The method of claim 1 , wherein
the other of the analysis stages analyzes the image information to determine whether the biometric object is three-dimensional.
6. The method of claim 5 , wherein determining whether the biometric object is three-dimensional includes determining whether patterns of the biometric object are uniform.
7. The method of claim 6 , wherein a Sobel filter is used to determine whether patterns of the biometric object are uniform.
8. The method of claim 5 , further comprising concluding that the biometric object is part of a live individual if:
(i) the temporal analysis stage indicates changes in the image information are similar to changes consistent with a live object; and
(ii) the other of the analysis stages indicates that the object is three-dimensional.
9. The method of claim 1 , wherein
the other of the analysis stages analyzes the image information to determine whether pattern types in the image information are among predetermined pattern types.
10. The method of claim 9 , wherein determining whether pattern types in the image information are among predetermined pattern types includes identifying pattern classifications presented in the image information, and then determining whether the pattern classifications are among predetermined pattern classifications.
11. The method of claim 10 , wherein identifying pattern classifications includes identifying whether patterns of the biometric object exhibit ridge flow.
12. The method of claim 9 , further comprising concluding that the biometric object is part of a live individual if:
(i) the temporal analysis stage indicates that a feature of the biometric object changed during the time period; and
(ii) the other of the analysis stages indicates that the pattern types in the image information are among predetermined pattern types.
13. The method of claim 1 , wherein determining whether the object includes at least one manufactured layer includes assessing image quality of the image information.
14. The method of claim 1 , further comprising concluding that the biometric object is part of a live individual if:
(i) the temporal analysis stage indicates that a feature of the biometric object changed during the time period; and
(ii) the other of the analysis stages indicates that the biometric object does not include at least one manufactured layer.
15. The method of claim 1 , wherein the ultrasonic sensor includes a platen on which the biometric object is placed.
16. The method of claim 1 , further comprising providing a liveness output signal based on the analyzed image information.
17. A non-transitory computer readable medium storing computer executable code, the executable code comprising instructions to:
acquire image information from the biometric object using an ultrasonic sensor, wherein acquiring the image information involves a plurality of range gate delays and a plurality of ultrasonic frequencies; and
analyze the image information in at least two analysis stages, wherein one of the analysis stages is a temporal analysis stage that analyzes changes in the image information obtained during a time period throughout which the biometric object was continuously available to the sensor and the other of the analysis stages involves analyzing the image information to determine whether the biometric object includes at least one manufactured layer.
18. The medium of claim 17 , wherein determining whether the object includes at least one manufactured layer includes assessing image quality of the image information.
19. The medium of claim 17 , wherein the executable code comprises instructions to determine whether
pattern types in the image information are among predetermined pattern types.
20. A system for determining whether a biometric object is part of a live individual, comprising:
an ultrasonic sensor configured to acquire image information from the biometric object using a plurality of range gate delays and a plurality of ultrasonic frequencies;
a processor configured to analyze the image information in at least two analysis stages, wherein one of the analysis stages is a temporal analysis stage that analyzes changes in the image information acquired during a time period throughout which the biometric object was continuously available to the sensor and the other of the analysis stages involves analyzing the image information to determine whether the biometric object includes at least one manufactured layer.
21. The system of claim 20 , wherein the processor is further configured to determine whether pattern types in the image information are among predetermined pattern types.
22. The system of claim 20 , wherein the processor is further configured to determine image quality of the image information.