CN114387674A - Liveness detection method, system, device, storage medium and program product - Google Patents

Abstract

A method, system, device, storage medium, and program product for living body detection are provided. The method comprises the following steps: acquiring a video of a target object; analyzing the video to obtain the actually used recording white balance sequence information in the recording process of the video; matching the recorded white balance sequence information with target white balance sequence information, wherein the target white balance sequence information is used for indicating that the white balance parameters of the camera device are adjusted in the video recording process; and checking whether the target object is a living body according to the matching result of the recorded white balance sequence information and the target white balance sequence information. The living body detection method can detect whether the video sent by the terminal equipment is the attack video shot by the camera device after being hijacked, and has the advantages of wide application range (for example, covering various use scenes of the user and various equipment of the user), high detection efficiency (the user does not need to spend a long time), and the like.

Description

Liveness detection method, system, device, storage medium and program product

technical field

The present application relates to the technical field of data processing, and in particular, to a method, system, device, storage medium and program product for detecting a living body.

Background technique

Liveness detection technology is a defense method used to detect whether a face is a real person, preventing others from using screen remakes, printing paper remakes or masks to fake portraits, and has important applications in consumer finance, entertainment games and other fields. In recent years, the deception methods of black products have become more and more sophisticated. By hijacking cameras and other means, they can directly upload videos or pictures that have been taken in advance, so as to deceive the living body detection model. This attack method cannot be defended by simply adopting the live detection model. At present, the live action, lip language and colorful live body can be well defended against this attack, but there are still many problems with the action live body, the lip language live body and the colorful live body. For example, the action live body and the lip language live body take a long time. Interacting with the user, the detection efficiency is low. The colorful living body has higher requirements on the brightness of the environment, the screen film of the mobile phone, the brightness of the mobile phone screen, etc., and the application scope is narrow.

SUMMARY OF THE INVENTION

The present application provides a living body detection method, system, device, storage medium and program product, which can realize the living body detection of a target object. The specific technical solutions are as follows:

In a first aspect of the embodiments of the present application, a method for detecting a living body is first provided, and the method includes:

Get the video of the target object;

Analyze the video to obtain the recording white balance sequence information actually used in the recording process of the video;

Matching the recording white balance sequence information with the target white balance sequence information, the target white balance sequence information is used to instruct to adjust the white balance parameters of the camera during the recording process of the video;

According to the matching result between the recorded white balance sequence information and the target white balance sequence information, it is verified whether the target object is a living body.

In a second aspect of the embodiments of the present application, a method for detecting a living body is provided, which is applied to a terminal device, and the method includes:

Acquire target white balance sequence information from the server in response to a living body detection request for performing living body detection on the target object; wherein the target white balance parameter sequence information is used to instruct to adjust the white balance parameters of the camera during the video recording process ;

Controlling the camera device to perform video recording on the target object according to the instruction of the target white balance sequence information;

Send the recorded video of the target object to the server, and the server parses the video to obtain the recording white balance sequence information actually used in the recording process of the video, and records the video. The white balance sequence information is matched with the target white balance sequence information, and whether the target object is a living body is determined according to the matching result.

In a third aspect of the embodiments of the present application, a live detection system is also provided, and the system includes a terminal device and a server;

The terminal device is configured to obtain target white balance sequence information from a server in response to a living body detection request for performing living body detection on the target object, and control the camera device to perform video recording on the target object according to the instructions of the target white balance sequence information, sending the recorded video of the target object to the server; wherein, the target white balance parameter sequence information is used to instruct the white balance parameters of the camera to be adjusted during the video recording process;

The server is configured to receive the video of the target object, parse the video, obtain the recording white balance sequence information actually used in the recording process of the video, and associate the recording white balance sequence information with the target object. The white balance sequence information is matched, and whether the target object is a living body is verified according to the matching result of the recorded white balance sequence information and the target white balance sequence information.

In a fourth aspect of the embodiments of the present application, a device for detecting a living body is also provided, and the device includes:

a first acquisition module, used for acquiring the video of the target object;

a parsing module for parsing the video to obtain the recording white balance sequence information actually used in the recording process of the video;

a matching module, configured to match the recording white balance sequence information with target white balance sequence information, where the target white balance sequence information is used to instruct the white balance parameters of the camera to be adjusted during the video recording process;

A verification module, configured to verify whether the target object is a living body according to the matching result of the recorded white balance sequence information and the target white balance sequence information.

In a fifth aspect of the embodiments of the present application, another apparatus for detecting a living body is provided, which is applied to a terminal device, and the apparatus includes:

The third obtaining module is configured to obtain target white balance sequence information from the server in response to a living body detection request for performing living body detection on the target object; wherein, the target white balance parameter sequence information is used to indicate that the camera is to be recorded during the video recording process. Adjust the white balance parameters of the device;

a first control module, configured to control the camera device to perform video recording on the target object according to the instruction of the target white balance sequence information;

The second sending module is configured to send the recorded video of the target object to the server, and the server parses the video to obtain the recording white balance sequence actually used in the video recording process information, and match the recorded white balance sequence information with the target white balance sequence information, and determine whether the target object is a living body according to the matching result.

In a sixth aspect of the embodiments of the present application, an electronic device is also provided, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus;

memory for storing computer programs;

The processor is configured to implement the steps in the living body detection method described in the first aspect or the second aspect of the embodiments of the present application when executing the program stored in the memory.

In a seventh aspect of the embodiments of the present application, a computer-readable storage medium is also provided, where instructions are stored in the computer-readable storage medium, and when the computer-readable storage medium is run on a computer, the computer is made to execute the first embodiment of the present application. The steps in the biological detection method of one aspect or the second aspect.

In an eighth aspect of the embodiments of the present application, a computer program product is also provided, including a computer program/instruction, and when the computer program/instructions are executed by a processor, the first aspect or the second aspect of the embodiments of the present application is implemented. The steps in the live detection method.

Using the living body detection method of the present application, the video of the target object is obtained first, and then the video is analyzed to obtain the recording white balance sequence information actually used in the video recording process. Next, the recording white balance sequence information is matched with the target white balance sequence information, and the target white balance sequence information is used to instruct to adjust the white balance parameters of the camera during the video recording process. Finally, according to the matching result of the recorded white balance sequence information and the target white balance sequence information, it is verified whether the target object is a living body. The living body detection method of the present application has the following multiple effects:

1. It can detect whether the video of the target object is an attack video shot after the camera device was hijacked.

2. The live detection method of the present application has a wider application range. On the one hand, it no longer requires the intensity of light (white balance has nothing to do with the intensity of light), it can still take effect in an environment with strong light, and can cover the user's various On the other hand, there is no longer any requirement for the light transmittance of the protective film of the device (such as mobile phone film) (white balance has nothing to do with the light transmittance of the film), which can cover various devices of the user; on the other hand, It is no longer necessary to brighten the screen of the device (white balance has nothing to do with screen brightness), which can avoid dazzling users and optimize the user experience.

3. The living body detection method of the present application has high detection efficiency, does not require users to spend a long time to interact, and at the same time reduces the difficulty for users to use, avoids individual users from being unable to complete the detection due to too much difficulty, and improves the overall detection rate .

Description of drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the accompanying drawings that are required to be used in the description of the embodiments or the prior art.

FIG. 1 is a schematic diagram of an implementation environment shown in an embodiment of the present application;

2 is a flowchart of a method for detecting a living body according to an embodiment of the present application;

3 is a flowchart of another method for detecting a living body according to an embodiment of the present application;

FIG. 4 is a complete schematic flowchart of a method for detecting a living body according to an embodiment of the present application;

FIG. 5 is a structural block diagram of a living body detection device shown in an embodiment of the present application;

6 is a structural block diagram of another living body detection apparatus shown in an embodiment of the present application;

FIG. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed ways

The technical solutions in the embodiments of the present application will be described below with reference to the accompanying drawings in the embodiments of the present application.

In recent years, important progress has been made in the research of artificial intelligence-based computer vision, deep learning, machine learning, image processing, image recognition and other technologies. Artificial Intelligence (AI) is an emerging science and technology that studies and develops theories, methods, technologies and application systems for simulating and extending human intelligence. Artificial intelligence is a comprehensive discipline, involving chips, big data, cloud computing, Internet of Things, distributed storage, deep learning, machine learning, neural networks and many other types of technologies. As an important branch of artificial intelligence, computer vision is to let machines recognize the world. Computer vision technology usually includes face recognition, living detection, fingerprint recognition and anti-counterfeiting verification, biometric recognition, face detection, pedestrian detection, target detection, pedestrian detection Recognition, Image Processing, Image Recognition, Image Semantic Understanding, Image Retrieval, Text Recognition, Video Processing, Video Content Recognition, Behavior Recognition, 3D Reconstruction, Virtual Reality, Augmented Reality, Simultaneous Localization and Mapping (SLAM), Computational Photography, Robotics Navigation and positioning technology. With the research and progress of artificial intelligence technology, this technology has been applied in many fields, such as security, city management, traffic management, building management, park management, face access, face attendance, logistics management, warehouse management, robotics , smart marketing, computational photography, mobile imaging, cloud services, smart home, wearable devices, driverless, autonomous driving, smart medical care, face payment, face unlock, fingerprint unlock, witness verification, smart screen, smart TV, Cameras, mobile Internet, webcasting, beauty, beauty, medical beauty, intelligent temperature measurement and other fields.

In the related art, the motion detection and the lip language detection require the user to perform a certain action according to the prompt, or to speak a random number on the screen, which is often impossible to achieve with photos. However, the interaction time of this type of detection is too long, and it is too difficult for some users to complete, resulting in a low overall detection rate.

In the dazzling live detection, the screen of the user terminal will emit light of various colors according to the random sequence sent by the server, and the server will compare the light sequence in the recorded video and the light sequence sent by the server when verifying the live body. Consistent, so as to determine whether the video is real shot, not hijacked injection. However, the colorful live detection has many defects, such as: (1) The screen lighting has very high requirements on the environment, and it must be effective in a dark environment; (2) The requirements for mobile phone screen film are high, and the light transmittance is not good. A good mobile phone film does not work well; (3) In order to make it as bright as possible, it is necessary to brighten the mobile phone screen and shorten the distance between the face and the mobile phone screen, causing dazzling.

In order to solve the above problems, the present application provides a white balance living body detection method, and the white balance living body detection method of the present application will be described in detail below.

FIG. 1 is a schematic diagram of an implementation environment according to an embodiment of the present application. In FIG. 1 , an SDK (Software Development Kit, software development kit) is installed inside the terminal device, and the terminal device can communicate and interact with the server through the SDK. The terminal device may be a smart phone, tablet, computer or other portable smart devices, and the server may be a server or a cluster of multiple servers. This embodiment does not specifically limit the types of terminal devices and servers.

The present application first provides a liveness detection method, which can be applied to the server in FIG. 1 or the terminal device in FIG. 1 . FIG. 2 is a flowchart of a method for detecting a living body according to an embodiment of the present application. Referring to FIG. 2 , the method for detecting a living body of the present application may specifically include the following steps:

Step S21: Obtain the video of the target object.

In this embodiment, the target object may be a human object, or may be other living objects (eg, animal objects) that are not human objects.

The video may be a video containing a specified area of the target object (area for living body detection). Taking the target object being a human object as an example, the video may be a video including a face area, a video including an upper body area, or a video including a hand area, which is not specifically limited in this embodiment.

When the living body detection method is applied to the server, the server can collect the video of the target object through other external camera devices connected to itself, and can also receive the video of the target object collected by the terminal device. When the living body detection method is applied to the terminal device, the terminal device can collect the video of the target object through its own built-in camera, and can also collect the video of the target object through other external camera connected to itself.

Step S22: Analyze the video to obtain recording white balance sequence information actually used in the video recording process.

In this embodiment, the camera generally uses a white balance sequence when recording a video. A white balance sequence contains multiple white balance parameters, each of which can be used to set the white balance of the camera device.

By analyzing the video in step S22, the sequence information of the white balance sequence actually used in the video recording process can be obtained.

Step S23: Match the recording white balance sequence information with target white balance sequence information, where the target white balance sequence information is used to instruct to adjust the white balance parameters of the camera during the video recording process.

In this embodiment, the target white balance sequence information is sequence information of a pre-designated white balance sequence used for anti-counterfeiting authentication. The purpose of the target white balance sequence is to instruct the camera device to adjust the white balance parameters when the camera device records video.

After obtaining the sequence information of the white balance sequence actually used in the video recording process, compare the sequence information of the actually used white balance sequence (recording white balance sequence information) with the sequence information of the pre-specified white balance sequence used for anti-counterfeiting authentication (target white balance sequence information) for matching to obtain a matching result.

Step S4: Check whether the target object is a living body according to the matching result between the recorded white balance sequence information and the target white balance sequence information.

In this embodiment, according to the matching result, if the matching degree between the recorded white balance sequence information and the target white balance sequence information is greater than a preset threshold, it can be determined that the target object is a living body, or it is determined that the target object is a living body and the video is legal, It was not filmed after the camera device was illegally hijacked. If the matching degree between the recorded white balance sequence information and the target white balance sequence information is not greater than the preset threshold, it can be determined that the target object living body detection has failed, and the user can be prompted to re-shoot the video.

In one embodiment, the execution device that executes the liveness detection method may install application software for anti-counterfeiting verification (hereinafter referred to as application A), and the application software may be a plurality of application software requiring security protection (hereinafter referred to as application B). ) for legal verification. When the user needs to log in to application B, application B obtains the target white balance sequence information from application A, sets the white balance of the camera device according to the target white balance sequence information, then captures the user's video, and sends the captured video to application A , Application A parses the video to obtain the recording white balance sequence information actually used in the video recording process, and matches the recording white balance sequence information with the target white balance sequence information sent to Application B in advance. If the matching degree between the sequence information and the target white balance sequence information is greater than the preset threshold, it can be determined that the video contains a living body and the video is legal. If the camera device is illegally hijacked, the target white balance sequence pre-sent by application A is not used when actually recording the video, and the match between the final recorded white balance sequence information and the target white balance sequence information is not greater than the preset threshold.

In one embodiment, the target white balance sequence may be the only white balance sequence pre-specified by the user for anti-counterfeiting authentication. When a user is using an application software, if the application software needs to verify the user's body, the application software can randomly generate a white balance sequence, and then send the white balance sequence to the camera device. Balance sequence information for white balance settings. The camera device sends the collected video to the application software, and the application software analyzes the video to obtain the recording white balance sequence information actually used in the video recording process. Specify the only white balance sequence for anti-counterfeiting authentication, then it can be determined that the video is illegal. In other words, even if the actually used recording white balance sequence information is consistent with the white balance sequence information received by the imaging device, as long as the recording white balance sequence information is inconsistent with the target white balance sequence information, it can be determined to be illegal. For example, when parents use some application software, they can pre-designate a white balance sequence for anti-counterfeiting authentication. In this way, children do not know the white balance sequence for anti-counterfeiting authentication. Even if the camera device is not illegally hijacked, they cannot Use these applications through normal operations, for example, children cannot log in to these applications or make payments.

With the living body detection method of the present application, the video of the target object is obtained first, and then the video is analyzed to obtain the recording white balance sequence information actually used in the video recording process. Next, the recording white balance sequence information is matched with the target white balance sequence information, and the target white balance sequence information is used to instruct to adjust the white balance parameters of the camera during the video recording process. Finally, according to the matching result of the recorded white balance sequence information and the target white balance sequence information, it is verified whether the target object is a living body. The living body detection method of the present application has the following multiple effects:

1. It can detect whether the video of the target object is an attack video shot after the camera device was hijacked.

2. The live detection method of the present application has a wider application range. On the one hand, it no longer requires the intensity of light (white balance has nothing to do with the intensity of light), it can still take effect in an environment with strong light, and can cover the user's various On the other hand, there is no longer any requirement for the light transmittance of the protective film of the device (such as mobile phone film) (white balance has nothing to do with the light transmittance of the film), which can cover various devices of the user; on the other hand, It is no longer necessary to brighten the screen of the device (white balance has nothing to do with screen brightness), which can avoid dazzling users and optimize the user experience.

3. The living body detection method of the present application has high detection efficiency, does not require users to spend a long time to interact, and at the same time reduces the difficulty for users to use, avoids individual users from being unable to complete the detection due to too much difficulty, and improves the overall detection rate .

In combination with the above embodiments, in one embodiment, the method for detecting a living body of the present application may further include the following steps:

Obtain white balance parameter information supported by the camera; wherein the white balance parameter information includes color temperature information corresponding to the white balance parameter; according to the color temperature information of the white balance parameter supported by the camera, The device generates at least one to-be-selected white balance sequence; wherein, the to-be-selected white balance sequence is used by the camera device to adjust white balance parameters when recording the video of the target object.

In this embodiment, the executing device may acquire the system (any type of system) version of the device to which the camera device belongs, and acquire white balance parameter information supported by the camera device according to the system version.

For example, taking the system of the device to which the camera device belongs is the Android system as an example, if the system version is less than 5.0 (the specific value of the white balance parameter that the user actively specifies) is not supported, then the white balance parameter information supported by the camera device is: Supported white balance list (including multiple white balance parameters), color temperature information corresponding to each white balance parameter in the white balance list. If the system version is not less than 5.0 (the user can actively specify the specific value of the white balance parameter), then the white balance parameter information supported by the camera device is: the white balance parameter range supported by the camera device and the color temperature range corresponding to the white balance parameter range. Among them, each white balance parameter corresponds to a unique color temperature value.

Of course, in actual implementation, the white balance parameter information supported by the camera device may also be obtained in other ways, which is not specifically limited in this embodiment.

After obtaining the white balance parameter information supported by the camera device, the executing device generates at least one white balance sequence to be selected for the camera device. The role of the white balance sequence to be selected is to instruct the camera device to adjust the white balance parameters when the camera device records the video of the target object.

When generating a white balance sequence to be selected, for each white balance sequence to be selected, the difference between the color temperatures corresponding to two adjacent white balance parameters can be greater than the target threshold, so as to avoid being unable to identify whether the video is used when the color temperature difference is small The white balance parameter is set to improve the accuracy of the recognition result.

In this embodiment, the white balance parameter information supported by the camera device can be obtained through the system version of the device to which the camera device belongs, and then the white balance sequence to be selected for instructing the camera device to adjust the white balance parameters when recording video can be obtained, which is a living body The smooth implementation of the detection method provides a guarantee.

In combination with the above embodiments, in an implementation manner, when the liveness detection method of the present application is applied to a server, before acquiring the video of the target object, the method may further include:

The at least one to-be-selected white balance sequence is sent to the terminal device, and the camera device of the terminal device adjusts the white balance parameters according to the at least one to-be-selected white balance sequence when recording the video.

Accordingly, obtain the video of the target object, including:

The video of the target object sent by the terminal device is received.

In this embodiment, after generating at least one to-be-selected white balance sequence, the server may send the at least one to-be-selected white balance sequence to the terminal device. The terminal device then determines a white balance sequence to be selected from the received at least one white balance sequence to be selected, and uses the determined white balance sequence to be selected during the process of recording the video of the target object by the camera of the terminal device. The camera device adjusts the white balance parameters.

In this embodiment, after parsing the video of the target object sent by the terminal device, if the server finds that the video does not use any white balance sequence, it means that the terminal device does not use the white balance sequence given by the server to capture the video according to the server's pre-requirements. Video of the target object. At this time, the subsequent process is not continued, and the server may directly send a notification message to the terminal device to remind the terminal device to use the white balance sequence given by the server to re-capture the video of the target object.

In actual implementation, even if the terminal device uses the white balance sequence when recording video, it does not necessarily mean that the terminal device uses the white balance sequence pre-given by the server when recording video, so the recorded video may also be illegal. . Therefore, after receiving the video of the target object sent by the terminal device, the server needs to detect whether the recorded white balance sequence actually used by the terminal device is a white balance sequence in at least one white balance sequence to be selected sent to it in advance. Only when the actually used recorded white balance sequence is the white balance sequence in at least one to-be-selected white balance sequence sent to it in advance can it be determined that the recorded video is legal, or that the target object is a living body.

In this embodiment, a pre-negotiated key may be used for interaction between the server and the terminal device, thereby ensuring communication security.

In this embodiment, if the server sends a to-be-selected white balance sequence to the terminal device, the terminal device determines the to-be-selected white balance sequence as the white balance sequence to be used, and then records the video of the target object on the camera device. , using each white balance parameter in the to-be-selected white balance sequence to adjust the white balance of the imaging device. During subsequent comparison, the server matches the information of the recorded white balance sequence actually used in the received video with the information of the pre-sent white balance sequence to be selected (target white balance sequence).

In this embodiment, if the server sends multiple to-be-selected white balance sequences to the terminal device, the terminal device determines any one of the multiple to-be-selected white balance sequences as the white balance sequence that needs to be used, Then, in the process of recording the video of the target object by the camera device, the white balance sequence to be used is used to adjust the white balance of the camera device. In the subsequent comparison, the server matches the information of the recorded white balance sequence actually used in the received video with the information of multiple pre-sent white balance sequences to be selected in turn. If one match is successful, it means that the video contains a living body and is legal .

For example, when user A transfers money from a bank card through the network, the backend server of the bank prompts the user to perform face video authentication. The user terminal obtains a plurality of white balance sequences to be used sent by the background server, determines a white balance sequence to be used among the plurality of white balance sequences to be used, and uses the camera to collect the face video of user A, according to the determination Each white balance parameter in the white balance sequence to be used is obtained to adjust the white balance of the camera device to obtain a recorded face video. Next, the user terminal sends the face video to the server. After receiving the face video, the server parses the face video to obtain the information of the recorded white balance sequence actually used in the face video, and compares the information of the recorded white balance sequence with the information of the multiple pre-sent white balance sequences to be selected. Matches are performed in sequence. If one match is successful, it means that the video is legal, not an attack video, allowing the user to continue with subsequent operations.

In this embodiment, if the system of the terminal device is the Android system, when the system version is less than 5.0 (the specific value of the white balance parameter cannot be actively specified by the user), the terminal device will use the information of the system API camera1 and the white balance supported by the current device. The list (including multiple white balance parameters) and the color temperature values corresponding to each white balance parameter in the white balance list are sent to the server as white balance parameter information supported by the camera device.

When the system version is not less than 5.0 (which supports the user to actively specify the specific value of the white balance parameter), the terminal device uses the system API camera2 information, the white balance parameter range supported by the current device, and the color temperature range corresponding to the white balance parameter range as the camera device. The supported white balance parameter information is sent to the server.

After receiving the white balance parameter information supported by the camera device, the server generates at least one white balance sequence to be selected according to the information. For details, refer to the foregoing description. In this embodiment, a terminal device can initiate a liveness detection request to the server, and the server processes the liveness detection request, so that the liveness detection method of the present application can be applied to any scene of interaction between a server and a terminal device, and the application of the liveness detection method is expanded. scope.

In combination with the above embodiments, in an implementation manner, if the number of white balance sequences to be selected sent by the server to the terminal device is multiple, before matching the recorded white balance sequence information with the target white balance sequence information, this The applied biometric method may also include:

The to-be-selected white balance sequence sent by the terminal device when recording the video of the target object is received, and used as the target white balance sequence.

In this embodiment, the server sends multiple to-be-selected white balance sequences to the terminal device, and the terminal device selects any one of the multiple to-be-selected white balance sequences as the actual white balance sequence to be used. In the subsequent comparison, if the server matches the information of the recorded white balance sequence actually used in the received video with the information of multiple pre-sent white balance sequences to be selected in turn, it will take a lot of time, resulting in the efficiency of living body detection. Low.

In fact, the actual white balance sequence that needs to be used determined by the terminal device is the target white balance sequence. Therefore, if the terminal device sends the determined white balance sequence that actually needs to be used to the server after recording the video, the server can directly use the received white balance sequence as the target white balance sequence, and then use the target white balance sequence as the target white balance sequence. Matching the information of the balance sequence with the information of the recorded white balance sequence used in the video will greatly reduce the matching time and improve the efficiency of live detection.

In this embodiment, the server can directly use the white balance sequence to be used when recording the video of the target object sent by the terminal device as the target white balance sequence, which can significantly improve the living body detection efficiency.

In combination with the above embodiments, in an implementation manner, the video is parsed to obtain the recording white balance sequence information actually used in the recording process of the video, which may specifically include:

extracting a plurality of video frames from the video;

According to the color value of the target pixel on each of the video frames, the color temperature change information corresponding to the recording white balance sequence is obtained; wherein, the target pixel is all the pixels of the video frame or the target pixel is The pixel points in the target area of the video frame; the target area is an area used for living body detection.

In this embodiment, the execution device first decomposes the video into single frames. For each video frame, the color value (eg, RGB value) of the target pixel is extracted, and then a corresponding color temperature value is calculated according to the color value. In this way, a plurality of video frames can be calculated to obtain a plurality of color temperature values, and the plurality of color temperature values can be drawn into a video color temperature change curve in sequence (that is, the color temperature change information corresponding to the recording white balance sequence).

On this basis, matching the color temperature change information corresponding to the recorded white balance sequence with the color temperature change information corresponding to the target white balance sequence may include:

Match the color temperature change information corresponding to the recording white balance sequence with the color temperature change information corresponding to the target white balance sequence. In this embodiment, one white balance parameter uniquely corresponds to one color temperature value. Therefore, according to each white balance parameter in the target white balance sequence, a color temperature change curve corresponding to the target white balance sequence can be drawn.

Next, compare the color temperature change curve corresponding to the recorded white balance sequence with the color temperature change curve corresponding to the target white balance sequence. If the similarity is greater than the preset threshold, it means that the recorded white balance sequence matches the target white balance sequence, and then indicates that The recording white balance sequence information matches the target white balance sequence information.

The target pixels may be all pixels of the video frame, or may be pixels in the target area of the video frame. The target area refers to an area for performing liveness detection. In actual implementation, the target pixel point may be defined according to actual requirements, which is not specifically limited in this embodiment. In this embodiment, whether the video recorded by the terminal device is legal is detected according to the similarity between the color temperature change information corresponding to the recorded white balance sequence and the color temperature change information corresponding to the target white balance sequence, and the video can be quickly and accurately detected. Whether it is real filming, not hijacked injection. In combination with the above embodiments, in an implementation manner, the target area may be the face area or the palm area of the target object. On this basis, according to the color value of the target pixel on each of the video frames, the color temperature change information corresponding to the recording white balance sequence is obtained, which may specifically include:

Obtain the color temperature change information corresponding to the recorded white balance sequence according to the color value of the pixel in the face area of the target object on each of the video frames; or

According to the color values of the pixels in the palm area of the target object on each of the video frames, the color temperature change information corresponding to the recorded white balance sequence is obtained.

In some application scenarios, when a user logs in to certain accounts, when the user needs to perform liveness detection on the user, the liveness detection must be performed on the user himself (the user to which the account belongs), but not on other people. For example, when making a payment, the user who performs the live detection must be the user to which the payment account belongs, and cannot be any other user. In this scenario, in addition to detecting whether the recorded white balance sequence information matches the target white balance sequence information, it is also necessary to detect whether the human object in the video is the user himself. This requires that the video of the target object must contain the identifying features of the target object, such as facial features, eye features, palm print features, etc.

Therefore, in order to realize the living body detection in this scene, the target area can be set as the face area or the palm area. In this way, when obtaining the color temperature change information corresponding to the recording white balance sequence according to the color value of the target pixel on each video frame, it can be obtained according to the color value of the pixel in the face area of the target object on each video frame. The color temperature change information corresponding to the recorded white balance sequence can also be obtained according to the color values of the pixels in the palm area of the target object on each video frame to obtain the color temperature change information corresponding to the recorded white balance sequence.

With this embodiment, the user's living body detection can be implemented in a living body detection scenario that has stricter requirements on the user's identity, thereby expanding the scope of application of the living body detection method of the present application.

In combination with the above embodiments, in one embodiment, when the target pixel is a pixel in the target area of the video frame, the color value of the target pixel on each video frame can be obtained in the following manner:

performing keypoint detection on each of the plurality of video frames;

According to the key point detection result, frame-select the target area from each video frame in the plurality of video frames;

Extracting the color value of each pixel in the frame-selected target area on each of the plurality of video frames.

In this embodiment, the key points may be facial features of a human face, and the target area may be a human face area.

In specific implementation, only the color value of each pixel of the face area in the video frame can be extracted, and the color temperature value corresponding to the face area can be calculated, so as to avoid the influence caused by the change of the background (for example, the change of the color temperature in the background is not due to the white balance caused by the setting, or the change of color temperature cannot be recognized in the background, which will eventually affect the detection accuracy).

During specific implementation, any key point detection means and target area identification means may be adopted, which are not specifically limited in this embodiment.

In this embodiment, when extracting the color values of multiple pixels on each video frame, only the color values of multiple pixels in the target area (for example, the face area) can be extracted, so as to avoid the influence caused by the background change , to improve the accuracy of detection results.

With reference to the above embodiments, in an implementation manner, before acquiring the video of the target object, the method for detecting a living body of the present application may further include:

determining a white balance sequence to be selected from the at least one white balance sequence to be selected whenever a living body detection request is received;

The determined to-be-selected white balance sequence is used as the target white balance sequence used by the camera device to record the video, wherein the two to-be-selected white balance sequences determined in two adjacent times are different.

In this embodiment, the server may generate at least one to-be-selected white balance sequence in advance, and each time a living body detection request is received, determine a to-be-selected white balance sequence from the at least one to-be-selected white balance sequence, and the determined A candidate white balance sequence is the white balance sequence to be used in response to the current liveness detection request.

In actual implementation, it is not arbitrary to determine a white balance sequence to be selected from at least one white balance sequence to be selected. Molecular fake white balance sequence.

The to-be-selected white balance sequence determined by the execution device in the at least one to-be-selected white balance sequence, that is, the target white balance sequence.

Taking the server processing the liveness detection request sent by the terminal device as an example, when the server receives the liveness detection request sent by the terminal device, it can send the first white balance sequence to be selected to the terminal device, so that when the terminal device uses the camera to record the video, Using the white balance parameters in the first to-be-selected white balance sequence, perform a white balance setting on the camera device for the camera device. When receiving a new living body detection request sent by the terminal device, the server may send the second to-be-selected white balance sequence of the camera to the terminal device, so that the terminal device uses the second to-be-selected white balance sequence when recording video with the camera device. Set the white balance of the camera device with the white balance parameter. The first to-be-selected white balance sequence and the second to-be-selected white balance sequence are both the to-be-selected white balance sequences pre-generated by the server. And the first white balance sequence to be selected is different from the second white balance sequence to be selected.

In this embodiment, the camera device can only use one white balance sequence when recording a live video. If a live video recording fails, the video recorded this time and the white balance sequence used (the first white balance to be selected) are discarded. sequence), again using a new white balance sequence (the second candidate white balance sequence) to prevent an attacker from hijacking the camera and using the historically sent white balance sequence for video recording. In this way, the server sends a new white balance sequence to be selected each time it receives a live detection request sent by the terminal device, and the previously sent white balance sequences to be selected will be invalidated, which can prevent attacks The attacker hijacks the camera and uses the historically sent white balance sequence to record video.

When the terminal device sends a living body detection request to the server, the server sends the determined target white balance sequence to the terminal device, so that the terminal device uses each white balance parameter in the target white balance sequence to match the target white balance sequence in the process of recording the video of the target object. The camera device performs white balance parameter setting. When the subsequent server analyzes the video sent by the terminal device, it can directly match the information of the recorded white balance sequence actually used in the video with the information of the target white balance sequence. In this way, the server does not need to compare the information of the recorded white balance sequence actually used in the video with the information of each pre-generated white balance sequence to be selected, which improves the living body detection efficiency, and does not need to use the white balance sequence sent by the terminal device. The target white balance sequence is determined by the sequence, which can prevent the terminal device from sending a fake target white balance sequence and ensure the smooth implementation of the living body detection. In this embodiment, each time the execution device receives a living body detection request, it determines a to-be-selected white balance sequence from at least one pre-generated to-be-selected white balance sequence, and the to-be-selected white balance sequence is different from the last response The to-be-selected white balance sequence determined by the living body detection request can prevent attackers from hijacking the camera device and use the historically sent to-be-selected white balance sequence for video recording, thereby ensuring the effective implementation of the living body detection method.

For the scenario where the server processes a liveness detection request initiated by a terminal device, the present application further provides a liveness detection method, which is applied to the terminal device in FIG. 1 , and is specifically applied to the SDK in the terminal device. FIG. 3 is a flowchart of another method for detecting a living body according to an embodiment of the present application. Referring to FIG. 3 , the method for detecting a living body of the present application may include the following steps:

Step S31: In response to a request for in vivo detection of the target object, obtain target white balance sequence information from the server; wherein the target white balance parameter sequence information is used to indicate the white balance of the camera during the video recording process parameters are adjusted.

Step S32: Control the camera to record video on the target object according to the instruction of the target white balance sequence information.

Step S33: Send the recorded video of the target object to the server, and the server parses the video to obtain the recording white balance sequence information actually used in the recording process of the video, and converts the recorded white balance sequence information to the server. The recorded white balance sequence information is matched with the target white balance sequence information, and whether the target object is a living body is determined according to the matching result.

In this embodiment, the living body detection request may be triggered by the user through the terminal device, or may be sent by the server to the terminal device.

After triggering the living body detection request, the server will send the target white balance sequence information to the terminal device, so that the terminal device can set the white balance of the camera according to each white balance parameter in the target white balance sequence during the process of recording the video of the target object. .

For example, in a camera device, a user needs to perform face verification when logging into an application, and the background server of the application actively sends a living body detection instruction to the terminal device where the user is located, and simultaneously sends the target white balance sequence to the terminal device. During the process of recording the user's face video, the terminal device uses each white balance parameter in the target white balance sequence to set the white balance of the camera device.

After completing the recording of the face video, the terminal device sends the recorded face video to the server, so that the server can verify whether the video is legal. For details about the process of the server verifying the video, please refer to the previous section.

With the living body detection method of the present application, the terminal device obtains target white balance sequence information from a server in response to a living body detection request for performing living body detection on a target object; wherein, the target white balance parameter sequence information is used to indicate the Adjust the white balance parameters of the camera. Next, the terminal device controls the camera to record the video of the target object according to the instruction of the target white balance sequence information; and then sends the recorded video of the target object to the server. The server analyzes the video, obtains the recording white balance sequence information actually used in the video recording process, matches the recording white balance sequence information with the target white balance sequence information, and determines whether the target object is a living body according to the matching result.

1. It can detect whether the video of the target object is an attack video shot after the camera device was hijacked.

2. The live detection method of the present application has a wider application range. On the one hand, it no longer requires the intensity of light (white balance has nothing to do with the intensity of light), it can still take effect in an environment with strong light, and can cover the user's various On the other hand, there is no longer any requirement for the light transmittance of the protective film of the device (such as mobile phone film) (white balance has nothing to do with the light transmittance of the film), which can cover various devices of the user; on the other hand, It is no longer necessary to brighten the screen of the device (white balance has nothing to do with screen brightness), which can avoid dazzling users and optimize the user experience.

3. The living body detection method of the present application has high detection efficiency, does not require users to spend a long time to interact, and at the same time reduces the difficulty for users to use, avoids individual users from being unable to complete the detection due to too much difficulty, and improves the overall detection rate .

With reference to the above embodiments, in an implementation manner, the controlling the camera device to perform video recording of the target object according to the instruction of the target white balance sequence information may include:

In response to the living body detection request, a white balance parameter is sequentially extracted from the target white balance sequence every target duration; the camera device is controlled to record a single-frame video or a continuous multi-frame video according to the extracted white balance parameter .

In this embodiment, a white balance sequence includes multiple white balance parameters. Since the camera device only supports setting one white balance parameter at the same time, in this embodiment, every target time period is selected from the target white balance sequence in order. Extract the next unused white balance parameter, and set the white balance of the camera device accordingly, and then perform continuous multi-frame video recording within the target duration according to the newly set white balance parameter.

For example, the target white balance sequence sent by the server in advance is: S1, S2, S3, ..., Sn, and the target duration is 2 seconds, then the user terminal first sets the white balance of the camera according to S1 at time t0, and then at t0- Capture and record video between (t0+2S), then, the terminal device sets the white balance of the camera according to S2 at (t0+2S) time, and then captures and records video between (t0+2S)-(t0+4S), And so on, until the video of the preset duration is captured. In this embodiment, it is also possible to sequentially extract the next unused white balance parameter from the target white balance sequence every target duration, and set the white balance of the camera device accordingly, and then set the white balance according to the newly set white balance. parameters to record a single frame of video.

For example, the target white balance sequence sent by the server in advance is: S1, S2, S3, ..., Sn, and the target duration is 2 seconds, then the user terminal first sets the white balance of the camera according to S1 at time t0, and then at time t0 Collect a video frame, and restore the white balance of the camera device to the state before setting after the video frame is collected. Next, the user terminal sets the white balance of the camera device according to S2 at time (t0+2S), then collects a video frame at time (t0+2S), and restores the white balance of the camera device to the state before setting after collecting the video frame. . And so on, until the video of the preset duration is captured.

In this embodiment, it is also possible to sequentially extract the next unused white balance parameter from the target white balance sequence every N video frames, and set the camera device accordingly, and then set the white balance parameter according to the newly set white balance parameter. Record continuous multi-frame video.

For example, the target white balance sequence sent by the server in advance is: S1, S2, S3, ..., Sn. The terminal device first sets the white balance of the camera according to S1, and then collects the first to fifth video frames, and then, After the 5th frame of video is collected, the terminal device sets the white balance of the camera according to S2, and then collects the 6th to 10th video frames, and so on, until the video of the preset duration is collected.

In this embodiment, each time a white balance sequence is used, it is not necessary to use up all the white balance parameters in the white balance sequence, as long as a video of a preset duration is recorded. If all the white balance parameters in the white balance sequence have been used up before recording the video of the preset length, you can either use the white balance parameters in another new white balance sequence, or you can no longer use the white balance parameters , or start over again with the white balance parameters in the white balance sequence.

In this embodiment, a white balance parameter is sequentially extracted from the target white balance sequence every target duration, and the camera is controlled to record a single-frame video or continuous multi-frame video according to the extracted white balance parameter, so that the The server can better identify the white balance parameters used in the recorded video (for example, a white balance parameter may not be accurately identified by the server when it is only used for a short period of time), thereby improving the accuracy of the server's liveness detection .

In combination with the above embodiments, in an implementation manner, during the control of the camera device to perform video recording of the target object according to the instruction of the target white balance sequence information, the living body detection method of the present application may further include the following steps:

Detecting whether the quality parameter of each video frame recorded by the camera device meets the target quality parameter; when detecting that the quality parameter of a video frame recorded by the camera device does not meet the target quality parameter, controlling the camera device to stop video recording, discarding the recorded video; and acquiring new target white balance sequence information from the server.

On this basis, controlling the camera device to record a single-frame video or continuous multi-frame video according to an extracted white balance parameter, including:

The camera is controlled to re-record the video of the target object according to the instruction of the new target white balance sequence information.

In this embodiment, the terminal device will detect in real time whether the quality parameters of each frame of video recorded by the camera device meet the target quality parameters, and if so, will retain the frame of video data, as long as it detects the quality of a video frame recorded by the camera device. If the quality parameter does not meet the target quality parameter, the terminal device will stop recording video, discard the recorded video, and re-record the video according to another target white balance sequence sent by the server.

The quality parameters may be measured by using multiple factors, such as clarity, alignment of the face and the face frame, brightness, etc. The selection and evaluation of the quality parameters are not specifically limited in this embodiment.

In this embodiment, if a certain recording of the live video fails (for example, the quality parameter of the video frame is unqualified), the video recorded this time and the target white balance sequence used at the time are discarded, and a new target white balance sequence is used again. Setting the white balance of the camera device can effectively prevent attackers from hijacking the camera device and recording video using the white balance sequence sent in the history.

In combination with the above embodiments, in an implementation manner, when the target white balance sequence information obtained from the server is multiple, the living body detection method of the present application may further include:

Sending the selected target white balance sequence to the server.

In this embodiment, if there are multiple target white balance sequences sent by the server to the terminal device, it will take a long time for the server to compare the information of the recorded white balance sequence with the information of each white balance sequence sent in advance. , reducing the detection efficiency. In order to improve the detection efficiency, after the terminal device selects a white balance sequence from the multiple white balance sequences sent by the server to record the video, it can send the used white balance sequence to the server. In this way, after receiving the video sent by the terminal device and a selected white balance sequence, the server can compare the information of the recorded white balance sequence used in the video with a white balance sequence sent by the terminal device (the selected white balance sequence) target white balance sequence), and if there is a match, the video is determined to be legitimate.

The method for detecting a living body of the present application will be described in detail below with reference to FIG. 1 and FIG. 4 with a complete embodiment. In this embodiment, the liveness detection request of the terminal device is processed by the server. FIG. 4 is a schematic diagram of a complete flow of a method for detecting a living body according to an embodiment of the present application.

Step 1: After detecting the operation of face verification triggered by the user, the SDK of the terminal device acquires multiple white balance parameters or white balance parameter ranges supported by the camera device of the terminal device. Then, a face verification request carrying these white balance parameters or white balance parameter ranges is sent to the server.

If the system of the terminal device is Android and the version is less than 5.0, the SDK selects API camera1, obtains multiple white balance parameters supported by the camera device, and then sends the information carrying API camera1, multiple white balance parameters and each white balance to the server. A face verification request for the color temperature value corresponding to the balance parameter (the color temperature value corresponding to each white balance parameter may not be carried, but determined by the server itself).

If the system of the terminal device is Android, and the version is not less than 5.0, the SDK selects API camera2, obtains the white balance parameter range supported by the camera device, and then sends the information carrying API camera2, white balance parameter range and white balance parameter to the server. The face verification request for the color temperature range corresponding to the range (the color temperature range corresponding to the white balance parameter range may also not be carried and determined by the server itself).

Among them, when the SDK sends the face verification request, it can be encrypted before sending, and the key can be determined by the SDK and the server through negotiation in advance.

Step 2: After the server receives the face verification request, it decrypts and obtains the current white balance information of the terminal device. Different white balance parameters correspond to different color temperature values. The server further selects the white balance parameter with a large color temperature difference according to the color temperature difference. Generate at least one white balance sequence. Wherein, in each white balance sequence, the difference between the color temperature values corresponding to two adjacent white balance parameters is greater than the target threshold.

Next, the server encrypts at least one white balance sequence and sends it to the SDK.

Step 3: After the SDK receives at least one white balance sequence, it decrypts it, then opens the face detection page, and calls the front camera of the terminal device to perform face quality detection (the face does not need to be moved). If the detection fails, Remind the user to make face adjustments, such as adjusting the orientation, or placing the face in the face frame.

Step 4: After the quality of the face meets the standard, the SDK selects a white balance sequence to be used in the at least one white balance sequence obtained by decryption. According to the sequence of each white balance parameter in the selected white balance sequence, every N frames , use a white balance parameter to set the white balance of the camera. For example, frames 1 to 5 are collected by the camera under the condition of white balance parameter 1, and for example, frames 6 to 10 are collected under the condition of white balance parameter 2 by the camera.

Step 5: In Step 4, the SDK continues to perform face quality detection on each video frame collected by the camera device, and pushes the video frame into the video after the detection is passed. If it is detected that the face quality of a certain video frame does not meet the standard, discard the currently captured video, select an unused white balance sequence from the at least one white balance sequence obtained by decryption, and repeat step 3-step 4Complete the video recording.

Step 6: After completing the white balance video recording, the SDK encrypts the recorded video and a selected white balance sequence, and then uploads it to the server.

Step 7: The server decrypts the data packet after receiving it, and obtains the recorded video and a white balance sequence selected when recording the video. Then, the server decomposes the video into a single frame, and extracts the RGB value of the face area in each frame of the image (a face frame will be displayed on the screen during the face quality detection to ensure that the image of the face area is extracted. In When extracting the RGB value of the face area, only the RGB value of the area within the face frame is extracted according to the coordinates of the face frame), and the corresponding color temperature value is calculated according to the RGB value. In this way, multiple frames of video images can obtain multiple color temperature values. . Next, draw all the color temperature values into a color temperature change curve according to the collection order. Finally, compare the color temperature change curve corresponding to the video with the color temperature change curve corresponding to the white balance sequence (target white balance sequence) sent by the terminal device. If the similarity is greater than the threshold, the verification is successful, that is, the video is legal. Not an attack video.

Step 8: The server sends the verification result to the terminal device.

The living body detection method of the present application has the following multiple effects:

1. It can detect whether the video of the target object is an attack video shot after the camera device was hijacked.

2. The live detection method of the present application has a wider application range. On the one hand, it no longer requires the intensity of light (white balance has nothing to do with the intensity of light), it can still take effect in an environment with strong light, and can cover the user's various On the other hand, there is no longer any requirement for the light transmittance of the protective film of the device (such as mobile phone film) (white balance has nothing to do with the light transmittance of the film), which can cover various devices of the user; on the other hand, It is no longer necessary to brighten the screen of the device (white balance has nothing to do with screen brightness), which can avoid dazzling users and optimize the user experience.

3. The living body detection method of the present application has high detection efficiency, does not require users to spend a long time to interact, and at the same time reduces the difficulty for users to use, avoids individual users from being unable to complete the detection due to too much difficulty, and improves the overall detection rate . It should be noted that, for the sake of simple description, the method embodiments are expressed as a series of action combinations, but those skilled in the art should know that the embodiments of the present application are not limited by the described action sequence, because According to the embodiments of the present application, certain steps may be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present application.

Based on the same inventive concept, an embodiment of the present application provides a living body detection system, the system includes a terminal device and a server;

The terminal device is configured to obtain target white balance sequence information from a server in response to a living body detection request for performing living body detection on the target object, and control the camera device to perform video recording on the target object according to the instructions of the target white balance sequence, and record the target object. The recorded video of the target object is sent to the server; wherein the target white balance parameter sequence information is used to instruct to adjust the white balance parameters of the camera during the video recording process;

The server is configured to receive the video of the target object, parse the video, obtain the recording white balance sequence information actually used in the recording process of the video, and associate the recording white balance sequence information with the target object. The white balance sequence information is matched, and whether the target object is a living body is verified according to the matching result of the recorded white balance sequence information and the target white balance sequence information.

Based on the same inventive concept, an embodiment of the present application provides a living body detection apparatus 500 . Referring to FIG. 5 , FIG. 5 is a structural block diagram of a living body detection apparatus according to an embodiment of the present application. As shown in FIG. 5 , the living body detection device 500 may include:

The first acquisition module 501 is used to acquire the video of the target object;

A parsing module 502, configured to parse the video to obtain the recording white balance sequence information actually used in the recording process of the video;

A matching module 503, configured to match the recording white balance sequence information with target white balance sequence information, where the target white balance sequence information is used to instruct the white balance parameters of the camera to be adjusted during the video recording process ;

The verification module 504 is configured to verify whether the target object is a living body according to the matching result between the recorded white balance sequence information and the target white balance sequence information.

Optionally, the apparatus 500 further includes:

a second obtaining module, configured to obtain white balance parameter information supported by the camera device; wherein the white balance parameter information includes color temperature information corresponding to the white balance parameter;

The generation module is configured to generate at least one white balance sequence to be selected for the camera device according to the color temperature information of the white balance parameter supported by the camera device; wherein, the white balance sequence to be selected is for the camera device to record the data. Adjust the white balance parameters when describing the video of the target object.

Optionally, the apparatus is applied to a server; the apparatus 500 further includes:

The first sending module is configured to send the at least one white balance sequence to be selected to the terminal device, and the camera device of the terminal device performs white balance parameters according to the at least one white balance sequence to be selected when recording the video. adjustment;

The first obtaining module 501 includes:

The receiving sub-module is configured to receive the video of the target object sent by the terminal device.

Optionally, the number of to-be-selected white balance sequences sent by the server to the terminal device is multiple; the apparatus 500 further includes:

A receiving module, configured to receive the white balance sequence to be used when recording the video of the target object sent by the terminal device, and use it as the target white balance sequence.

Optionally, the apparatus 500 further includes:

a first determining module, configured to determine a to-be-selected white balance sequence from the at least one to-be-selected white balance sequence whenever a living body detection request is received;

The second determination module is configured to use the determined to-be-selected white balance sequence as the target white balance sequence used by the camera device to record the video, wherein the two to-be-selected white balance sequences determined in two adjacent times are different.

Optionally, the parsing module 502 includes:

Extracting submodules for extracting multiple video frames from the video;

The first obtaining sub-module is used to obtain the color temperature change information corresponding to the recording white balance sequence according to the color value of the target pixel point on each of the video frames; wherein, the target pixel point is all the pixels of the video frame The point or the target pixel point is the pixel point in the target area of the video frame; the target area is the area used for living body detection;

The matching module 503 includes:

The matching submodule is configured to match the color temperature change information corresponding to the recording white balance sequence with the color temperature change information corresponding to the target white balance sequence.

Optionally, the first obtaining submodule includes:

The second obtaining sub-module is used to obtain the color temperature change information corresponding to the recorded white balance sequence according to the color value of the pixel in the face area of the target object on each of the video frames; or

The third obtaining sub-module is configured to obtain the color temperature change information corresponding to the recording white balance sequence according to the color values of the pixels in the palm area of the target object on each of the video frames.

Based on the same inventive concept, an embodiment of the present application provides another living body detection apparatus 600, which is applied to a terminal device. Referring to FIG. 6 , FIG. 6 is a structural block diagram of another living body detection apparatus shown in an embodiment of the present application. As shown in FIG. 6 , the living body detection device 600 may include:

The third obtaining module 601 is configured to obtain target white balance sequence information from a server in response to a live body detection request for performing live body detection on the target object; wherein, the target white balance parameter sequence information is used to indicate the Adjust the white balance parameters of the camera device;

a first control module 602, configured to control the camera device to perform video recording on the target object according to the instruction of the target white balance sequence information;

The second sending module 603 is configured to send the recorded video of the target object to the server, and the server parses the video to obtain the recording white balance actually used in the video recording process sequence information, and match the recorded white balance sequence information with the target white balance sequence information, and determine whether the target object is a living body according to the matching result.

Optionally, the first control module 602 includes:

An extraction submodule, used for sequentially extracting a white balance parameter from the target white balance sequence every target duration in response to the living body detection request;

The first control submodule is configured to control the camera device to record a single-frame video or a continuous multi-frame video according to an extracted white balance parameter.

Optionally, the apparatus 600 further includes:

a detection module, configured to detect whether the quality parameter of each video frame recorded by the camera device conforms to the target quality parameter;

a second control module, configured to control the camera to stop video recording and discard the recorded video when it is detected that the quality parameter of a video frame recorded by the camera does not meet the target quality parameter; and, from the camera The server obtains the new target white balance sequence information;

The first control submodule includes:

The second control submodule is configured to control the camera to re-record the video of the target object according to the instruction of the new target white balance sequence information.

Optionally, when the target white balance sequence information obtained from the server is multiple, the apparatus 600 further includes:

The third sending module is configured to send the selected target white balance sequence to the server.

The embodiment of the present application further provides an electronic device, as shown in FIG. 7 . FIG. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present application. 7, the electronic device includes a processor 71, a communication interface 72, a memory 73 and a communication bus 74, wherein the processor 71, the communication interface 72, and the memory 73 complete the communication with each other through the communication bus 74;

memory 73 for storing computer programs;

When the processor 71 is used to execute the program stored in the memory 73, the following steps are implemented:

Get the video of the target object;

Analyze the video to obtain the recording white balance sequence information actually used in the recording process of the video;

Matching the recording white balance sequence information with the target white balance sequence information, the target white balance sequence information is used to instruct to adjust the white balance parameters of the camera during the recording process of the video;

According to the matching result between the recorded white balance sequence information and the target white balance sequence information, it is verified whether the target object is a living body.

Or, when the processor 71 executes the program stored in the memory 73, all the steps in the above-mentioned other method embodiments are implemented.

The communication bus mentioned by the above terminal may be a Peripheral Component Interconnect (PCI for short) bus or an Extended Industry Standard Architecture (Extended Industry Standard Architecture, EISA for short) bus or the like. The communication bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of presentation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

The communication interface is used for communication between the above-mentioned terminal and other devices.

The memory may include random access memory (Random Access Memory, RAM for short), and may also include non-volatile memory (non-volatile memory), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

The above-mentioned processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, referred to as CPU), a network processor (Network Processor, referred to as NP), etc.; may also be a digital signal processor (Digital Signal Processing, referred to as DSP) , Application Specific Integrated Circuit (ASIC for short), Field-Programmable Gate Array (FPGA for short) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

In yet another embodiment provided by the present application, a computer-readable storage medium is also provided, where instructions are stored in the computer-readable storage medium, and when the computer-readable storage medium is run on a computer, the computer is made to execute any one of the foregoing embodiments. The described living body detection method.

In yet another embodiment provided by the present application, there is also provided a computer program product containing instructions, which, when executed on a computer, cause the computer to execute the method for detecting a living body in any of the foregoing embodiments.

In the above-mentioned embodiments, it may be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented in software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated. The computer may be a general purpose computer, special purpose computer, computer network, or other programmable device. The computer instructions may be stored in or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions may be downloaded from a website site, computer, server or data center Transmission to another website site, computer, server, or data center is by wire (eg, coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (eg, infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, a data center, or the like that includes an integration of one or more available media. The usable media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes), optical media (eg, DVDs), or semiconductor media (eg, Solid State Disk (SSD)), among others.

It should be noted that, in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any relationship between these entities or operations. any such actual relationship or sequence exists. Moreover, the terms "comprising", "comprising" or any other variation thereof are intended to encompass a non-exclusive inclusion such that a process, method, article or device that includes a list of elements includes not only those elements, but also includes not explicitly listed or other elements inherent to such a process, method, article or apparatus. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in a process, method, article or apparatus that includes the element.

Each embodiment in this specification is described in a related manner, and the same and similar parts between the various embodiments may be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and for related parts, please refer to the partial descriptions of the method embodiments.

The above descriptions are only preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this application are included in the protection scope of this application.

Claims (15)

1. A method of in vivo detection, the method comprising:

acquiring a video of a target object;

analyzing the video to obtain the actually used recording white balance sequence information in the recording process of the video;

matching the recorded white balance sequence information with target white balance sequence information, wherein the target white balance sequence information is used for indicating that the white balance parameters of a camera device are adjusted in the video recording process;

and checking whether the target object is a living body according to the matching result of the recorded white balance sequence information and the target white balance sequence information.

2. The method of claim 1, further comprising:

acquiring white balance parameter information supported by a camera device; the white balance parameter information comprises color temperature information corresponding to the white balance parameters;

generating at least one white balance sequence to be selected for the camera device according to the color temperature information of the white balance parameter supported by the camera device; and adjusting white balance parameters when the to-be-selected white balance sequence is used for the camera device to record the video of the target object.

3. The method of claim 2, wherein the method is applied to a server; prior to the obtaining the video of the target object, the method further comprises:

sending the at least one white balance sequence to be selected to terminal equipment, and adjusting white balance parameters according to the at least one white balance sequence to be selected when a camera of the terminal equipment records the video;

acquiring a video of a target object, comprising:

and receiving the video of the target object sent by the terminal equipment.

4. The method according to claim 3, characterized in that the number of the white balance sequences to be selected sent by the server to the terminal device is multiple; before the matching the recorded white balance sequence information with the target white balance sequence information, the method further includes:

and receiving a white balance sequence to be selected, which is sent by the terminal equipment and used when the terminal equipment records the video of the target object, and taking the white balance sequence as the target white balance sequence.

5. The method of any of claims 2-4, wherein prior to said obtaining the video of the target object, the method further comprises:

determining a white balance sequence to be selected from the at least one white balance sequence to be selected whenever a liveness detection request is received;

and taking the determined white balance sequence to be selected as a target white balance sequence used by the camera device for recording the video, wherein two determined white balance sequences to be selected in two adjacent times are different.

6. The method according to any one of claims 1 to 5, wherein parsing the video to obtain recording white balance sequence information actually used in the recording process of the video comprises:

extracting a plurality of video frames from the video;

obtaining color temperature change information corresponding to the recorded white balance sequence according to the color value of the target pixel point on each video frame; the target pixel points are all pixel points of the video frame or the target pixel points are pixel points in a target area of the video frame; the target area is an area for performing living body detection;

matching the recorded white balance sequence information with target white balance sequence information, comprising:

and matching the color temperature change information corresponding to the recorded white balance sequence with the color temperature change information corresponding to the target white balance sequence.

7. The method of claim 6, wherein obtaining color temperature change information corresponding to the recorded white balance sequence according to the color value of the target pixel point on each video frame comprises:

obtaining color temperature change information corresponding to the recorded white balance sequence according to color values of pixel points in the face area of the target object on each video frame; or

And obtaining color temperature change information corresponding to the recorded white balance sequence according to the color values of the pixel points in the palm region of the target object on each video frame.

8. A method for detecting a living body is applied to a terminal device, and comprises the following steps:

acquiring target white balance sequence information from a server in response to a liveness detection request for liveness detection of a target object; the target white balance parameter sequence information is used for indicating that the white balance parameters of the camera device are adjusted in the video recording process;

controlling the camera device to record the video of the target object according to the indication of the target white balance sequence information;

and sending the recorded video of the target object to the server, analyzing the video by the server to obtain recorded white balance sequence information actually used in the recording process of the video, matching the recorded white balance sequence information with the target white balance sequence information, and determining whether the target object is a living body according to a matching result.

9. The method according to claim 8, wherein the controlling the camera to record video of the target object according to the indication of the target white balance sequence information comprises:

in response to the living body detection request, sequentially extracting a white balance parameter from a target white balance sequence every target duration;

and controlling the camera device to record a single-frame video or continuous multi-frame video according to the extracted white balance parameter.

10. The method according to claim 8 or 9, wherein during controlling the image pickup apparatus to perform video recording of the target object in accordance with the instruction of the target white balance sequence information, the method further comprises:

detecting whether the quality parameter of each video frame recorded by the camera device meets a target quality parameter or not;

when detecting that the quality parameter of one video frame recorded by the camera device does not accord with the target quality parameter, controlling the camera device to stop video recording and discarding the recorded video; and obtaining new target white balance sequence information from the server;

controlling the camera device to record a single-frame video or a continuous multi-frame video according to the extracted white balance parameter, comprising:

and controlling the camera device to re-record the video of the target object according to the indication of the new target white balance sequence information.

11. The method according to any one of claims 8 to 10, wherein when there are a plurality of pieces of target white balance sequence information acquired from the server, the method further comprises:

and sending the selected target white balance sequence to the server.

12. A living body detection system is characterized by comprising terminal equipment and a server;

the terminal equipment is used for responding to a living body detection request for carrying out living body detection on a target object, acquiring target white balance sequence information from a server, controlling a camera device to record a video of the target object according to the indication of the target white balance sequence information, and sending the recorded video of the target object to the server; the target white balance parameter sequence information is used for indicating that the white balance parameters of the camera device are adjusted in the video recording process;

the server is used for receiving the video of the target object, analyzing the video to obtain recording white balance sequence information actually used in the recording process of the video, matching the recording white balance sequence information with the target white balance sequence information, and checking whether the target object is a living body according to the matching result of the recording white balance sequence information and the target white balance sequence information.

13. An electronic device comprising a memory, a processor and a computer program stored on the memory, wherein the processor executes the computer program to implement the steps in the liveness detection method of any one of claims 1-7 or wherein the processor executes the computer program to implement the steps in the liveness detection method of any one of claims 8-11.

14. A computer readable storage medium having stored thereon a computer program/instructions, characterized in that the computer program/instructions, when executed by a processor, implement the steps in the liveness detection method of any one of claims 1-7, or the computer program/instructions, when executed by a processor, implement the steps in the liveness detection method of any one of claims 8-11.

15. A computer program product comprising computer programs/instructions, characterized in that the computer programs/instructions, when executed by a processor, implement the steps in the liveness detection method of any one of claims 1 to 7, or the computer programs/instructions, when executed by a processor, implement the steps in the liveness detection method of any one of claims 8 to 11.

Images