CN104992085A - A human body detection method and device based on touch trajectory tracking - Google Patents

Abstract

The invention discloses a human body living body detection method based on touch trajectory tracking, which comprises the following steps: randomly calling an instruction subject preset in an instruction database in advance, and designating a response in a touch track tracking mode; acquiring a touch track through a touch screen; matching and comparing the acquired touch track with a standard track preset in the instruction database, and if the matching is successful, judging that the response user is a living human body; otherwise, the non-human living body is judged. According to the invention, the command questions which need to be responded by a human living body brain thinking party in the command database are randomly called, and the response is completed in a touch track tracking mode, so that a counterfeiter can be effectively prevented from attacking the human living body detection system by using a series of technical means such as photos, cut videos, synthesized voice, automatic OCR (optical character recognition) of a machine and the like, and the safety and the reliability of the human living body detection system are higher.

Description

A human body detection method and device based on touch trajectory tracking

technical field

The invention relates to the technical fields of computer application and image recognition, in particular to a human living body detection method and device based on touch trajectory tracking.

Background technique

The demand for commercial application of identity verification is becoming more and more intense, and people have taken biometric technology as an important means of identity verification. Human biological characteristics include physiological characteristics and behavioral characteristics. Physiological characteristics usually include face, fingerprint, iris, retina, palmprint, hand shape, voice, DNA, auricle shape, etc.; behavioral characteristics usually include handwriting, gait, etc. However, the problem of counterfeiting poses a great threat to authentication systems based on human biometrics. Counterfeiters can attack existing biometric-based identity verification systems by using fake fingerprint films, stealing other people's photos or videos, and synthesized voices. If the identity verification system based on human biometrics lacks the ability to effectively prevent counterfeit attacks, it will bring disastrous consequences. One of the means to avoid the aforementioned counterfeiting problem is to ensure that the acquisition of human biological characteristics is obtained from a living human body. Therefore, an effective human living body detection technology is required.

For identity verification systems based on face recognition, counterfeiters can use photos of legitimate users to carry out fraudulent attacks. Therefore, the current existing living body detection methods mainly adopt the idea of automatically distinguishing between photo faces and live faces. The main difference between a photo face and a living human face is that the human face in the photo does not perform actions such as blinking or opening the mouth, while the living human face can perform actions such as opening the mouth, blinking eyes, and shaking the head. However, criminals can further conduct fraudulent attacks on the liveness detection system by recording the face video images of legitimate users in advance or editing and synthesizing the face video images of legitimate users.

For identity verification systems based on speech recognition, counterfeiters can conduct fraudulent attacks on human body detection systems through pre-recorded voices, edited or synthesized voices.

Contents of the invention

The first purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art, and provide a human body detection method based on touch trajectory tracking. The user thinks about the instruction questions that can be answered, and then the terminal randomly calls the instruction questions in the instruction database, and requires the user to complete the answer by touching the terminal touch screen with a finger or a touch tool (such as a touch pen) within the specified time , if the user completes the response correctly within the specified time, it is judged as a living human body, otherwise, it is a non-human living body.

Another object of the present invention is to overcome the shortcomings and deficiencies of the prior art, and provide a human body living body detection device based on touch trajectory tracking.

First purpose of the present invention is achieved through the following technical solutions:

A human body detection method based on touch trajectory tracking, applied to a terminal configured with a touch screen, comprising the following steps:

Randomly call the pre-set command questions in the command database that need to be answered by living brain thinking;

Designate the answering method of the instruction topic as touch track tracking;

Obtaining the user's touch trajectory through the collection of the terminal touch screen;

Matching and comparing the user's touch trajectory with the preset standard trajectory in the command database, if the matching is successful, it is judged that the responding user is a living body; otherwise, it is judged that the responding user is a non-human living body.

Preferably, the instruction topics include:

A maze is displayed on the touch screen, and the user is required to draw a path diagram for getting out of the maze;

Displaying half of a symmetrical figure on the touch screen, requiring the user to draw the other half of the symmetrical figure;

Displaying a sequence of patterns on the touch screen, stipulating that the user draws the pattern that should be displayed next according to the changing trend or rule of the patterns;

The patterns are sequentially displayed on the touch screen, and the user is required to sequentially touch and click the display patterns according to the display sequence;

Displaying a moving track of a target on the touch screen, requiring the user to draw the moving track of the target;

The stroke writing order of one or more characters is displayed on the touch screen, and the user is required to write corresponding characters in this order.

Preferably, the user touch track collected by the terminal touch screen is generated by the user touching the terminal touch screen with a finger or a touch tool.

Preferably, the method also includes:

The step of setting the response time, which is used to set the response time of the command topic, if the response time of the command topic exceeds the preset response time, it is judged that the responding user is a non-human living body.

Preferably, before the step of randomly calling the pre-set instruction subject in the instruction database, the method further includes:

An instruction database entry step, which is used to enter the instruction topic into the instruction database, and enter the standard trajectory corresponding to the instruction topic.

Preferably, in the step of matching and comparing the user touch trajectory with the preset standard trajectory in the instruction database, the similarity judgment is performed by setting a matching similarity threshold, if the matching similarity is greater than or equal to the If the similarity threshold is matched, it is judged that the matching is successful; otherwise, it is judged that the matching fails.

Another object of the present invention is achieved through the following technical solutions:

A human body detection device based on touch trajectory tracking, applied to a terminal configured with a touch screen, the device includes:

An instruction topic calling unit, which is used to randomly call pre-set instruction topics in the instruction database;

An answer mode specifying unit, which is used to specify the answer mode of the instruction topic as touch track tracking;

A touch track acquisition unit, which acquires the user's touch track through the touch screen of the terminal;

A matching judging unit, which is used to match and compare the user's touch trajectory with the preset standard trajectory in the instruction database, if the matching is successful, then judge that the responding user is a living body; otherwise, judge that the responding user is not a human body living body.

Preferably, the device also includes:

A response time setting unit, which is used to set the response time of the command topic, if the response time of the command topic exceeds the preset response time, it is judged that the responding user is a non-human living body.

Preferably, the device also includes:

An instruction database entry unit, which is used to enter instruction topics into the instruction database, and enter standard tracks corresponding to the instruction topics.

Compared with the prior art, the present invention has the following advantages and effects:

(1) The present invention uses a finger or a touch tool (such as a touch pen) to perform liveness detection, so as to prevent counterfeiters from cheating liveness detection through photos or videos when performing liveness detection based on face detection system. Secondly, it also prevents counterfeiters from deceiving the living body detection system by synthesizing voice when liveness detection is performed based on speech recognition.

(2) The instruction questions called from the instruction database in the present invention are randomly generated, and the traces to be drawn in the randomly called instruction questions still need to be drawn after living brain thinking. Therefore, it is possible to effectively prevent counterfeiters from invading the human body detection system through machines.

Description of drawings

Fig. 1 is a flow chart of a human body detection method based on touch trajectory tracking in Embodiment 1 of the present invention;

FIG. 2 is a block diagram of a human body detection device based on touch trajectory tracking in Embodiment 2 of the present invention.

Detailed ways

In order to make the object, technical solution and advantages of the present invention more clear and definite, the present invention will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

Embodiment one

Please refer to FIG. 1. FIG. 1 is a flow chart of a human body detection method based on touch trajectory tracking in Embodiment 1, which is applied to a terminal equipped with a touch screen, wherein the terminal includes a mobile phone, a notebook computer, a tablet computer, a palm Internet device, Multimedia devices, streaming media devices, servers or other smart devices connected to the cloud, etc., terminals include but are not limited to the above examples.

The application example of the human living body detection method based on touch trajectory tracking will be described in detail below in conjunction with FIG. 1 . As can be seen in Figure 1, the human body detection method based on touch track tracking includes the following steps:

Step S1, randomly calling pre-set instruction titles in the instruction database.

The instruction database of the terminal saves various instruction topics in advance, and the terminal calls the instruction topics randomly, requiring the user to complete the corresponding answer according to the answer method prompted on the touch screen.

Since the biggest difference between a living person and a machine is that the living person can think in the brain, the command questions invoked by the terminal in this embodiment need to be answered by the brain of the living person. This directive includes but is not limited to:

(1) A maze is displayed on the touch screen, and the user is required to touch with a finger or a touch tool (such as a touch pen) to draw a path diagram for getting out of the maze.

(2) Displaying half of a symmetrical figure on the touch screen, it is stipulated that the user draws the other half of the symmetrical figure by touching with a finger or a touch tool (such as a touch pen).

(3) Displaying a sequence of patterns on the touch screen, stipulating that the user draws the next pattern to be displayed according to the changing trend or rule of the pattern with a finger or a touch tool (such as a touch pen). For example, the change trend of the displayed pattern sequence is a gradually larger or smaller polygon (such as a triangle, a quadrilateral), and the user correspondingly draws a larger or smaller polygon (such as a triangle, a quadrilateral). For another example, if the change trend of the displayed pattern sequence is a gradually longer or shorter line segment or arc, the user draws a longer or shorter line segment or arc accordingly. For example, the rule of the displayed pattern sequence is a pattern of a certain color, and the user should select a brush of the corresponding color to draw the pattern of the corresponding color. Another example is that the displayed pattern sequence is that an object is rotated by 45°, 90°, and 135° respectively, and the user correspondingly needs to draw the figure after the object is rotated by 180°.

(4) The patterns are displayed sequentially on the touch screen, and it is stipulated that the user touches and clicks the displayed patterns sequentially with a finger or a touch tool (such as a touch pen) according to the display sequence.

(5) Displaying a moving track of an object on the touch screen, and specifying that the user touches with a finger or a touch tool (such as a touch pen) to draw the moving track of the target.

(6) Displaying the stroke writing sequence of one or more characters on the touch screen, and specifying that the user writes corresponding characters according to the stroke writing sequence with fingers or a touch tool (such as a touch pen).

If the method is used for the first time, the method also includes an instruction database entry step, which is used to enter the instruction topic into the instruction database, and enter the standard trajectory corresponding to the instruction topic. Of course, during the use process stage, the instruction title in the instruction database can also be added or updated.

Step S2, designate the answering method of the instruction question as touch track tracking.

Step S3, acquiring the user's touch track through the touch screen of the terminal.

In this step, the user's touch track collected by the terminal touch screen is generated by the user touching (that is, writing, drawing, or swiping) on the terminal touch screen with a finger or a touch tool (such as a touch pen).

Step S4. Matching and comparing the user's touch trajectory with the preset standard trajectory in the command database, if the matching is successful, it is judged that the responding user is a living body; otherwise, it is judged that the responding user is a non-human living body.

In addition, in the actual use process, before judging whether the answering user is engaged in physical activity, the answering time of the command topic can also be set, and it can be judged whether the user's answer exceeds the answering time. That is, the step of setting the response time, which is used to set the response time of the command topic, if the response time of the command topic exceeds the preset response time, it is judged that the responding user is a non-human living body.

Below in conjunction with Fig. 1, the human living body detection method based on touch trajectory tracking in the present embodiment is further described in detail:

The terminal randomly selects the instruction topic from the instruction database, and sends out an instruction prompt, asking the user to answer the instruction topic with a finger or a touch pen within the specified time; the user uses the finger or the touch tool ( For example, a touch pen) touch writing to complete the touch operation required by the terminal; the touch track is collected through the terminal touch screen; the obtained touch track is matched with the standard track preset in the command database, and if the match is successful, then judge the answering user is a living body; otherwise, it is judged that the responding user is not a living body. When actually judging whether there is a match, a matching degree or similarity threshold can be predefined in advance. For example, if the matching degree or similarity reaches more than 90%, the matching is considered successful, and the responding user is judged to be a living body.

(1) Implementation of maze command

A maze is displayed on the touch screen, and the user is required to touch with a finger or a touch tool (such as a touch pen) to draw a path map for getting out of the maze. For the maze command topic, display the maze on the touch screen, let the living body touch with fingers or touch tools (such as touch pen), and draw the path out of the maze. The user's touch trajectory is obtained through the touch screen, and the user's touch trajectory is matched and compared with the preset standard trajectory in the command database. If the match is successful, it is judged that the responding user is a living body; otherwise, it is judged that the responding user is a non-human living body.

(2) Draw a symmetrical semicircle according to the semicircle implementation

That is, displaying half of a symmetrical figure on the touch screen, it is stipulated that the user draws the other half of the symmetrical figure by touching with a finger or a touch tool (such as a touch pen).

The terminal prompts to give a circle drawing command. At this time, the left semicircle or right semicircle is displayed on the terminal touch screen, and the user is required to touch and draw a symmetrical right semicircle or left semicircle with a finger or a touch tool (such as a touch pen). After the user draws according to the instruction, the touch trajectory can be collected through the touch screen, and then the obtained touch trajectory is matched with the preset standard trajectory in the command database. If the matching is successful, it is determined to be a human body, otherwise it is not a living body.

(3) Click on the mouse pattern appearing on the touch screen in sequence (similar to whack-a-mole game) implementation

That is, the patterns are sequentially displayed on the touch screen, and it is stipulated that a user touches and clicks the displayed patterns sequentially with a finger or a touch tool (such as a touch pen) according to the display sequence.

Mouse patterns appear sequentially on the touch screen of the terminal, prompting the user to click on the sequentially appearing mouse patterns. After the user performs a touch and click operation, the touch track is collected through the touch screen. According to the collected touch track, it is judged whether the area clicked by the user is correct. If the user completes the clicks correctly in sequence according to the prompts within the specified time, it is determined to be a living body, otherwise it is not a living body.

(4) Track the motion path of the animal for specific implementation

The movement track of an animal is displayed on the touch screen of the terminal, and the living body is allowed to touch with a finger or a touch tool (such as a touch pen) to draw the movement track of the target. According to the motion trajectory displayed on the touch screen, the user touches and tracks with a finger or a touch tool (such as a touch pen). If the matching is successful, it is determined to be a living body, otherwise it is not a living body.

In the present invention, the terminal randomly calls the instruction questions in the instruction database that need to be answered by the living brain, and specifies that the user completes the answer by tracking the touch track of the finger or touch tool (such as a touch pen), so as to judge the current user. Whether it is a living body or not, the purpose is to prevent criminals from attacking the human living body detection system through a series of technical means such as video synthesis, speech synthesis, and machine automatic OCR recognition.

Embodiment two

As shown in FIG. 2 , FIG. 2 is a block diagram of a human body detection device based on touch trajectory tracking in Embodiment 2, which is applied to a terminal configured with a touch screen.

As can be seen from Figure 2, the device includes:

1. An instruction topic calling unit, which is used to randomly call pre-set instruction topics in the instruction database.

Since the biggest difference between a living person and a machine is that the living person can think with the brain, the command topic called by the terminal in this embodiment needs to be thought through the brain. This directive includes but is not limited to:

(1) A maze is displayed on the touch screen, and the user is required to touch with a finger or a touch tool (such as a touch pen) to draw a path diagram for getting out of the maze.

(2) Displaying half of a symmetrical figure on the touch screen, it is stipulated that the user draws the other half of the symmetrical figure by touching with a finger or a touch tool (such as a touch pen).

(3) Displaying a pattern sequence on the touch screen, it is stipulated that the user touches with a finger or a touch tool (such as a touch pen), and draws the next pattern to be displayed according to the changing trend and rule of the pattern.

(4) The patterns are displayed sequentially on the touch screen, and it is stipulated that the user touches and clicks the displayed patterns sequentially with a finger or a touch tool (such as a touch pen) according to the display sequence.

(5) Displaying a moving track of an object on the touch screen, and specifying that the user touches with a finger or a touch tool (such as a touch pen) to draw the moving track of the target.

(6) Display the stroke writing order of one or more characters on the touch screen, and stipulate that the user writes corresponding characters in this order by touching with a finger or a touch tool (such as a touch pen).

2. An answer mode specifying unit, which is used to specify the answer mode of the instruction topic as touch track tracking.

3. A touch track acquiring unit, which acquires the touch track through the touch screen.

4. A matching judging unit, which is used to match and compare the user’s touch trajectory with the preset standard trajectory in the instruction database. If the matching is successful, it is judged that the responding user is a living body; otherwise, it is judged that the responding user is non-human body.

5. A response time setting unit, which is used to set the response time of the command topic. If the response time of the command topic exceeds the preset response time, it is judged that the responding user is a non-human living body.

7. An instruction database entry unit, which is used to enter the instruction topic into the instruction database, and enter the standard trajectory corresponding to the instruction topic.

If the device is used for the first time, the instruction database entry unit must be used, and the instruction topic is entered into the instruction database by calling the unit, and the standard trajectory corresponding to the instruction topic is entered. Of course, during the use process stage, the instruction title in the instruction database can also be added or updated by calling this unit.

In a word, the device of the present invention can effectively prevent counterfeiters from deceiving the living body detection and authentication system by using methods such as photos, clipped videos, and synthesized voices, so that the safety and reliability of the living body detection and authentication system are higher.

It is worth noting that in the above-mentioned device embodiments, the included units and modules are only divided according to functional logic, but are not limited to the above-mentioned divisions, as long as the corresponding functions can be realized; in addition, each device and unit The specific names are only for the convenience of distinguishing each other, and are not used to limit the protection scope of the present invention.

The above-mentioned embodiment is a preferred embodiment of the present invention, but the embodiment of the present invention is not limited by the above-mentioned embodiment, and any other changes, modifications, substitutions, combinations, Simplifications should be equivalent replacement methods, and all are included in the protection scope of the present invention.

Claims (9)

1. A human body living body detection method based on touch trajectory tracking, applied to a terminal configured with a touch screen, characterized in that, comprising the following steps: Randomly call the pre-set command questions in the command database that need to be answered by living brain thinking; Designate the answering method of the instruction topic as touch track tracking; Obtaining the user's touch trajectory through the collection of the terminal touch screen; Matching and comparing the user's touch trajectory with the preset standard trajectory in the command database, if the matching is successful, it is judged that the responding user is a living body; otherwise, it is judged that the responding user is a non-human living body. 2. The human living body detection method based on touch trajectory tracking according to claim 1, wherein the instruction topic comprises: A maze is displayed on the touch screen, and the user is required to draw a path diagram for getting out of the maze; Displaying half of a symmetrical figure on the touch screen, requiring the user to draw the other half of the symmetrical figure; Displaying a sequence of patterns on the touch screen, stipulating that the user draws the pattern that should be displayed next according to the changing trend or rule of the pattern; The patterns are sequentially displayed on the touch screen, and the user is required to sequentially touch and click the display patterns according to the display order; Displaying a moving track of a target on the touch screen, requiring the user to draw the moving track of the target; The stroke writing sequence of one or more characters is displayed on the touch screen, and the user is required to write corresponding characters according to the stroke writing sequence. 3. according to claim 1 or 2 described based on the human body living body detection method of touch trajectory tracking, it is characterized in that, The user touch track collected by the terminal touch screen is generated by the user touching the terminal touch screen with a finger or a touch tool. 4. according to claim 1 or 2 described human body detection methods based on touch trajectory tracking, it is characterized in that, described method also comprises: The step of setting the response time, which is used to set the response time of the command topic, if the response time of the command topic exceeds the preset response time, it is judged that the responding user is a non-human living body. 5. according to claim 1 or 2 described human body living body detection method based on touch track tracking, it is characterized in that, Before the step of randomly calling the pre-set instruction subject in the instruction database, the method also includes: An instruction database entry step, which is used to enter the instruction topic into the instruction database, and enter the standard trajectory corresponding to the instruction topic. 6. according to claim 1 or 2 described human body living body detection methods based on touch trajectory tracking, it is characterized in that, In the step of matching and comparing the user touch trajectory with the preset standard trajectory in the instruction database, the similarity judgment is performed by setting the matching similarity threshold, if the matching similarity is greater than or equal to the matching similarity threshold, it is judged that the matching is successful; otherwise, it is judged that the matching fails. 7. A human body detection device based on touch trajectory tracking, applied to a terminal configured with a touch screen, characterized in that the device includes: An instruction topic calling unit, which is used to randomly call pre-set instruction topics in the instruction database; An answer mode specifying unit, which is used to specify the answer mode of the instruction topic as touch track tracking; A touch track acquisition unit, which acquires the user's touch track through the touch screen of the terminal; A matching judging unit, which is used to match and compare the user's touch trajectory with the preset standard trajectory in the instruction database, if the matching is successful, then judge that the responding user is a living body; otherwise, judge that the responding user is not a human body living body. 8. The human living body detection device based on touch trajectory tracking according to claim 7, which is applied to a terminal configured with a touch screen, wherein the device further comprises: A response time setting unit, which is used to set the response time of the command topic, if the response time of the command topic exceeds the preset response time, it is judged that the responding user is a non-human living body. 9. The human living body detection device based on touch trajectory tracking according to claim 7, which is applied to a terminal configured with a touch screen, wherein the device further comprises: An instruction database entry unit, which is used to enter the instruction topic into the instruction database, and enter the standard trajectory corresponding to the instruction topic.