WO2021097829A1

WO2021097829A1 - Method and apparatus for locking target object, computer device and storage medium

Info

Publication number: WO2021097829A1
Application number: PCT/CN2019/120364
Authority: WO
Inventors: 吴晓; 胡龙耀; 耿路; 吴靖; 范育宸; 刘怀国; 刘润声
Original assignee: 连云港伍江数码科技有限公司
Priority date: 2019-11-22
Filing date: 2019-11-22
Publication date: 2021-05-27

Abstract

A method and apparatus for locking a target object, a computer device and a storage medium, which relate to the technical field of video recognition. The method comprises: when a plurality of objects enter a specific region, acquiring authentication feature information of the plurality of objects (S201); authenticating the plurality of objects according to the authentication feature information, and using the authenticated objects as objects to be locked (S202); receiving a lock trigger instruction, and prompting the objects to perform a specific action according to the lock trigger instruction (S203); and acquiring lock feature information when the objects perform the specific action, and determining an object the lock feature information of which meets a preset condition as a target object (S204). The described method solves the problem in which target object determination is inaccurate. The target object may be accurately locked insofar as the plurality of objects are in one environment.

Description

Target object locking method, device, computer equipment and storage medium

Technical field

The present invention relates to the technical field of video recognition, in particular to a method, device, computer equipment and storage medium for locking a target object.

Background technique

With the development of society, unmanned supermarkets have gradually emerged. At present, unmanned supermarkets mainly use RFID electronic identification technology to identify products. In the process of implementing the present invention, the inventor found that the prior art has at least the following problems: when multiple consumers are in the same RFID unmanned vending device or the same unmanned supermarket environment, the traditional identification method may not be accurate To determine which of these consumers is the dominant or settler of the current shopping, it is easy to confuse the correspondence between products and consumers.

Summary of the invention

Based on this, the present invention provides a method, a device, a computer device and a storage medium for locking a target object, which can accurately lock a shopping leader when multiple objects are in the same environment.

The content of the embodiment of the present invention is as follows:

A method for locking a target object includes the following steps: when multiple objects enter a specific area, acquiring authentication feature information of the multiple objects; authenticating the multiple objects according to the authentication feature information, and passing the authentication The object to be locked as the object to be locked; receives the lock trigger instruction, prompts the object to be locked to perform a specific action according to the lock trigger instruction; obtains the lock feature information when the object to be locked performs the specific action, and matches the lock feature information to The object to be locked with the preset condition is determined as the target object.

In one of the embodiments, the prompting the object to be locked to perform a specific action according to the lock trigger instruction; the step of acquiring the locking feature information when the object to be locked performs the specific action includes: according to the The lock trigger instruction prompts the object to be locked to perform a specific head movement; obtains the movement characteristic information and the first biological characteristic information of the to-be-locked object in the head movement process, according to the movement characteristic information and the first biological characteristic Information to obtain the locking feature information of the object to be locked.

In one of the embodiments, the step of prompting the object to be locked to perform a specific action according to the lock trigger instruction; the step of acquiring the lock feature information when the object to be locked performs the specific action further includes: The lock trigger instruction acquires preset first encoding information, displays the encoding pattern corresponding to the first encoding information, and prompts the object to be locked to scan the encoding pattern; when acquiring the object to be locked scans the encoding pattern According to the code scan information, the lock feature information of the object to be locked is obtained according to the code scan information.

In one of the embodiments, the step of determining the object to be locked whose locking feature information meets a preset condition as the target object includes: when the locking feature information matches the pre-stored candidate locking feature information, determining the target object. The lock feature information meets preset conditions.

In one of the embodiments, the step of authenticating the multiple objects according to the authentication characteristic information includes: comparing the authentication characteristic information of any one of the objects with pre-stored candidate authentication characteristic information, It is judged whether there is candidate authentication characteristic information that matches the authentication characteristic information; if it exists, it is determined that the corresponding object is authenticated.

In one of the embodiments, the step of comparing the authentication feature information of any object with pre-stored candidate authentication feature information includes: comparing the second biometric information of any object with the pre-stored candidate authentication feature information. Comparison of biometric information.

In one of the embodiments, the step of comparing the authentication feature information of any object with pre-stored candidate authentication feature information includes: comparing the second code information of any object with the pre-stored second code information. The coded information is compared.

In one of the embodiments, the specific area includes an unmanned supermarket; after the step of determining the target object whose lock feature information meets preset conditions as the target object, the method further includes: acquiring the target object in the unmanned supermarket. Based on the real-time information of the human supermarket, the shopping information of the target object is determined according to the real-time information.

Correspondingly, an embodiment of the present invention provides a device for locking a target object, including: an information acquisition module for acquiring authentication feature information of multiple objects when they enter a specific area; an authentication module for obtaining authentication feature information of multiple objects according to The authentication feature information authenticates the multiple objects, and uses the authenticated object as the object to be locked; the lock trigger module is configured to receive a lock trigger instruction, and prompt the object to be locked to perform a specific action according to the lock trigger instruction And, the target locking module is used to obtain the locking feature information when the object to be locked performs the specific action, and to determine the object to be locked whose locking feature information meets a preset condition as the target object.

In the above method and device for locking a target object, when multiple objects enter a specific area, the authenticated object is used as the object to be locked, and when the locking feature information of the object to be locked performing a specific action meets the preset conditions, the object will be locked The object is locked as the target object. In the case of determining the object to be locked, the locking feature information when the object to be locked performs a specific action is acquired. The locking feature information can effectively distinguish different objects, so that the target object can be accurately determined from multiple objects.

A computer device includes a memory, a processor, and a computer program stored on the memory and running on the processor. The processor implements the following steps when executing the computer program: when multiple objects enter a specific area, obtain The authentication feature information of the multiple objects; the multiple objects are authenticated according to the authentication feature information, and the authenticated object is regarded as the object to be locked; the lock trigger instruction is received, and the lock trigger instruction is prompted to prompt the waiting The locked object performs a specific action; the locking feature information when the to-be-locked object performs the specific action is acquired, and the to-be-locked object whose locking feature information meets a preset condition is determined as the target object.

The above-mentioned computer equipment, in the case of determining the object to be locked, acquires the locking characteristic information when the object to be locked performs a specific action. The locking characteristic information can effectively distinguish different objects, and therefore can accurately determine the target object from multiple objects .

A computer-readable storage medium has a computer program stored thereon, and when the computer program is executed by a processor, the following steps are implemented: when multiple objects enter a specific area, obtain authentication feature information of the multiple objects; The authentication feature information authenticates the plurality of objects, and uses the authenticated object as the object to be locked; receives a lock trigger instruction, and prompts the object to be locked to perform a specific action according to the lock trigger instruction; obtains the object to be locked The lock feature information when the specific action is executed, and the object to be locked whose lock feature information meets the preset condition is determined as the target object.

The above-mentioned computer-readable storage medium acquires the locking feature information when the object to be locked performs a specific action when the object to be locked is determined. The locking feature information can effectively distinguish different objects, so it can be accurately determined from multiple objects Out of the target audience.

Description of the drawings

FIG. 1 is an application environment diagram of a method for locking a target object in an embodiment;

FIG. 2 is a schematic flowchart of a method for locking a target object in an embodiment;

Figure 3 is a schematic diagram of an object entering a specific area in an embodiment;

4 is a schematic flowchart of a method for locking a target object in another embodiment;

Fig. 5 is a structural block diagram of a locking device of a target object in an embodiment;

FIG. 6 is a structural block diagram of a video detection system in an embodiment;

Fig. 7 shows the internal structure of a computer device in an embodiment.

Detailed ways

In order to make the objectives, technical solutions, and advantages of the present invention clearer, the following further describes the present invention in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, but not to limit the present invention.

The method for locking the target object provided in this application can be applied to the application environment as shown in FIG. 1. The application environment can include a video detection system 101, an information input device 102, and an information output device 103, and the entire application environment can be integrated into a video detection system. The information input device 102 obtains the user's biological characteristics, videos, images and other information in a specific area. The video detection system 101 locks the target object from multiple objects according to the information input by the processing information input device 102, and tracks and locks the target object. At the same time, the video detection system 101 controls the information output device 103 to output a corresponding signal when there is a signal output. Among them, the video detection system 101, the information input device 102, and the information output device 103 may all be terminals or servers. In the embodiment of the present invention, the terminal may be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices. The server may be implemented by an independent server or a server cluster composed of multiple servers.

In one embodiment, the information input device 102 obtains the human body characteristic information and the like in the covered specific area, and transmits the human body characteristic information and the like to the video detection system 101. The video detection system processes the human body characteristic information acquired by the information input device 102 and stores it in the memory, and at the same time authenticates and identifies the object, and locks the target object according to the actions of the object. The information output device 103 is connected to the video detection system 101, and feeds back the image and other output of the video detection system in real time.

The embodiment of the present invention provides a method, a device, a computer device, and a storage medium for locking a target object. Detailed descriptions are given below.

In one embodiment, as shown in FIG. 2, a method for locking a target object is provided. Taking the method applied to the video detection system in FIG. 1 as an example, the method includes the following steps:

S201: When multiple objects enter a specific area, obtain authentication feature information of the multiple objects.

In this step, when multiple objects enter a specific area, the authentication feature information of these objects is obtained through video recognition technology.

Among them, the object can refer to a natural person, a robot, etc., or an animal, etc. At the same time, these objects can refer to consumers who enter a specific area. As shown in FIG. 3, there is more than one object 302 in the dotted area (specific area 301), and the information input device 102 obtains authentication feature information of the object 302.

Specific areas refer to areas such as unmanned vending equipment, unmanned supermarkets, and unmanned freezers. When multiple objects enter these specific areas, because there are no business managers, etc., it is easy to be unable to lock the specific settlement leader, resulting in confusion in the settlement of goods. Therefore, it is necessary to determine the leader of this shopping among multiple objects (the shopping behavior between objects entering a specific area and leaving a specific area can be considered as a shopping). The embodiment of the present invention does not limit the shape and size of a specific area.

Authentication feature information refers to feature information used to authenticate an object, and these authentication feature information can be unique information such as biological feature information, physical feature information, and two-dimensional code information of the object.

In addition, the authentication feature information of the object can be obtained through video recognition technology. Video recognition technology is a relatively mature and stable recognition technology so far, which can realize the recognition of human body, biological body, etc. Video recognition technology uses certain optical modules to obtain environmental images, video information, etc., and uses image analysis and other technologies to extract the required information in environmental images and video information and make corresponding judgments to make up for the incomprehensibility of conventional recognition technology . The current video recognition technology includes face recognition, video recognition, image recognition, text recognition, behavior recognition, iris recognition, eyeball focus positioning and other recognition technologies. The first information of the object can be acquired through a camera device or the like. The camera device can also be a camera or the like installed in a specific area, or a mobile terminal of the object, or the like.

S202: Authenticate the multiple objects according to the authentication feature information, and use the authenticated objects as the objects to be locked.

Among them, the authentication of the object can be to determine whether the corresponding authentication feature information is legal, and if it is legal, it can be determined that the corresponding object is authenticated; it can also be judging whether the corresponding authentication feature information is stored, and if it is stored, then the corresponding object is determined Pass the certification. Taking shopping in an unmanned supermarket as an example, in order to distinguish between different objects, the object needs to be authenticated before the object leaves the unmanned supermarket or the product is settled. For authentication, the object is not sure of the settler of this purchase. Objects that have passed authentication are objects that have been authenticated or that meet the corresponding authentication conditions. Objects that do not meet the authentication conditions will not be the target objects. Therefore, in this step, the objects that fail the authentication are filtered, and the objects that pass the authentication are determined as the objects to be locked.

In addition, there can be one or two objects to be locked, or multiple, or even zero. If all objects in a specific area pass the authentication, these objects can be determined as objects to be locked.

In this step, multiple objects are authenticated, and the authenticated object is determined as the object to be locked, which can narrow the scope of the target object from the multiple objects and improve the efficiency of the target object locking process.

S203. Receive a lock trigger instruction, and prompt the object to be locked to perform a specific action according to the lock trigger instruction.

Among them, prompting the object to be locked to perform a specific action can be done through broadcast, display, bracelet, etc. Among them, the display can be installed in a fixed position in a specific area, or it can be a display screen on the user's mobile terminal. The embodiment of the present invention does not limit the way of prompting the object.

In addition, the specific action can be an action preset by the system, or an action customized by the object to be locked. Further, the specific motion may be a head motion, a body motion, and the like.

S204: Obtain locking feature information when the object to be locked performs the specific action, and determine the object to be locked whose locking feature information meets a preset condition as a target object.

When the object to be locked performs a specific action, it can generate information such as action posture, action amplitude, and biological characteristics. Therefore, the lock feature information can refer to information such as action feature information, biological feature information, and so on. Among them, the action feature information refers to information such as action posture and action amplitude. Biometric information refers to fingerprints, facial features, iris, body shape and other information.

Among them, the target object refers to the shopping leader among multiple objects in a specific area (for example, when multiple objects enter an unmanned supermarket together, one or more objects may be accompanied, and the true leader It is the target object that needs to be locked), the leader is the settler of this shopping, and the commodities selected by these objects can be determined as the commodities selected by the leader, and the leader will settle these commodities.

In addition, the target object can be one, two or more. If multiple objects in a specific area are entered in batches, the corresponding target objects can be determined from the objects in each batch.

In this embodiment, when the object to be locked is determined, the locking characteristic information of the object to be locked when performing a specific action is acquired. The locking characteristic information can effectively distinguish different objects, so that an accurate target object can be determined from multiple objects. , In order to lock the target object to complete this shopping behavior.

In one embodiment, the step of prompting the object to be locked to perform a specific action according to the lock trigger instruction; the step of acquiring the lock feature information when the object to be locked performs the specific action includes: according to the lock The trigger instruction prompts the object to be locked to perform a specific head movement; obtains the movement characteristic information and the first biological characteristic information of the object to be locked in the head movement process, according to the movement characteristic information and the first biological characteristic information Obtain the locking feature information of the object to be locked.

Among them, head movement can be head movement, for example, swinging the head left and right, nodding, raising the head, etc. Some restrictions can be imposed on the frequency range of these movements, of course, there is no restriction; in addition, head movement can also refer to Changes in facial expressions, such as smiling, blinking, etc.

What this embodiment implements is the process of acquiring the lock feature information of the object to be locked. The locking feature information is the action feature information and the first biological feature information of the object to be locked in the process of performing a specific action, and is information that can uniquely indicate the corresponding object to be locked. Therefore, the target object can be accurately determined from the objects to be locked through the lock feature information.

In one embodiment, the process of determining the target object may be: when it is detected that the consumer is standing in front of the confirmation device (the confirmation device may be a camera, a computer, etc.), prompting the consumer to move the head left and right in a certain order, Acquire the lock feature information of the consumer during the movement, and when the lock feature information matches the previously determined candidate lock feature information, confirm that the consumer is the target object, that is, the leader of this shopping.

In one embodiment, the step of prompting the object to be locked to perform a specific action according to the lock trigger instruction; and the step of obtaining the locking feature information when the object to be locked performs the specific action further includes: The lock trigger instruction acquires the preset first encoding information, displays the encoding pattern corresponding to the first encoding information, prompts the object to be locked to scan the encoding pattern; acquires the information when the object to be locked scans the encoding pattern Encoding scanning information, and obtaining the locking feature information of the object to be locked according to the encoding scanning information.

In this embodiment, the locking feature information is determined by scanning the coding pattern corresponding to the first coding information by the object to be locked. The first encoding information of different objects is different, so the unique locking feature information can be determined by encoding the scanning information. Therefore, the target object can be accurately determined from the objects to be locked through the lock feature information.

In an embodiment, the first code information may refer to the object number, user name, barcode, one-dimensional code, two-dimensional code, and so on.

Further, the object to be locked can provide its two-dimensional code during its authentication process, and the system saves the corresponding two-dimensional code information. When the lock feature information needs to be obtained, the QR code of the object is read and displayed. When the corresponding object to be locked scans the QR code and confirms the QR code information, the corresponding QR code information is used as the lock feature information.

In an embodiment, the first encoding information of the object may not be stored in advance. When it is necessary to obtain the locking feature information of the object to be locked, the object to be locked is prompted to show its terminal two-dimensional code information. After the system scans, the two-dimensional code is determined as the locking feature information.

In one embodiment, the process of obtaining the first encoding information of the object may be: prompting the object to provide the encoding information, and the object inputs the corresponding encoding information through an external interface. It can also be: prompt the object to provide coding information, the object displays the two-dimensional code on the mobile terminal, and the system scans the two-dimensional code to obtain the corresponding two-dimensional code information. It can also be: after receiving the registration confirmation information sent by the object, the unique coding information is automatically generated for the object and fed back to the object.

In one embodiment, the step of determining the object to be locked whose locking characteristic information meets a preset condition as the target object includes: when the locking characteristic information matches the pre-stored candidate locking characteristic information, determining the The lock feature information meets the preset conditions.

Wherein, the lock feature information and the candidate lock feature information can match either because they are consistent, or the similarity is greater than a certain threshold, and so on.

The candidate locking feature information can be information pre-stored by the system, or can be customized feature information of the object to be locked.

In this embodiment, the acquired lock feature information is compared with the candidate lock feature information, and when the two match, it is determined that the lock feature information meets the preset condition, and the corresponding object to be locked is determined as the target object. The process of determining the lock feature information is simple and accurate.

In one embodiment, after acquiring the lock characteristic information of the object to be locked, the corresponding object to be locked may also be directly determined as the target object. That is, if a certain object to be locked does not perform a specific action, it can be considered that the corresponding object to be locked is not the target object, and as long as it performs the corresponding specific action, it is considered to be the target object.

In one embodiment, after the lock feature information is determined, the target object can be determined from the objects to be locked based on the lock feature information. Therefore, taking the object as the consumer as an example, the process of determining the target object can also be: when one of the consumers P1 to be locked stands in front of the confirmation device, acquiring and displaying the locking feature information corresponding to the consumer P1, and prompting the consumer P1 confirms whether it is the leader of this shopping. When receiving the confirmation message from the consumer P1, it is considered that the consumer P1 confirms that he is the leader of this shopping. Lock the consumer P1 and feed back its characteristic information in real time through the display. When another consumer P2 to be locked stands in front of the confirmation device, the lock feature information corresponding to the consumer P2 is obtained, and the consumer P2 is prompted to confirm whether it is the leader of this shopping. When receiving the denial information of the consumer P2 or the confirmation information of the consumer P2 has not been received after a certain period of time, the consumer P2 is considered to confirm that he is not the leader of this shopping, and the information of the leader of this shopping is displayed. When the consumer P3 who has not passed the authentication stands in front of the confirmation device, the characteristic information corresponding to the consumer and/or the information of the current shopping leader is displayed.

In one embodiment, the process of determining the target object may also be: when it is detected that the consumer is standing in front of the confirmation device, displaying a QR code on the display and prompting the consumer to scan the QR code through the mobile terminal. When the consumer scans the corresponding QR code, it is confirmed that the consumer is the leader of the shopping.

In one embodiment, the step of authenticating the plurality of objects according to the authentication characteristic information includes: comparing the authentication characteristic information of any one object with pre-stored candidate authentication characteristic information to determine Whether there is candidate authentication feature information that matches the authentication feature information; if so, it is determined that the corresponding object has passed the authentication.

In this embodiment, the authentication feature information is compared with the candidate authentication feature information to determine whether there is candidate authentication feature information that matches the authentication feature information. If it exists, it means that the corresponding object is authenticated, and the authenticated object is determined as the object to be locked; if it does not exist, it means that the corresponding object has not been authenticated. The process of determining the lock feature information is simple and accurate.

In an embodiment, the candidate authentication feature information may be information pre-defined by the system, or information customized by the object to be locked.

For the self-defined candidate authentication feature information, in one embodiment, before the step of comparing the authentication feature information of any object with the pre-stored candidate authentication feature information, the method further includes: acquiring the biological feature information of the object, Store the biometric feature information as candidate authentication feature information; or, prompt the object to perform a specific action, obtain the action feature information during the execution of the specific action by the object, and store the action feature information as candidate authentication feature information.

In one embodiment, after the step of judging whether there is candidate authentication feature information matching the authentication feature information, the method further includes: if it does not exist, displaying the registration request information through the authentication device; when the object is received according to the When the registration confirmation information is sent by the registration request information, the third biometric information and/or coding information of the object is acquired, and the third biometric information and/or coding information is stored as candidate authentication characteristic information.

In this embodiment, when the object fails to pass the authentication, it is determined according to the willingness of the object whether to register and authenticate the object that has not passed the authentication. When an object that has not passed authentication is willing to perform registration authentication (send registration confirmation information), the third biometric information and/or coding information of the object that has passed authentication is used as candidate authentication characteristic information.

In one embodiment, when the registration confirmation information is received, the object may also be prompted to perform a specific action, and the candidate authentication feature information is determined according to the action feature information and/or biometric information in the process of the object performing the specific action. If the candidate authentication feature information contains action feature information, the corresponding object can be required to perform a specific action in the process of authenticating the object.

In an embodiment, after the step of authenticating the plurality of objects according to the first characteristic information, the method further includes: storing the authentication results of authenticating the plurality of objects; obtaining the object display information of the objects that have not passed the authentication, and displaying The object displays information.

For example, when the object is authenticated, information such as "Authenticated" can be displayed. When the object is not authenticated, the object display information of the object can be displayed, such as "first visit to our store" and so on.

In an embodiment, the step of comparing the authentication characteristic information of any object with pre-stored candidate authentication characteristic information includes: comparing the second biological characteristic information of any object with the pre-stored candidate biological information The feature information is compared.

In this embodiment, the biological feature information is used as the pre-stored authentication feature information.

In an embodiment, the step of comparing the authentication feature information of any object with pre-stored candidate authentication feature information includes: comparing the second code information of any object with the pre-stored second code information Information is compared.

In this embodiment, the encoded information is determined as pre-stored authentication feature information.

This embodiment and the previous embodiment compare the authentication feature information with the pre-stored candidate authentication feature information, which can accurately authenticate the object and improve the accuracy of target object determination.

In one embodiment, the specific area includes an unmanned supermarket; after the step of determining the target object whose lock feature information meets preset conditions as the target object, the method further includes: acquiring the target object in the unmanned supermarket. The real-time information of the supermarket determines the shopping information of the target object according to the real-time information.

In this embodiment, after the target object is determined, the target object is locked, and the state of the target object is marked as a locked state. Obtain real-time information in the process of shopping in the unmanned supermarket, and determine the shopping list of the target object based on the real-time information.

In one embodiment, the process of tracking (locking) the target object may be: when the target object selects a product in a specific area, as long as the camera device can capture the target object (the target object is not completely blocked by obstacles, etc.), the processor If it is judged that the key information of the target object can still be recognized in real time, the target object is still marked in the locked state, and the relevant information of the target object is displayed in real time on the display.

In one embodiment, after the target object is locked, when the target object cannot be normally recognized for a certain period of time, it is determined that the current locked state of the target object is abnormal, and the target object is prompted through the display. If the recognition is restored within the t1 time period, the lock on the target object is restored. The period of abnormal identification of the target object is determined as the identity information pending state, and any commodity selected by the target object during this period of time is judged to be illegal. If the recognition is resumed in the t1 time period, the product selection information of the target object can be confirmed, the product information added by the target object in the pending status of the identity information can be determined, and the shopping list of the target object can be generated based on the product information.

If the recognition has not been restored after t1, the locked state and related orders are deemed to be automatically abandoned. At the same time, in order to avoid the loss of goods, the target object is prompted to return the corresponding goods through audio and video methods. After returning, the target person can purchase again. If the target object does not return the goods, the account can be automatically deducted from the account bound to the target object (the account can be bound during the authentication process).

In one embodiment, the shopping list of the determined target object may be the product information in the shopping cart of the determined target object, and the shopping list is generated according to the product information. It can also be to determine the product information of the target object and its accompaniment (the target object and its escort enter a specific area within the same time period) in a specific area, and determine the products they take as the target object’s Commodities and generate the corresponding shopping list. If there are target objects in the companions of a target object, and the products they choose are placed in the same shopping cart, you can request these target objects to determine the products they want to settle.

In an embodiment, as shown in FIG. 4, a method for locking a target object is provided, which includes the following steps:

S401: When multiple objects enter a specific area, obtain authentication feature information of the multiple objects.

S402. Compare the authentication feature information of any object with pre-stored candidate authentication feature information, and determine whether there is candidate authentication feature information that matches the authentication feature information; if it exists, determine that the corresponding object passes Authentication, the object that has passed authentication is regarded as the object to be locked.

S403. Receive a lock trigger instruction, and prompt the object to be locked to perform a specific head movement according to the lock trigger instruction.

S404: Obtain the movement characteristic information and the first biometric information of the object to be locked during the head movement, and obtain the locking characteristic information of the object to be locked according to the movement characteristic information and the first biometric information.

S405: When the lock feature information matches the pre-stored candidate lock feature information, determine that the lock feature information meets a preset condition, and determine an object to be locked whose lock feature information meets the preset condition as a target object.

S406. Acquire real-time information of the target object in an unmanned supermarket, and determine the shopping information of the target object according to the real-time information.

In this embodiment, in the case of determining the object to be locked, the locking characteristic information when the object to be locked performs a specific action is acquired. The locking characteristic information can effectively distinguish different objects, so the target object can be accurately determined from multiple objects .

In order to better understand the above method, an application example of the method for locking the target object of the present invention is described in detail below. The camera is arranged in an unmanned supermarket, and the method of locking the target object is described from the perspective of the video detection system, taking the object as the consumer as an example.

When two consumers A and B enter an unmanned supermarket, the gate of the unmanned supermarket is equipped with a camera device, an authentication display, and a confirmation display (the authentication display and the confirmation display can be the same display), and the target object is locked. The locking process can be:

1) The video detection system receives authentication feature information such as facial and body characteristics of A and B collected by a camera device facing the door of an unmanned supermarket.

2) The video detection system judges whether there is candidate authentication feature information matching these authentication feature information among the stored candidate authentication information. If there is candidate authentication feature information that matches A's authentication feature information, it means that A is authenticated; if there is no candidate authentication feature information that matches B's feature information, it means that B has not been authenticated.

3) The video detection system determines the authenticated A as the target to be locked. The video detection system stores the certification results of A and B, and controls the display to display the certification results of A and B simultaneously.

4) Ask the uncertified B through the broadcast whether it needs to be certified and ask B to move to the certification display at the door of the unmanned supermarket for certification. When B stands on the authentication display, prompts B to move his head left and right, extract biometric information such as B’s facial features, and use the biometric information as the candidate authentication feature information for the consumer, and set the authenticated flag for B. The authenticated identifier and candidate authentication feature information are stored, and the authentication display prompts B that the authentication is successful.

5) When A is standing in front of the confirmation display, the head is prompted to move left, right, up and down in a certain order to obtain the lock feature information of the nail during the movement. Display the lock feature information of A, prompting A to confirm whether he is the leader of this shopping. When A presses the confirmation button on the display, it is determined that A is the leader of the shopping, locks A and feeds back his characteristic information in real time through the display.

When B stands in front of the confirmation display, it prompts B to move his head up and down in a certain order, obtains the locking feature information of B during the head movement, and prompts B to confirm whether he is the leader of this shopping. When B’s confirmation message is not received after a certain period of time, B is deemed not to be the leader of this shopping. The display shows that the leader of this shopping is A.

6) When A selects a product in an unmanned supermarket, as long as the camera device can obtain A’s information, the processor judges that it still recognizes A’s key information in real time, and its identity information is always locked, and A’s information is displayed in real time on the monitor. Related Information.

7) A completes shopping and enters the checkout process, and the checkout ends. When it is determined that A and B have left the unmanned supermarket according to the information of the camera device facing the gate, the shopping process is completed.

In this embodiment, when there are multiple consumers in an unmanned supermarket environment, the video recognition technology can realize the identification and identity lock of the shopping leader, and perform identity tracking during the shopping selection process, and provide real-time feedback through the display , In order to promote the consumer's consumption behavior.

It should be noted that for the foregoing method embodiments, for simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described sequence of actions, because According to the present invention, certain steps can be performed in other order or simultaneously.

Based on the same idea as the method for locking the target object in the foregoing embodiment, the present invention also provides a device for locking the target object, which can be used to execute the method for locking the target object as described above. For ease of description, in the structural schematic diagram of the embodiment of the locking device of the target object, only the parts related to the embodiment of the present invention are shown. Those skilled in the art can understand that the illustrated structure does not constitute a limitation on the device, and may include comparisons. More or fewer parts are shown, or some parts are combined, or different parts are arranged.

As shown in FIG. 5, the locking device of the target object includes an information acquisition module 501, an authentication module 502, a locking trigger module 503, and a target locking module 504. The detailed description is as follows:

The information obtaining module 501 is configured to obtain authentication feature information of multiple objects when the multiple objects enter a specific area.

The authentication module 502 is configured to authenticate the multiple objects according to the authentication feature information, and use the authenticated objects as the objects to be locked.

The lock trigger module 503 is configured to receive a lock trigger instruction, and prompt the object to be locked to perform a specific action according to the lock trigger instruction.

And, the target locking module 504 is configured to obtain the locking characteristic information when the target to be locked performs the specific action, and determine the target to be locked whose locking characteristic information meets a preset condition as the target object.

In one embodiment, it includes: a head movement module, configured to prompt the object to be locked to perform a specific head movement according to the lock trigger instruction; a first information acquisition module, configured to obtain the head movement of the object to be locked According to the action feature information and the first biological feature information during the movement of the part, the lock feature information of the object to be locked is obtained according to the action feature information and the first biological feature information.

In one embodiment, it further includes: a code scanning module, configured to obtain preset first code information according to the lock trigger instruction, display the code pattern corresponding to the first code information, and prompt the object to be locked to scan The coding pattern; a second information acquisition module for acquiring the coding scan information when the object to be locked scans the coding pattern, and obtaining the locking feature information of the object to be locked according to the code scanning information.

In one embodiment, the target locking module 504 is further configured to determine that the locking characteristic information meets a preset condition when the locking characteristic information matches the pre-stored candidate locking characteristic information.

In one embodiment, the authentication module 502 includes: an information comparison sub-module, which is used to compare the authentication feature information of any object with pre-stored candidate authentication feature information, and determine whether there is an authentication feature that matches the authentication feature information. The candidate authentication feature information that matches the information; the authenticated sub-module is used to determine that the corresponding object has passed the authentication if it exists.

In one embodiment, the information comparison sub-module is also used to compare the second biometric information of any object with pre-stored candidate biometric information.

In one embodiment, the information comparison sub-module is also used to compare the second code information of any object with the pre-stored second code information.

In one embodiment, it further includes: a shopping information obtaining module, configured to obtain real-time information of the target object in the unmanned supermarket, and determine the shopping information of the target object according to the real-time information.

It should be noted that the target object locking device of the present invention corresponds to the target object locking method of the present invention one-to-one, and the technical features and beneficial effects described in the foregoing embodiment of the target object locking method are applicable to the target object's locking method. In the embodiment of the locking device, the specific content can be referred to the description in the embodiment of the method of the present invention, which will not be repeated here, and it is hereby declared.

In addition, in the implementation of the locking device of the target object in the above example, the logical division of each program module is only an example. In actual applications, it can be based on needs, for example, due to the configuration requirements of the corresponding hardware or the convenience of software implementation. The above function allocation is completed by different program modules, that is, the internal structure of the locking device of the target object is divided into different program modules to complete all or part of the functions described above.

In one embodiment, as shown in FIG. 6, a video detection system 101 is provided, which includes a video processing module, a video detection module, an identity information locking module, an identity information verification module, an identity information authentication module, and a video information storage module , Display output module, identity information classification module. These modules are connected through a software link, a hardware link and a communication bus to realize a locking method of a target object.

Among them, the video processing module is used to process the detection data transmitted by the video detection module. The video processing module has a video output interface that is connected to a display, a player, or a speaker through a physical or wireless connection.

The video detection module is used to detect images, videos, texts, QR codes, human faces and other information within the detection range.

The video information storage module (the video information storage module may be a non-volatile memory) is used to store data and user identity information in the video processing module.

The identity information audit module is used to analyze the information output by the video-conferencing processor module, and determine the new users and/or old users among them.

The identity information classification module is used to classify and classify the identity information of these users after the identity information audit module determines the new and old users.

The identity information locking module is used to lock and track one or more users in real time after the user information classification module calibrates the identity information.

The display output module (the display output module can be wired or wirelessly connected to external output devices such as displays) is used for real-time output of video recognition results, commodity settlement and other information through the display.

The identity information authentication module is used to authenticate unauthenticated users and mark them as authenticated.

In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. 7. The computer equipment includes a processor, a memory, a network interface, and a database connected through a system bus. Among them, the processor of the computer device is used to provide calculation and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer equipment is used to store data such as authentication feature information, action feature information, and biological feature information. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method of locking the target object is realized.

Those skilled in the art can understand that the structure shown in FIG. 7 is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may Including more or fewer parts than shown in the figure, or combining some parts, or having a different arrangement of parts.

In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the following steps are implemented: Area, obtain the authentication feature information of the multiple objects; authenticate the multiple objects according to the authentication feature information, and use the authenticated object as the object to be locked; receive a lock trigger instruction, and according to the lock trigger instruction The object to be locked is prompted to perform a specific action; the locking feature information when the object to be locked performs the specific action is acquired, and the object to be locked whose locking feature information meets a preset condition is determined as the target object.

In one embodiment, the processor further implements the following steps when executing the computer program: prompting the object to be locked to perform a specific head movement according to the lock trigger instruction; acquiring the movement of the object to be locked in the process of head movement The characteristic information and the first biological characteristic information, and the locking characteristic information of the object to be locked is obtained according to the action characteristic information and the first biological characteristic information.

In one embodiment, the processor further implements the following steps when executing the computer program: acquiring preset first encoding information according to the lock trigger instruction, displaying the encoding pattern corresponding to the first encoding information, and prompting the to-be-locked The object scans the coding pattern; the code scan information when the object to be locked scans the code pattern is acquired, and the lock feature information of the object to be locked is obtained according to the code scan information.

In an embodiment, the processor further implements the following step when executing the computer program: when the lock feature information matches the pre-stored candidate lock feature information, it is determined that the lock feature information meets a preset condition.

In one embodiment, when the processor executes the computer program, the processor further implements the following steps: compare the authentication feature information of any object with pre-stored candidate authentication feature information, and determine whether there is a match with the authentication feature information Candidate authentication feature information; if it exists, it is determined that the corresponding object is authenticated.

In an embodiment, the processor further implements the following step when executing the computer program: comparing the second biometric information of any object with the pre-stored candidate biometric information.

In an embodiment, the processor further implements the following step when executing the computer program: comparing the second code information of any object with the pre-stored second code information.

In an embodiment, the processor further implements the following steps when executing the computer program: acquiring real-time information of the target object in the unmanned supermarket, and determining the shopping information of the target object according to the real-time information.

In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented: when multiple objects enter a specific area, obtain the information of the multiple objects Authentication feature information; authenticate the multiple objects according to the authentication feature information, and use the authenticated object as the object to be locked; receive a lock trigger instruction, and prompt the object to be locked to perform a specific action according to the lock trigger instruction; Acquire locking feature information when the object to be locked performs the specific action, and determine the object to be locked whose locking feature information meets a preset condition as the target object.

In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: according to the lock trigger instruction, prompt the object to be locked to perform a specific head movement; obtain information about the object to be locked in the process of head movement. The action feature information and the first biological feature information, and the lock feature information of the object to be locked is obtained according to the action feature information and the first biological feature information.

In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: acquiring preset first encoding information according to the lock trigger instruction, displaying the encoding pattern corresponding to the first encoding information, and prompting the waiting The locking object scans the coding pattern; acquiring the coding scan information when the to-be-locked object scans the coding pattern, and obtaining the locking feature information of the to-be-locked object according to the coding scanning information.

In one embodiment, when the computer program is executed by the processor, the following step is further implemented: when the lock feature information matches the pre-stored candidate lock feature information, it is determined that the lock feature information meets a preset condition.

In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: comparing the authentication feature information of any object with pre-stored candidate authentication feature information, and judging whether there is any object that is compatible with the authentication feature information. The matching candidate authentication feature information; if it exists, it is determined that the corresponding object passes the authentication.

In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: comparing the second biometric information of any object with the pre-stored candidate biometric information.

In one embodiment, when the computer program is executed by the processor, the following step is further implemented: comparing the second code information of any object with the pre-stored second code information.

In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: obtaining real-time information of the target object in the unmanned supermarket, and determining the shopping information of the target object according to the real-time information.

A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing relevant hardware through a computer program, and the program can be stored in a computer readable storage medium, as Independent product sales or use. More specific examples (non-exhaustive list) of computer-readable media include the following: electrical connections (electronic devices) with one or more wiring, portable computer disk cases (magnetic devices), random access memory (RAM), Read only memory (ROM), erasable and editable read only memory (EPROM or flash memory), fiber optic devices, and portable compact disk read only memory (CDROM). In addition, the computer-readable medium can even be paper or other suitable media on which the program can be printed, because it can be done, for example, by optically scanning the paper or other media, and then editing, interpreting, or other suitable media if necessary. The program is processed in a way to obtain the program electronically and then stored in the computer memory.

It should be understood that each part of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if it is implemented by hardware, as in another embodiment, it can be implemented by any one or a combination of the following technologies known in the art: Discrete logic circuits, application-specific integrated circuits with suitable combinational logic gates, programmable gate arrays (PGA), field programmable gate arrays (FPGA), etc.

The terms "including" and "having" and any variations thereof in the embodiments of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or device that includes a series of steps or (module) units is not limited to the listed steps or units, but optionally also includes unlisted steps or units, or optional The ground also includes other steps or units inherent to these processes, methods, products, or equipment.

The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the various technical features in the above-mentioned embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, All should be considered as the scope of this specification.

The above-mentioned embodiments only express several embodiments of the present invention, and cannot be understood as a limitation of the patent scope of the present invention. It should be pointed out that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can be made, and these all fall within the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention should be subject to the appended claims.

Claims

A method for locking a target object is characterized in that it comprises the following steps:

When multiple objects enter a specific area, acquiring authentication feature information of the multiple objects;

Authenticate the multiple objects according to the authentication feature information, and use the authenticated objects as the objects to be locked;

Receiving a lock trigger instruction, and prompt the object to be locked to perform a specific action according to the lock trigger instruction;

Acquire locking feature information when the object to be locked performs the specific action, and determine the object to be locked whose locking feature information meets a preset condition as the target object.
The method for locking a target object according to claim 1, wherein the prompting the object to be locked to perform a specific action according to the lock trigger instruction; acquiring the lock when the object to be locked performs the specific action The steps of characteristic information include:

Prompting the object to be locked to perform a specific head movement according to the lock trigger instruction;

Obtain the movement characteristic information and the first biological characteristic information of the object to be locked during the head movement, and obtain the locking characteristic information of the object to be locked according to the movement characteristic information and the first biological characteristic information.
The method for locking a target object according to claim 1, wherein the prompting the object to be locked to perform a specific action according to the lock trigger instruction; acquiring the lock when the object to be locked performs the specific action The step of feature information also includes:

Acquiring preset first encoding information according to the lock trigger instruction, displaying an encoding pattern corresponding to the first encoding information, and prompting the object to be locked to scan the encoding pattern;

Obtain code scan information when the object to be locked scans the code pattern, and obtain lock feature information of the object to be locked according to the code scan information.
The method for locking a target object according to claim 1, wherein the step of determining an object to be locked whose locking feature information meets a preset condition as the target object comprises:

When the lock feature information matches the pre-stored candidate lock feature information, it is determined that the lock feature information meets a preset condition.
The method for locking a target object according to claim 1, wherein the step of authenticating the multiple objects according to the authentication characteristic information comprises:

The authentication feature information of any object is compared with the pre-stored candidate authentication feature information to determine whether there is candidate authentication feature information that matches the authentication feature information; if it exists, it is determined that the corresponding object is authenticated.
The method for locking a target object according to claim 5, wherein the step of comparing the authentication characteristic information of any object with pre-stored candidate authentication characteristic information comprises:

Compare the second biometric information of any object with the pre-stored candidate biometric information;

and / or,

The second coding information of any object is compared with the pre-stored second coding information.
The method for locking a target object according to any one of claims 1 to 6, wherein the specific area includes an unmanned supermarket;

After the step of determining the object to be locked whose locking feature information meets the preset condition as the target object, the method further includes:

Acquire real-time information of the target object in the unmanned supermarket, and determine the shopping information of the target object according to the real-time information.
A locking device for a target object, characterized in that it comprises:

An information acquisition module, configured to acquire authentication feature information of multiple objects when the multiple objects enter a specific area;

An authentication module, configured to authenticate the multiple objects according to the authentication feature information, and use the authenticated objects as the objects to be locked;

The lock trigger module is configured to receive a lock trigger instruction, and prompt the object to be locked to perform a specific action according to the lock trigger instruction;

And, the target locking module is configured to obtain the locking characteristic information of the object to be locked when performing the specific action, and determine the object to be locked whose locking characteristic information meets a preset condition as the target object.
A computer device, comprising a memory, a processor, and a computer program stored on the memory and running on the processor, wherein the processor implements the computer program described in any one of claims 1 to 7 when the processor executes the computer program. The steps of the method described.
A computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the method according to any one of claims 1 to 7 when the computer program is executed by a processor.