TWI795724B - An identity verification method and an identity verification system - Google Patents

Abstract

An identity verification method is implemented by a processing unit and includes the step of: receiving register data including a register license plate number and a raw facial image; bluring a least a part of the raw facial image to generate a blured facial image; deleting the raw facial image; adding the blured facial image into the register data; receiving a license plate image to be verified and a facial image to be verified; recognizing the license plate image to be verified to generate a license plate number to be verified; and, upon determining that license plate number to be verified matches the register license plate number of the register data and the similarity of the facial image to be verified and the blured facial image of the register data reaches a predetermined similarity threshold, generating a successful verification result.

Description

Identity verification method and identity verification system

The present invention relates to an identity verification method, in particular to an identity verification method involving image recognition technology. The invention also relates to an identity verification system.

Image recognition is a commonly used technical means in current identity verification mechanisms. The identity verification system determines whether to pass the verification by comparing and analyzing the similarity between the face image to be verified and the pre-stored face image. However, storing the pre-stored facial images in the cloud poses a security risk of being stolen by hackers. How to improve the above-mentioned shortcoming of the prior art is a subject to be further discussed in the present invention.

Therefore, the object of the present invention is to provide an identity verification method.

Another object of the present invention is to provide an identity verification system.

Therefore, the identity verification method of the present invention is executed by a processing unit, the The method includes: receiving registration data, the registration data including a registered license plate number and an original facial image; at least partially blurring the original facial image to generate a blurred facial image, and deleting the original facial image, and add the blurred facial image to the registration information; receive a license plate image to be verified and a face image to be verified; identify the license plate image to be verified to generate a license plate number to be verified, and determine the license plate number to be verified and the Whether the registered license plate number in the registration data matches; whether the similarity between the facial image to be verified and the blurred facial image in the registered data reaches a predetermined similarity threshold; and when it is determined that the license plate number to be verified and The registered license plate number of the registration data matches, and the similarity between the facial image to be verified and the blurred facial image of the registration data reaches the predetermined similarity threshold, and a successful verification result is generated.

In some implementations, the received registration information also includes a payment information. In addition to receiving the license plate image to be verified and the face image to be verified, a consumption amount is also received. The method also includes: after generating the successful verification result , executing a payment procedure according to the payment information of the registration information and the consumption amount.

In some implementation aspects, the identity verification method further includes: after judging whether the similarity between the facial image to be verified and the blurred facial image of the registration data reaches the predetermined similarity threshold, The image of the face to be verified is deleted.

In some implementations, the received registration data also includes a security level setting data, which is used to determine whether the similarity between the facial image to be verified and the blurred facial image in the registration data reaches the predetermined similarity threshold Previously from multiple pairs From the similarity thresholds of multiple security level data, select the corresponding security level data that is consistent with the security level setting data as the predetermined similarity threshold.

In some implementation aspects, after receiving the facial image to be verified and before judging whether the similarity between the facial image to be verified and the blurred facial image of the registration data reaches the predetermined similarity threshold, The facial image to be verified is at least partially blurred in the same manner as the original facial image is blurred. In the step of judging whether the similarity between the facial image to be verified and the blurred facial image of the registration data reaches the predetermined similarity threshold, it is judged that the facial image to be verified after blurring and the registration data Whether the similarity of the blurred facial image reaches the predetermined similarity threshold.

The identity verification system of the present invention includes an image capturing device, a terminal computer device, and a verification server. The verification server stores a registration data, and the registration data includes a registered license plate number and an original face image.

The verification server at least partially blurs the original facial image to generate a blurred facial image, deletes the original facial image, and adds the blurred facial image to the registration data.

The image capturing device photographs a license plate of a vehicle to generate a license plate image to be verified, and photographs a driver of the vehicle to generate a facial image to be verified, and transmits the license plate image to be verified and the facial image to be verified Give the terminal computer device.

The terminal computer device uses the image of the license plate to be verified and the image of the face to be verified image to the authentication server.

The verification server identifies the license plate image to be verified to generate a license plate number to be verified, and judges whether the license plate number to be verified matches the registered license plate number of the registration data.

The verification server determines whether the similarity between the facial image to be verified and the blurred facial image in the registration data reaches a predetermined similarity threshold.

When the verification server determines that the license plate number to be verified matches the registered license plate number in the registration data, and the similarity between the facial image to be verified and the blurred facial image in the registration data reaches the predetermined similarity threshold value, the authentication server generates a successful authentication result.

The identity verification system of the present invention includes an image capture device, a terminal computer device, an in-vehicle image capture device and a verification server.

The in-vehicle image capture device is suitable for being arranged in a vehicle and stores a device identification data.

The verification server stores a registration data, which includes a registration license plate number, an original face image and the device identification data.

The verification server at least partially blurs the original facial image to generate a blurred facial image, deletes the original facial image, and adds the blurred facial image to the registration data.

The image capturing device captures a license plate of the vehicle to generate a vehicle to be verified license plate image, and the image of the license plate to be verified is transmitted to the terminal computer device.

The terminal computer device transmits the license plate image to be verified to the verification server.

The verification server identifies the license plate image to be verified to generate a license plate number to be verified, and judges whether the license plate number to be verified matches the registered license plate number of the registration data.

When the verification server judges that the license plate number to be verified matches the registered license plate number in the registration data, the verification server sends a driving image request to the in-vehicle image capture device according to the device identification data in the registration data.

When the in-vehicle image capture device receives the driving image request, the in-vehicle image capture device captures a driver of the vehicle to generate a facial image to be verified, and transmits the facial image to be verified and the device identification data to the authentication server.

The verification server determines whether the similarity between the facial image to be verified and the blurred facial image in the registration data reaches a predetermined similarity threshold.

When the verification server determines that the license plate number to be verified matches the registered license plate number in the registration data, and the similarity between the facial image to be verified and the blurred facial image in the registration data reaches the predetermined similarity threshold value, the authentication server generates a successful authentication result.

The effect of the present invention lies in: using the verification server to at least partially blur the original facial image to generate the blurred facial image, and delete the original facial image It can prevent hackers from obtaining the original facial image when invading the system, thereby improving system security. Furthermore, by deleting the facial image to be verified, it can also prevent hackers from obtaining the facial image to be verified when intruding into the system. In addition, after the verification server generates the successful verification result, the payment procedure is executed according to the payment information of the registration information and the consumption amount, so that the driver does not need to get out of the car, and does not need to take cash, credit card or mobile phone. Payment can be completed, thereby increasing the convenience of payment.

100: Identity Verification System

1: Image capture device

2: Terminal computer device

3: Verify the server

4: In-vehicle image capture device

S01~S03: Steps

S11~S18: steps

S21~S31: Steps

Other features and effects of the present invention will be clearly presented in the implementation manner with reference to the drawings, wherein: Fig. 1 is a schematic diagram of a hardware connection relationship of a first embodiment of the identity verification system of the present invention; Fig. 2 is the schematic diagram of the hardware connection A flow chart of the first embodiment, illustrating an image processing program; FIG. 3 is another flow chart of the first embodiment, illustrating a verification and payment program; FIG. 4 is a second embodiment of the identity verification system of the present invention A schematic diagram of hardware connections; and FIG. 5 (including FIG. 5A and FIG. 5B ) is a flowchart of the second embodiment, illustrating the verification and payment procedure.

Before the present invention is described in detail, it should be noted that in the following description, similar elements are denoted by the same numerals.

Referring to FIG. 1 and FIG. 2 , a first embodiment of the identity verification system 100 of the present invention includes an image capture device 1 , a terminal computer device 2 and a verification server 3 . The terminal computer device 2 is electrically connected with the image capture device 1 to receive the content captured by the image capture device 1 and communicate with the verification server 3 through a communication network. The image capture device 1 and the terminal computer device 2 are, for example, set up in a business (such as a gas station, a fast food restaurant consumption lane or a parking lot).

The verification server 3 stores a registration data, a plurality of security level data and a plurality of similarity thresholds respectively corresponding to the security level data. The registration data includes a registered license plate number, an original face image, a payment data and a security level setting data.

The registration information is, for example, received from an electronic device operated by a user (such as a smart phone, a tablet computer, a laptop computer or a desktop computer). The security level data respectively indicate low security level, medium security level and high security level, and the similarity threshold values respectively indicate 40%, 60% and 80% similarity, but not limited thereto. The security level setting data of the registration data is used for the user to set the security level as one of low security level, medium security level and high security level. The payment information of the registration information is, for example, a credit card information.

Referring to Fig. 1 and Fig. 2, the following describes the implementation of the identity verification system 100 like handler. First, as shown in step S01 , the verification server 3 at least partially blurs the original facial image to generate a blurred facial image. For example, the verification server 3 may mosaic the eye parts in the original facial image, or process the entire original facial image with a low-pass filter. The degree of image blurring mentioned above can be the degree preset by the system, or the degree specified by the user through the registration information.

Next, as shown in step S02, the verification server 3 deletes the original facial image.

Next, as shown in step S03, the verification server 3 adds the blurred facial image into the registration data.

By means of the image processing program, it is possible to prevent hackers from obtaining the original face image when intruding into the system, thereby improving system security.

Referring to FIG. 1 and FIG. 3 , the identity verification system 100 performs a verification and payment procedure as follows. First, as shown in step S11, the image capture device 1 photographs a license plate of a vehicle to generate a license plate image to be verified, and photographs a driver of the vehicle to generate a face image to be verified, and takes the license plate to be verified The image and the facial image to be verified are sent to the terminal computer device 2 .

Next, as shown in step S12, the terminal computer device 2 transmits the license plate image to be verified and the face image to be verified to the verification server 3 via the communication network. In this embodiment, in addition to the image of the license plate to be verified and the image of the license plate to be verified, the terminal computer device 2 In addition to sending the verification facial image to the verification server 3 , a consumption amount (such as fuel for refueling, fast food consumption or parking fee) is also sent to the verification server 3 .

Next, as shown in step S13, the verification server 3, for example, uses optical character recognition technology, to identify the image of the license plate to be verified to generate a license plate number to be verified.

Next, as shown in step S14, the verification server 3 judges whether the license plate number to be verified matches the registered license plate number in the registration data, and the facial image to be verified is consistent with the blurred facial image in the registration data. Whether the similarity reaches a predetermined similarity threshold. In this embodiment, the verification server 3 selects from the similarity thresholds the one whose corresponding security level data matches the security level setting data as the predetermined similarity threshold.

If the determination in step S14 is negative, step S15 is executed, and the verification server 3 generates a verification failure result.

If step S14 is judged to be yes, that is, when the verification server 3 judges that the license plate number to be verified matches the registered license plate number of the registration data, and the face image to be verified matches the blurred face of the registration data. When the image similarity reaches the predetermined similarity threshold, step S16 is executed, and the verification server 3 generates a verification success result.

Next, as shown in step S17, after the verification server 3 generates the successful verification result, it executes a payment procedure according to the payment data and the consumption amount of the registration data. After successfully executing the payment procedure, the verification server 3 may return a payment The terminal computer device 2 is notified of success.

Finally, as shown in step S18, the verification server 3 deletes the facial image to be verified.

The dual verification of the license plate and face in the verification and payment program can provide a good security verification mechanism. Furthermore, by deleting the face image to be verified, it is possible to prevent hackers from obtaining the face image to be verified when they invade the system , so as to improve the security of the system. In addition, by matching the payment steps after successful verification, the driver can complete the payment without getting out of the car, without taking cash, credit card or mobile phone, thereby increasing the convenience of payment.

It is supplemented that, in another embodiment, before the terminal computer device 2 transmits the facial image to be verified to the verification server 3 (before step S12), it also blurs the original facial image In the same manner, at least a part of the face image to be verified is blurred. When the terminal computer device 2 transmits the facial image to be verified to the verification server 3 (in step S12 ), it transmits the fuzzy facial image to be verified to the verification server 3 . When the verification server 3 judges whether the similarity between the facial image to be verified and the blurred facial image of the registration data reaches the predetermined similarity threshold (in step S14), it judges that the blurred facial image Whether the similarity between the facial image to be verified and the blurred facial image of the registration data reaches the predetermined similarity threshold. Thereby, the situation that the unfuzzy facial image to be verified is stolen by hackers in the cloud can be avoided.

Referring to FIG. 4 and FIG. 5 (including FIG. 5A and FIG. 5B ), a second embodiment of the identity verification system 100 of the present invention is similar to the second embodiment, and the similarities will not be repeated. The differences are described as follows.

In this embodiment, the identity verification system 100 also includes an in-vehicle image capture device 4 . The in-vehicle image capture device 4 has a networking function and can communicate with the verification server 3 through the Internet. The in-vehicle image capturing device 4 is suitable for being arranged in the vehicle, and stores a device identification data (such as a device serial number). The registration data stored in the verification server 3 also includes the device identification data.

The image processing procedure of the second embodiment is the same as that of the first embodiment. The differences in the verification and payment procedures are described below.

Step S21 is similar to step S11, except that the image capture device 1 is only used to capture the license plate of the vehicle to generate the image of the license plate to be verified, not to capture the driver of the vehicle, and only transmits the image of the license plate to be verified Give the terminal computer device 2.

Step S22 is similar to step S12, except that the terminal computer device 2 only transmits the image of the license plate to be verified to the verification server 3 .

Step S23 is the same as step S13.

In step S24, the verification server 3 judges whether the license plate number to be verified is consistent with the registered license plate number in the registration data. If not, execute step S25 to generate a verification failure result. If yes, execute step S26.

Step S26 is when the verification server 3 determines that the license plate number to be verified matches the registered license plate number of the registration data, the verification server 3 sends a driving image request to the vehicle interior image according to the device identification data of the registration data Camera 4.

Next, as shown in step S27, when the in-vehicle image capturing device 4 receives the driving image request, the in-vehicle image capturing device 4 captures a driver of the vehicle to generate a face image to be verified, and the image to be verified The verification facial image and the device identification data are sent to the verification server 3 .

Next, as shown in step S28, the verification server 3 judges whether the similarity between the facial image to be verified and the blurred facial image in the registration data reaches a predetermined similarity threshold, if not, proceed to the following steps S25, if yes, then execute steps S29-S31. Steps S29-S31 are the same as steps S16-S18. In short, the main difference between the second embodiment and the first embodiment is that the face image to be verified is provided to the verification server 3 by the in-vehicle image capture device 4 instead of by the image capture device 1 is provided to the authentication server 3 .

It is supplemented that, in another embodiment, before the in-vehicle image capture device 4 transmits the facial image to be verified to the verification server 3, it also blurs the original facial image in the same manner as the original facial image. The image of the face to be verified is at least partially blurred. When the in-vehicle image capture device 4 sends the facial image to be verified to the verification server 3 (in step S27), it is to send the facial image to be verified after the paste to the verification server. Device 3. When the verification server 3 judges whether the similarity between the facial image to be verified and the blurred facial image of the registration data reaches the predetermined similarity threshold (in step S28), it judges that the blurred facial image Whether the similarity between the facial image to be verified and the blurred facial image of the registration data reaches the predetermined similarity threshold. Thereby, the situation that the unfuzzy facial image to be verified is stolen by hackers in the cloud can be avoided.

A third embodiment of the identity verification system 100 of the present invention is similar to the first embodiment and the second embodiment, and the differences are described as follows. In the third embodiment, the identity verification system 100 also includes a user terminal electronic device (not shown in the figure, such as a smart phone, a tablet computer, a laptop computer or desktop computer). The user-end electronic device stores the original facial image, and the user-end electronic device at least partially blurs the original facial image to generate the blurred facial image, and the user-end electronic device includes the blur The registered data including the personalized facial image, the registered license plate number, the payment data and the security level setting data are transmitted to the identity verification system 100 . That is to say, the verification server 3 does not store the original facial image from the beginning to the end, but only stores the blurred facial image generated by the user-end electronic device, thereby further reducing the risk of the original facial image being stolen by hackers .

In summary, the identity verification system 100 of the present invention uses the verification server 3 to at least partially blur the original facial image to generate the blurred facial image, and delete the original facial image, which can prevent hackers from intruding. The original face shadow is obtained when the system image, thereby improving system security, moreover, by deleting the facial image to be verified, it can also prevent hackers from obtaining the facial image to be verified when invading the system, and in addition, the verification is successful through the verification server 3 After the result, the payment program is executed according to the payment information of the registration information and the consumption amount, which can allow the driver to complete the payment without getting out of the car, without taking cash, credit card or mobile phone, thereby increasing the convenience of payment, so it is true Can reach the purpose of the present invention.

But the above-mentioned ones are only embodiments of the present invention, and should not limit the scope of the present invention. All simple equivalent changes and modifications made according to the patent scope of the present invention and the content of the patent specification are still within the scope of the present invention. Within the scope covered by the patent of the present invention.

S11~S18: steps

Claims (15)

An identity verification method, performed by a processing unit, the method includes: receiving a registration data, the registration data includes a registration license plate number and an original facial image; at least partially blurring the original facial image to generate a blurred facial image, deleting the original facial image, and adding the blurred facial image to the registration data; Receive a license plate image to be verified and a face image to be verified; Identify the license plate image to be verified to generate a license plate number to be verified, and determine whether the license plate number to be verified matches the registered license plate number in the registration data; judging whether the similarity between the facial image to be verified and the blurred facial image in the registration data reaches a predetermined similarity threshold; and When it is determined that the license plate number to be verified matches the registered license plate number of the registration data, and the similarity between the facial image to be verified and the blurred facial image of the registration data reaches the predetermined similarity threshold, a Validate successful results. The identity verification method as described in Claim 1, wherein the received registration information also includes a payment information, in addition to receiving the license plate image to be verified and the face image to be verified, a consumption amount is also received, and the method also includes: After the successful verification result is generated, a payment procedure is executed according to the payment information of the registration information and the consumption amount. The identity verification method as described in Claim 1, further comprising: After judging whether the similarity between the facial image to be verified and the blurred facial image of the registration data reaches the predetermined similarity threshold, the facial image to be verified is deleted. The identity verification method as described in Claim 1, wherein the received registration data also includes a security level setting data, and is used to determine whether the similarity between the facial image to be verified and the blurred facial image of the registration data is reached Before the predetermined similarity threshold value, the one whose corresponding security level data is consistent with the security level setting data is selected from a plurality of similarity threshold values respectively corresponding to a plurality of security level data as the predetermined similarity threshold value. The identity verification method as described in Claim 1, after receiving the facial image to be verified and judging whether the similarity between the facial image to be verified and the blurred facial image of the registration data reaches the predetermined similarity threshold Before the value, also include: at least partially blurring the image of the face to be verified in the same manner as blurring the original image of the face; Wherein, in the step of judging whether the similarity between the facial image to be verified and the blurred facial image of the registration data reaches the predetermined similarity threshold, it is judged that the facial image to be verified after blurring and the Whether the similarity of the blurred facial image of the registered data reaches the predetermined similarity threshold. An identity verification system comprising: an image capture device; a terminal computer device; and A verification server, storing a registration data, the registration data includes a registered license plate number and an original facial image; The verification server at least partially blurs the original facial image to generate a blurred facial image, deletes the original facial image, and adds the blurred facial image to the registration data; The image capturing device photographs a license plate of a vehicle to generate a license plate image to be verified, and photographs a driver of the vehicle to generate a facial image to be verified, and transmits the license plate image to be verified and the facial image to be verified to the terminal computer device; The terminal computer device transmits the license plate image to be verified and the face image to be verified to the verification server; The verification server identifies the license plate image to be verified to generate a license plate number to be verified, and determines whether the license plate number to be verified matches the registered license plate number in the registration data; The verification server judges whether the similarity between the facial image to be verified and the blurred facial image in the registration data reaches a predetermined similarity threshold; When the verification server determines that the license plate number to be verified matches the registered license plate number in the registration data, and the similarity between the facial image to be verified and the blurred facial image in the registration data reaches the predetermined similarity threshold value, the authentication server generates a successful authentication result. The identity verification system as described in claim 6, wherein the registration data also includes a payment data, and the terminal computer device not only transmits the license plate image to be verified and the face image to be verified to the verification server, but also Send a consumption amount to the verification server; After the verification server generates the successful verification result, it executes a payment procedure according to the payment information of the registration information and the consumption amount. The identity verification system as described in Claim 6, wherein, after the verification server determines whether the similarity between the facial image to be verified and the blurred facial image in the registration data reaches the predetermined similarity threshold, The image of the face to be verified is deleted. The identity verification system as described in Claim 6, wherein, the verification server stores a plurality of security level data and a plurality of similarity thresholds respectively corresponding to the security level data, and the registration data also includes a security level setting data, the verification server selects from the similarity thresholds the corresponding security level data that matches the security level setting data as the predetermined similarity threshold. The identity verification system as described in Claim 6, wherein, before the terminal computer device transmits the face image to be verified to the verification server, the face image to be verified is also blurred in the same manner as the original face image Part of the image is at least partially blurred; Wherein, when the terminal computer device transmits the facial image to be verified to the verification server, it transmits the fuzzed facial image to be verified to the verification server; Wherein, when the verification server judges whether the similarity between the facial image to be verified and the blurred facial image in the registration data reaches the predetermined similarity threshold, it judges that the facial image to be verified after blurring Whether the similarity with the blurred facial image of the registration data reaches the predetermined similarity threshold. An identity verification system comprising: an image capture device; a terminal computer device; An in-vehicle image capture device, suitable for being installed in a vehicle, and storing a device identification data; and A verification server, storing a registration data, the registration data includes a registration license plate number, an original facial image and the device identification data; The verification server at least partially blurs the original facial image to generate a blurred facial image, deletes the original facial image, and adds the blurred facial image to the registration data; The image capture device captures a license plate of the vehicle to generate a license plate image to be verified, and transmits the license plate image to be verified to the terminal computer device; The terminal computer device transmits the license plate image to be verified to the verification server; The verification server identifies the license plate image to be verified to generate a license plate number to be verified, and determines whether the license plate number to be verified matches the registered license plate number in the registration data; When the verification server determines that the license plate number to be verified matches the registered license plate number in the registration data, the verification server sends a driving image request to the in-vehicle image capture device according to the device identification data in the registration data; When the in-vehicle image capture device receives the driving image request, the in-vehicle image capture device captures a driver of the vehicle to generate a facial image to be verified, and transmits the facial image to be verified and the device identification data to the authentication server; The verification server judges whether the similarity between the facial image to be verified and the blurred facial image in the registration data reaches a predetermined similarity threshold; When the verification server determines that the license plate number to be verified matches the registered license plate number in the registration data, and the similarity between the facial image to be verified and the blurred facial image in the registration data reaches the predetermined similarity threshold value, the authentication server generates a successful authentication result. The identity verification system as described in claim 11, wherein the registration information also includes payment information, and the terminal computer device not only transmits the license plate image to be verified to the verification server, but also transmits a consumption amount to the verification server. authentication server; After the verification server generates the successful verification result, it executes a payment procedure according to the payment information of the registration information and the consumption amount. The identity verification system as described in claim 11, wherein the verification server will The image of the face to be verified is deleted. The identity verification system as described in claim 11, wherein the verification server stores a plurality of security level data and a plurality of similarity thresholds respectively corresponding to the security level data, and the registration data also includes a security level setting data, the verification server selects from the similarity thresholds the corresponding security level data that matches the security level setting data as the predetermined similarity threshold. The identity verification system according to claim 11, wherein, before the in-vehicle image capturing device transmits the face image to be verified to the verification server, the face image to be verified is also blurred in the same manner as the original face image. verifying that the face image is at least partially blurred; Wherein, when the in-vehicle image capture device transmits the facial image to be verified to the verification server, it transmits the fuzzy facial image to be verified to the verification server; Wherein, when the verification server judges whether the similarity between the facial image to be verified and the blurred facial image in the registration data reaches the predetermined similarity threshold, it judges that the facial image to be verified after blurring Whether the similarity with the blurred facial image of the registration data reaches the predetermined similarity threshold.

Images