WO2019109322A1

WO2019109322A1 - Biometric feature detection method and identification system

Info

Publication number: WO2019109322A1
Application number: PCT/CN2017/115142
Authority: WO
Inventors: 杨孟达; 江志伟
Original assignee: 深圳市汇顶科技股份有限公司
Priority date: 2017-12-08
Filing date: 2017-12-08
Publication date: 2019-06-13
Also published as: CN108140120A

Abstract

Provided is a biometric feature detection method, comprising: according to a rapid detection, determining whether a user biometric feature satisfies a comparison condition of the rapid detection (204); if the biometric feature satisfies the comparison condition of the rapid detection, then determining that said biometric feature passes detection (206); if the biometric feature does not satisfy the comparison condition of the rapid detection, then, according to a complete detection, determining whether the biometric feature satisfies the comparison condition of the complete detection (208). The method has the advantages of rapid and correct detection of biometric features as well as a reduction of the false rejection rate.

Description

Biometric detection method and identification system

Technical field

The present invention relates to a biometric detection method and an identification system, and more particularly to a biometric detection method and an identification system for reducing false rejection rate (FRR).

Background technique

With the development and advancement of technology, mobile electronic devices such as mobile phones, digital cameras, tablet computers, and notebook computers have become indispensable tools in people's lives. In order to effectively avoid theft of internal data of the personal mobile electronic device, various protection methods are also generated, such as password authentication, and user biometric detection methods including voiceprint recognition, fingerprint recognition, facial recognition, and the like. However, when the biometric detection method currently used collects biometric images of a user, if a biometric feature such as a fingerprint or a face is collected for a long exposure time, a clear image can be obtained for subsequent biometric comparison, but The required time is longer, which in turn affects the recognition efficiency. Conversely, if the biometric features such as fingerprints or faces are collected with a short exposure time, the obtained image resolution is lower, which may lead to an increase in the rejection rate.

Therefore, how to provide a detection method to improve the identification efficiency and reduce the misjudgment caused by the unclear image has become one of the goals of the industry.

Summary of the invention

Therefore, the main purpose of the present application is to provide a biometric detection method and an identification system, which can quickly and correctly detect user biometrics to reduce the rejection rate.

In order to solve the above technical problem, the embodiment of the present application provides a biometric detection method, including: determining, according to a fast detection, whether a biometric feature of a user satisfies the comparison condition of the fast detection; when the biometric feature When the comparison condition of the rapid detection is satisfied, determining that the biometric feature passes the detection; and when the biometric feature does not meet the comparison condition of the rapid detection, determining whether the biometric feature conforms to the Alignment conditions for complete detection.

For example, the fast detection has a first detection time length, and the complete detection has a second detection time length, wherein the first detection time length is less than the second detection time length.

For example, the fast detection has the same alignment conditions as the complete detection.

For example, the fast detection has a first detection resolution, the complete detection has a second detection resolution, and the first detection resolution is smaller than the second detection resolution.

The biometric detection method provided by the present application has the advantages of quickly and correctly detecting biometrics and reducing the biometric rejection rate.

DRAWINGS

1 is a schematic diagram of an identification system according to an embodiment of the present application;

FIG. 2 is a schematic diagram of a biometric detection process according to an embodiment of the present application.

Detailed ways

In order to make the objects, technical solutions, and advantages of the present application more comprehensible, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are merely illustrative of the application and are not intended to be limiting.

Please refer to FIG. 1. FIG. 1 is a schematic diagram of an identification system 10 according to an embodiment of the present application. The identification system 10, for example, a fingerprint identification system or a facial recognition system, can be used to perform feature comparison according to a user's biometrics to confirm whether the compared biometrics belong to the authenticated user, thereby completing the identification process. The identification system 10 includes a processor 102 and a storage device 104. The storage device 104 can store program codes to instruct the processor 102 to perform related operations. In order to achieve a fast and correct feature comparison, the identification system 10 is used to perform the identification process. The embodiment of the present application provides a biometric detection method to complete the identification process.

Please refer to FIG. 2. FIG. 2 is a schematic diagram of a biometric detection process 20 according to an embodiment of the present application. The detection process 20 can be used to identify the system 10 for biometric alignment. The detection process 20 can be compiled into program code and stored in the storage device 104 to instruct the processor 102 to perform corresponding operations. In the following embodiments, fingerprint images are taken as an example. In other embodiments, the biometrics of the present application The detection method can also be used for detecting and recognizing user biometrics such as faces and pupils. The detection process 20 includes the following steps:

Step 202: Start.

Step 204: Determine whether the fingerprint image meets a comparison condition according to a fast detection.

Step 206: When the fingerprint image meets a comparison condition, it is determined that the fingerprint image passes the detection.

Step 208: When the fingerprint image does not meet the comparison condition, determine whether the fingerprint image meets the comparison condition according to a complete detection.

Step 210: Determine that the fingerprint image does not pass the detection.

Step 212: End.

According to the detection process 20, in an embodiment, the identification system 10 is configured to detect a fingerprint image. In step 204, according to the fast detection, it is determined whether the fingerprint image meets the comparison condition, and then, when the fingerprint meets the comparison condition, the step is performed. In 206, it is determined that the fingerprint image passes the detection. For example, the comparison condition may be that the fingerprint image is compared with a pre-stored fingerprint template to determine whether the fingerprint image has a fingerprint. The feature points included in the template, which meet certain conditions, or other methods and conditions for comparing images, are not limited thereto. Further, the detection process 20 can perform detection based on one or more alignment conditions. Then, as described in step 208, when the fingerprint image of the user does not meet the comparison condition of the fast detection (for example, the feature points included in the fingerprint image do not satisfy the preset condition), whether the fingerprint image meets the comparison condition according to the complete detection is determined. . In this way, when the fingerprint image of the user meets the comparison condition of the fast detection, the fingerprint can be quickly confirmed to pass the detection, and when the fingerprint does not meet the comparison condition of the rapid detection, the judgment can still be performed according to the complete detection to ensure The correctness of the comparison, in turn, reduces the False Rejection Rate (FRR).

In detail, the difference between fast detection and complete detection is that the length of time for rapid detection of image acquisition, image processing and comparison process is shorter than that of complete detection of image acquisition, image processing and image comparison process. . In an embodiment, it is assumed that the fast detection and the complete detection are the same in the image processing and the comparison, and the fast detection has a first detection time length T1, and the complete detection has a second detection time length T2. The fast detection can perform image acquisition with a shorter exposure time, while the complete detection is performed with a longer exposure time. Therefore, the first detection time length T1 is smaller than the second detection time length T2. In an embodiment, the length of the exposure time for setting the complete detection is more than twice the length of the rapid detection exposure time to increase the difference between the first detection time length T1 and the second detection time length T2. For example, in the detection process 20, the fast detection acquires the fingerprint image for detection by the exposure time of 20 milliseconds, and when the fingerprint image meets the comparison condition, it determines that the fingerprint image passes the detection; otherwise, when the fingerprint image does not meet the comparison In the case of the condition, the fingerprint image is re-acquired by the complete detection with an exposure time of 45 milliseconds for detection, and it is judged whether the comparison condition is met. The comparison conditions corresponding to the fast detection and the complete detection may be the same, and will not change due to the length of the detection time.

In another embodiment, the fast detection and the full detection may have different detection resolutions to capture the fingerprint image. In detail, the fast detection has a first detection resolution R1, and the complete detection tool There is a second detection resolution R2, and the first detection resolution R1 is smaller than the second detection resolution R2. Therefore, the fast detection can take a lower resolution fingerprint image in a shorter time for feature comparison. That is to say, in the detection process 20, the fast detection detects the fingerprint image with the first detection resolution R1 to perform the feature comparison of the fingerprint image. When the fingerprint image meets the comparison condition, the fingerprint image is determined to pass the detection; When the fingerprint image does not meet the comparison condition, the complete detection is started to acquire the fingerprint image with a higher second detection resolution R2, and the feature comparison is performed. The comparison conditions corresponding to the fast detection and the complete detection may be the same, that is, the fingerprint image is detected under the same comparison condition regardless of the resolution of the captured fingerprint image. When the fingerprint image passes the fast detection, the time for detecting the fingerprint image can be greatly reduced compared to the complete detection. In this way, the detection method of the present application can reduce the time taken to detect biometrics without changing the alignment conditions.

It is to be noted that the foregoing embodiments are intended to illustrate the spirit of the present application, and those of ordinary skill in the art can be appropriately modified. For example, different detection time lengths or resolutions corresponding to fast detection and complete detection may be adjusted according to the needs of the computer system or the user. Alternatively, the comparison conditions can be adjusted according to other detection methods as a basis for the detection process. Alternatively, the results of the fast detection and the complete detection may be referred to at the same time, and then it is determined whether the feature meets the comparison condition, and is not limited thereto, and is applicable to the present application.

In summary, the present application performs biometric comparison of users according to the rapid detection and complete detection methods, improves the efficiency at the time of comparison, and reduces the rejection rate.

The above is only a part of the embodiments of the present application, and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application are included in the scope of protection of the present application. within.

Claims

A biometric detection method, comprising:

Determining whether a biometric of a user satisfies the comparison condition of the fast detection according to a fast detection;

Determining that the biometric passes the detection when the biometric satisfies the comparison condition of the rapid detection;

When the biometric feature does not meet the comparison condition of the rapid detection, it is determined whether the biometric feature meets the comparison condition of the complete detection according to a complete detection.
The biometric detecting method according to claim 1, wherein the fast detecting time first detecting time, the complete detecting time second detecting time, wherein the first detecting time length is smaller than the second detecting time length of time.
The biometrics detecting method according to claim 2, wherein the first detecting time includes a first exposure time, and the second detecting time includes a second exposure time, the length of the second exposure time It is more than twice the length of the first exposure time.
The biometric detection method according to claim 1, wherein said fast detection has a first detection resolution, said complete detection has a second detection resolution, wherein said first detection resolution is smaller than The second detection resolution is described.
The biometric detecting method according to any one of claims 1 to 4, wherein the comparison condition of the rapid detection is the same as the comparison condition of the complete detection.
An identification system characterized by detecting a biometric using the detection method of any of claims 1-5.