Attack test method, device and equipment for biological feature recognition

Abstract

The embodiment of the invention discloses an attack test method, device and equipment for biometric feature recognition, and the method comprises the steps: obtaining a biometric feature assembly needed for the biometric feature recognition attack of biometric recognition equipment, and the biometric feature assembly comprising a feature attachment part attached with a biometric feature, and a conductive part; controlling a mechanical arm to place the biological characteristic assembly in an identification area of the biological identification equipment, controlling the mechanical arm to perform pressing operation on the biological characteristic assembly so as to trigger the biological characteristic assembly to input biological characteristics in the characteristic attachment part into the biological recognition equipment through the conductive part to perform an attack test of biological characteristic identification; and obtaining an attack test result corresponding to the biometric feature component, and determining a test result of performing biometric feature recognition attack on the biometric recognition equipment based on the attack test result corresponding to the biometric feature component.

Description

technical field [0001] This specification relates to the field of computer technology, and in particular to an attack testing method, device and equipment for biometric identification. Background technique [0002] Biometric recognition (such as fingerprint recognition, palmprint recognition, etc.) is often implemented using corresponding biometric recognition algorithms. There are many machine learning algorithms applied to biometric recognition algorithms, and the results of machine learning algorithms are highly probabilistic. [0003] In the process of testing the biometric capabilities of biometric devices, a manual test can usually be used, that is, a number of different test volunteers are called, and then the above-mentioned biometric tests are performed separately through the above-mentioned test volunteers, until all The test volunteers all completed the test to obtain the test results. However, since there are many machine learning algorithms applied in the biomet...

AI Technical Summary

Problems solved by technology

[0003] In the process of testing the biometric capabilities of biometric devices, a manual test can usually be used, that is, a number of different test volunteers are called, and then the above-mentioned biometric tests are performed separately through the above-mentioned test volunteers, until all The test volunteers all completed the test and obtained the test results. However, since there are many machine learning algorithms applied in the biometric identification algorithm, and the machine learning algorithm is characterized by the strong probability of the results, the biometric identification of biometric equipment In the process of ability testing, a lot of attempts are required, which consumes more human resources, and the test cycle will be very long, and manual testing will also cause large errors in each manual operation, and it is difficult to reproduce. Therefore, there is a need to provide a better biometric testing scheme, so that biometrics can be tested more efficiently and accurately

Images