CN111223219A

CN111223219A - Identity recognition method and storage medium

Info

Publication number: CN111223219A
Application number: CN201911416728.1A
Authority: CN
Inventors: 区国雄; 林峰
Original assignee: Shenzhen Fushi Technology Co Ltd
Current assignee: Shenzhen Fushi Technology Co Ltd
Priority date: 2019-12-31
Filing date: 2019-12-31
Publication date: 2020-06-02

Abstract

The application is applicable to the technical field of optics and electronics, and provides an identity recognition method for recognizing the identity of a visitor according to a preset voice password template, a two-dimensional face data template and a three-dimensional face data template which are related to the identity of a target object, wherein the identity recognition method comprises the following steps: acquiring voice information of a visitor; extracting the voice password spoken by the visitor from the acquired voice information of the visitor, and comparing the voice password with a voice password template of a preset target object to carry out password identification; acquiring image information of a visitor after the password identification is passed; extracting two-dimensional face data or three-dimensional face data of the visitor from the acquired image information of the visitor, and comparing the two-dimensional face data or the three-dimensional face data with a preset two-dimensional face data template or a preset three-dimensional face data template of a target object to perform face recognition; and confirming the identity of the visitor as a target object after the face recognition is passed. The present application also provides a storage medium.

Description

Identity recognition method and storage medium

Technical Field

The present application relates to biometric identification, and more particularly, to an identity recognition method and a storage medium.

Background

The traditional identity recognition device generally needs to recognize the identity by means of an identification card, so that the recognition efficiency is low, the use is very inconvenient under the condition that both hands are occupied, and the problem that the identification card which needs to be carried about is easy to lose is also solved.

Disclosure of Invention

The technical problem to be solved by the present application is to provide an identity recognition method and a storage medium, which can recognize the identity of a visitor without an identification card.

The embodiment of the application provides an identity recognition method, which is used for recognizing the identity of a visitor according to a preset voice password template, a two-dimensional face data template and a three-dimensional face data template which are related to the identity of a target object, and comprises the following steps: acquiring voice information of a visitor; extracting the voice password spoken by the visitor from the acquired voice information of the visitor, and comparing the voice password with a voice password template of a preset target object to carry out password identification; acquiring image information of a visitor after the password identification is passed; extracting two-dimensional face data or three-dimensional face data of the visitor from the acquired image information of the visitor, and comparing the two-dimensional face data or the three-dimensional face data with a preset two-dimensional face data template or a preset three-dimensional face data template of a target object to perform face recognition; and confirming the identity of the visitor as a target object after the face recognition is passed.

In some embodiments, before obtaining the voice information of the visitor, whether the visitor approaches is sensed, and after the visitor approaches is sensed, the voice information of the visitor is obtained.

In some embodiments, the method further comprises the following steps after the approach of the visitor is sensed: and reminding the visitor to identify the identity by displaying a reminder or making a sound.

In some embodiments, an alarm password, alarm information, and alarm object data are preset, and the identity recognition method further includes the steps of:

and if the voice password of the visitor is identified as an alarm password in the password identification, respectively sending the alarm information to corresponding alarm objects according to preset alarm object data, and identifying through the password.

In some embodiments, after the voice password of the visitor is identified as the alarm password in the password identification, the method further comprises the following steps: and recording the image information and the sound information of the scene.

In some embodiments, a two-dimensional and/or three-dimensional image of the face of the visitor is acquired for face motion recognition when password recognition is performed to determine whether the recognized visitor is living, and password recognition is allowed when the recognized visitor is living.

In some embodiments, the three-dimensional face data of the visitor is acquired according to a structured light pattern sensing principle, a binocular vision sensing principle, and/or a time-of-flight sensing principle.

In some embodiments, after confirming the identity of the visitor as the target object, the identity authority corresponding to the target object is granted to execute the corresponding operation.

In some embodiments, a voiceprint feature template of the target object is also preset, and the identity recognition method further includes the steps of:

after the password identification is passed, voiceprint characteristics of the visitor are extracted from the obtained voice information of the visitor and are compared with a preset voiceprint characteristic template of the target object to carry out voiceprint identification; and

confirming that the identity of the visitor is a target object after the voiceprint recognition and the face recognition are passed;

wherein the voiceprint recognition and face recognition are performed simultaneously; or, the voiceprint recognition is executed first, and the face recognition is executed after the voiceprint recognition is passed.

A storage medium is provided for storing program code executable by one or more processors. The program is executable by the processor to implement the identification device control method as described in the above embodiment.

The identity recognition device, the identity recognition method and the storage medium provided by the application carry out identity recognition to open the door by collecting various characteristic information of a visitor, and a target object does not need to carry a key or an access card. And different characteristic information is set to carry out combined identification according to a reasonable logic sequence, so that the safety and convenience of identity identification can be considered.

Additional aspects and advantages of embodiments of the present application will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of embodiments of the present application.

Drawings

Fig. 1 is a schematic functional block diagram of an identification apparatus according to a first embodiment of the present application.

Fig. 2 is a schematic position diagram of the image capturing module shown in fig. 1.

FIG. 3 is one embodiment of the identification module of FIG. 1.

Fig. 4 is one implementation of the identification module of fig. 1.

FIG. 5 is one embodiment of the identification module of FIG. 1.

Fig. 6 is a flowchart illustrating steps of an identification method according to a first embodiment of the present disclosure.

Fig. 7 is one implementation of the identification method shown in fig. 6.

Fig. 8 is another implementation of the identification method of fig. 6.

Fig. 9 is a further implementation of the identification method of fig. 6.

Fig. 10 is a functional block diagram of an identification device according to a second embodiment of the present application.

Fig. 11 is a flowchart illustrating steps of an identification method according to a second embodiment of the present application.

Detailed Description

Reference will now be made in detail to embodiments of the present application, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the accompanying drawings are illustrative and are only for the purpose of explaining the present application and are not to be construed as limiting the present application. In the description of the present application, it is to be understood that the terms "first" and "second" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implying any order or number of technical features indicated. Thus, features defined as "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the present application, "a plurality" means two or more unless specifically limited otherwise.

In the description of the present application, it should be noted that, unless explicitly stated or limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; either mechanically or electrically or in communication with each other; either directly or indirectly through intervening media, either internally or in any other relationship or combination of two or more elements. The specific meaning of the above terms in the present application can be understood by those of ordinary skill in the art as appropriate.

The following disclosure provides many different embodiments or examples for implementing different structures of the application. In order to simplify the disclosure of the present application, only the components and settings of a specific example are described below. Of course, they are merely examples and are not intended to limit the present application. Moreover, the present application may repeat reference numerals and/or letters in the various examples, such repeat use is intended to provide a simplified and clear description of the present application and is not intended to suggest any particular relationship between the various embodiments and/or arrangements discussed. In addition, the various specific processes and materials provided in the following description of the present application are only examples of implementing the technical solutions of the present application, but one of ordinary skill in the art should recognize that the technical solutions of the present application can also be implemented by other processes and/or other materials not described below.

Further, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to provide a thorough understanding of embodiments of the application. One skilled in the relevant art will recognize, however, that the subject technology can be practiced without one or more of the specific details, or with other structures, components, and so forth. In other instances, well-known structures or operations are not shown or described in detail to avoid obscuring the focus of the application.

As shown in fig. 1, a first embodiment of the present application provides an identification apparatus 1 for identifying the identity of a visitor. The identification device 1 can be disposed on a device requiring identification, such as: monitoring equipment, access control equipment or automobiles and the like. The identification device 1 comprises an information collector 13, a sensor 14, an interactive device 15, a power supply 17, a storage medium 18, a processor 19, a network connector 101 and a control system 11. The information collector 13, the sensor 14, the interactive device 15, the power supply 17, the storage medium 18, the processor 19 and the network connector 101 are connected to each other through a bus to transmit data and signals to each other. The information collector 13 is used for obtaining information related to the visitor so as to extract features related to the identity of the visitor. The information collector 13 includes, but is not limited to, an image collecting module 132 and a sound collecting module 134, and correspondingly, the obtained information includes, but is not limited to, image information and sound information of the visitor. The sensor 14 is used for sensing whether a visitor approaches and awakening the information collector 13 through the control system 11 to perform identification operation when the visitor approaches.

Optionally, in some embodiments, the control system 11 is stored in the storage medium 18 and executed by the processor 19, and is configured to analyze the image information and the sound information of the visitor sensed by the information collector 13 to obtain a biometric feature of the visitor, and compare the biometric feature with an identity feature template of a target object pre-stored in the storage medium 18 to determine whether the visitor is the target object.

The interaction device 15 is used to provide an interaction interface. The storage medium 18 is used for storing the identity template of the target object and the data generated in the identification process. The network connector 101 is used for connecting the internet to realize remote control. The power supply 17 supplies power to the above components.

The sensor 14 may be, but is not limited to, an acoustic sensor, a proximity sensor, an electromagnetic sensor, and a touch sensor. Correspondingly, the sensor 14 senses whether someone wants to open the door in different ways, for example: sensing whether a sound is generated in front of the door through a sound sensor, and if the sound is sensed, indicating that further identity identification is possibly needed; or whether an object approaches within a preset distance is sensed through the proximity sensor, if so, the object is indicated to possibly need further identity recognition, and the proximity sensor can realize proximity sensing within the preset distance by emitting and receiving infrared rays; or by providing a touch sensor to sense whether there is a touch, sensing sometimes indicating that further identification may be required. The touch sensor can be integrated, for example, in a door bell and a door handle; or the electromagnetic sensor establishes a detection magnetic field, when a visitor approaches the magnetic sensor, the detection magnetic field changes, and whether the visitor approaches the magnetic sensor is sensed by sensing the detection magnetic field changes.

The sensor 14 may sense at a preset frequency. The preset frequency may be adjusted in a range of 0 to 2 seconds/time, for example: the sensor 14 emits infrared sensing light every 1 second, and determines whether there is a visitor in front by sensing whether the infrared sensing light is reflected back. The adjustment range of the preset distance may be 0 to 3 meters. Because the sensor 14 does not need to perform complex identification and analysis, and the power consumption during operation is low, the sensor 14 is used to sense that a visitor appears before the door 10 and then awakens the information collector 13 with high power consumption, such as: the image collection module 132 and the sound collection module 134 can effectively reduce the overall power consumption of the identification device 1.

It is understood that in other embodiments, the sensor 14 may be omitted, and the information collector 13 is turned on periodically according to a preset frequency to directly obtain visitor information for identification. In addition, the sensor 14 may also be replaced by a switch key, and the target object starts the information collector 13 to operate by pressing or touching the switch key, so as to save power consumption.

The image capturing module 132 is configured to obtain three-dimensional image information of a visitor. The three-dimensional image information can be used for extracting image characteristics related to the identity of the visitor. The image features include, but are not limited to, three-dimensional face data, gestures, stature, body pose, iris, etc. of the visitor. The image capture module 132 includes a three-dimensional camera.

The three-dimensional camera acquires a three-dimensional (3D) stereoscopic image of a visitor. The three-dimensional camera may comprise components corresponding to the three-dimensional sensing principle used, according to different three-dimensional sensing principles. For example: a structured light projector and a structured light pattern sensor according to the structured light pattern sensing principle, at least two image sensors according to the binocular vision sensing principle, or a light emitter and a light receiver according to the Time Of Flight (TOF) sensing principle.

It is understood that the image capturing module 132 may capture a whole body image, an upper body image, or a specific part of the visitor, such as: images of the face, hands and/or eyes. The processor 19 may perform corresponding identification according to the content of the acquired image. For example: the processor 19 may perform face recognition and face motion recognition based on the acquired two-dimensional face data or three-dimensional face data, gesture recognition and stature recognition based on the acquired three-dimensional whole body image, gesture recognition based on the acquired three-dimensional hand image, iris recognition based on the acquired eye image, and the like.

Alternatively, in some embodiments, the image capturing module 132 may further comprise a color camera for acquiring a two-dimensional color (RGB) image of the visitor. According to actual needs, the color camera can comprise a light supplement element, or a lighting device nearby is started through the internet or remote control technology to supplement the shortage of ambient light.

The number of the image capturing modules 132 may be one or more. A plurality of the image capturing modules 132 may be disposed at different positions, respectively. For example: the identity recognition device 1 is arranged on the access control equipment 3. The access control device 3 may be a device that is broadly installed at a doorway of a specific area such as a factory, a company, an office building, an exhibition hall, a residential quarter, or the like, and is used for identity recognition, or may be a door that is installed at a doorway of an individual home or room in the conventional sense.

As shown in fig. 2, the access control device 3 may include a door 10 and a door lock 12. The image capturing module 132 may be disposed at one or more different positions on the door 10, including but not limited to the first region 102 on the upper center of the door 10, the second region 104 on the lower center of the door 10, and the four corner regions 105 of the door 10. The different positions of the image capturing module 132 are set for capturing image information for visitors with different heights, or for obtaining more comprehensive three-dimensional image information of the visitors from different angles. The door lock 12 is used for locking the door body 10. The door lock 12 may include a mechanical lock and/or an electronic lock. The mechanical lock can be locked or unlocked through an electric drive mechanical structure, and can also be manually locked or unlocked through a key so as to be used in case of power failure or failure of related electronic components. The electronic lock can be locked or unlocked by electric signals, such as: an electromagnetic attraction type door lock 12. The sound collection module 134 is used for acquiring the voice information sent by the visitor. The voice information includes but is not limited to voice characteristics related to the visitor identity, such as: voiceprint features and voice passwords. The collected voiceprint features include, but are not limited to, the spectrum, timbre, prosody, etc. of the sound. The voice password refers to the content spoken by the voice information and can be recognized after being converted into characters through a voice recognition technology. The control system 11 determines the identity of the visitor based on the voiceprint feature and the voice password in the obtained voice message.

The interaction device 15 includes, but is not limited to, a touch screen, a doorbell, and/or a speaker for displaying or sending prompt information and receiving the operation of the visitor to realize the interaction. For example, when the voiceprint characteristics of the visitor need to be sensed, the specific content which needs to be read out by the visitor can be displayed through the touch screen or sent out through the loudspeaker. Alternatively, the visitor may select a different identification mode to unlock the door lock 12 through operation of the touch screen. Alternatively, the visitor may initiate an internet connection by pressing a doorbell to enable the target object to remotely identify its identity.

The power supply 17 may provide power for each component by connecting with the mains supply and performing corresponding adaptation processing. The power supply 17 may also include an energy storage element such as a battery to continue to provide power to the various components during a power outage.

The storage medium 18 includes, but is not limited to, a Flash Memory (Flash Memory), a charged Erasable Programmable read only Memory (EEPROM), a Programmable Read Only Memory (PROM), and a hard disk. The storage medium 18 is used for storing an identity template of a preset target object, intermediate data generated in the identification process, computer software codes for realizing identification and control functions, and the like.

The control system 11 includes one or more functional modules 110, and the functional modules 110 may be firmware solidified in the corresponding storage medium 18 or computer software codes stored in the storage medium 18. The functional modules 110 are executed by the corresponding one or more processors 19 to control the relevant components to implement the corresponding functions. The control system 11 includes, but is not limited to, a setup module 111, an interaction module 112, a connection module 113, an identification module 114, a control module 115, and a sensing module 116.

The setting module 111 is configured to set an identity template and an identification mode of the target object. The identification mode refers to the combination of relevant parameters such as the type, the times and the execution sequence of the identification operation needing to be executed. The target object can set recognition modes with different types of identity recognition functions according to self conditions. The identification mode may include one or more identification operations. The execution sequence of any different types of identification operations included in the same identification pattern may be a relationship caused in sequence, such as: the first identification operation is passed first and then the second identification operation is carried out. The execution sequence between any different types of identification operations included in the same identification pattern may also be a relationship that is satisfied simultaneously, such as: the first identification operation and the second identification operation are carried out synchronously and must pass to carry out the next step. And the identity characteristic template of the target object corresponds to the set type of the identity recognition operation. In this embodiment, the identification types include, but are not limited to, two-dimensional face recognition, three-dimensional face recognition, voiceprint recognition, voice password recognition, gesture recognition, stature recognition, body gesture recognition, and iris recognition. Correspondingly, the identity feature template of the target object includes, but is not limited to, a three-dimensional face data template of the target object, a two-dimensional face data template of the target object, a voiceprint feature template of the target object, a preset voice password template of the target object, a preset gesture template of the target object, a figure template of the target object, a body posture template of the target object, and an iris feature template of the target object.

It is understood that in the present embodiment and the modified embodiment, a plurality of three-dimensional face data templates, two-dimensional face data templates, voiceprint feature templates, and iris feature templates of different target objects may be respectively set, or a plurality of different password templates and gesture templates may be respectively set, so as to improve the applicability of the identity recognition apparatus 1.

It is understood that, in the present embodiment and the modified embodiments, the identification of the visitor may be performed, and the face motion recognition may be performed based on the acquired two-dimensional and/or three-dimensional face image to determine whether the detection target is a living body. For example: whether the target object is a living body is judged by identifying whether the eye blinks or whether the mouth moves when the password is spoken, so as to prevent the visitor from attacking the target object by imitating the characteristics of the target object by using a three-dimensional mask and a recording.

The interaction module 112 is used for controlling the interaction device 15 to interact with the target object. For example: different recognition modes are set and/or selected through the touch screen. Or, the visitor is informed to read out specific contents through a touch screen and/or a loudspeaker when the voiceprint recognition is carried out. Or when the voice password is identified, the visitor is informed to speak the voice password through the touch screen and/or the loudspeaker. Or, communicate with the target object through a touch screen and/or a speaker, and the like.

The connection module 113 is used to control the network connector 101 to connect to the internet, so that the target object can remotely control the identification apparatus 1 through the mobile terminal device. For example: the visitor is viewed through the mobile terminal. The network connector 101 may be connected to the internet in a wired or wireless manner. The connection module 113 may also control the network connector 101 to disconnect the internet connection according to a preset setting, so as to reduce the risk of a hacker capturing the control right of the identification apparatus 1 through the internet connection. Specifically, for example: when a person is at home, the network connection of the access control device 1 can be cut off by controlling the network connector 101 through the connection module 113. Alternatively, the first countdown is set by the mobile terminal device, such as: 10 to 30 seconds, the connection module 113 disconnects the network connection when the first countdown is over. Such a function can be applied to a scenario where the identification apparatus 1 automatically disconnects the internet within a period of time after the target object identity is successfully identified. In this embodiment, the network connection can be disconnected by the connection module 113 controlling the power supply 17 to stop supplying power to the network connector 101.

The sensing module 116 is used to control the sensor 14 to sense whether there is a visitor who needs to identify. The sensing module 116 wakes up the information collector 13 and the identification module 114 for the corresponding type of identification through the control module 115 when sensing that the visitor needs identification.

The recognition module 114 is configured to extract image features related to the identity of the visitor from the image information acquired by the image acquisition module 132 and voice features related to the identity of the visitor from the voice information acquired by the voice acquisition module 134, and implement a corresponding type of identity recognition operation according to a comparison between the image features and the voice features of the visitor and an identity feature template of a preset target object, and when the acquired image features and/or the voice features of the visitor are consistent with the identity feature template of the preset target object, the recognition module determines that the visitor passes the corresponding identity recognition operation.

It can be understood that the setting module 111 may also be configured to respectively preset matching degree thresholds corresponding to different types of identity recognition operations, and when the matching degree between the acquired image features and/or voice features of the visitor and the preset identity feature template of the target object meets the preset matching degree threshold and is not necessarily completely consistent, the reasonable fault tolerance rate may also be achieved through the identity recognition. The threshold matching degree corresponding to each identification operation is not specifically limited herein.

The recognition module 114 includes a plurality of recognition units 1140 to perform corresponding types of identification operations, such as: three identification units 1140 are included, a first identification unit 1141, a second identification unit 1142, and a third identification unit 1143.

As shown in fig. 3, in the present embodiment, the identification module 114 includes two identification units 1140, which are a first identification unit 1141 and a second identification unit 1142, respectively. The first identification unit 1141 is a human face identification unit. The face recognition unit 1141 obtains the sensed two-dimensional face data or three-dimensional face data of the visitor from the image acquisition module 132, compares the two-dimensional face data or three-dimensional face data with a preset two-dimensional face template and a preset three-dimensional face template of the target object, and determines that the visitor passes face recognition when the matching degree between the two-dimensional face data or three-dimensional face data of the visitor's face and the preset two-dimensional face data template or three-dimensional face data template of the target object respectively satisfies a preset matching degree threshold. Because the three-dimensional data of the human face is far richer than the two-dimensional image data of the human face, the counterfeiting difficulty and the cost are higher, and the reliability and the safety are higher when the three-dimensional data of the human face is adopted for identity recognition.

The second recognition unit 1142 is a voice password recognition unit. The voice password recognition unit 1142 performs semantic analysis on the voice information of the visitor acquired by the voice acquisition module 134 to obtain the content of the voice password spoken by the visitor, compares the analyzed content of the voice password with the voice password template preset by the target object, and determines that the visitor passes the voice password recognition when the matching degree between the content of the voice password spoken by the visitor and the voice password template preset by the target object meets the preset matching degree threshold.

As shown in fig. 4, in a modified embodiment, the identification module 114 may further include a third identification unit 1143. The third recognition unit 1143 is a voiceprint recognition unit. The voiceprint recognition unit 1143 analyzes the voice information of the visitor acquired by the voice acquisition module 134 to extract the voiceprint feature of the visitor, compares the voiceprint feature with a preset voiceprint feature template of the target object, and determines that the visitor passes voiceprint recognition when the matching degree of the voiceprint feature of the visitor and the preset voiceprint feature template of the target object meets a preset matching degree threshold. The voiceprint feature templates include, but are not limited to, the frequency spectrum, timbre, prosody, etc. of the sound. The voice print recognition mode comprises text correlation and text independence. When the voice print recognition is performed in a text-related manner, the voice print recognition unit 1143 controls the interaction device 15 through the interaction module 112 to prompt the visitor to read the preset text content, and the voice print recognition unit 1143 obtains the voice data of the visitor reading the preset text content through the voice collection module 134 to perform analysis and recognition. When the voiceprint recognition is performed in a text-independent manner, the voiceprint recognition unit 1143 directly obtains the voice data of the visitor through the voice collection module 134 for analysis and recognition, and at this time, the visitor can read the preset voice password and can speak any content at any time.

As shown in fig. 5, in a modified embodiment, the identification module 114 further includes a fourth identification unit 1144. The fourth recognition unit 1144 is a face motion recognition unit. The facial action recognition unit 1144 analyzes the two-dimensional and/or three-dimensional image of the visitor acquired by the image acquisition module 132 to determine whether the visitor is a living body. For example: the voice information of the visitor is acquired through the voice acquisition module 134, and simultaneously the three-dimensional information of the face of the visitor is acquired through the image acquisition module 132 and analyzed to identify whether the face of the visitor moves or not so as to judge whether the visitor is a living body or not. The visitor's facial movements include, but are not limited to, blinking and/or mouth opening and closing. The face action recognition can be used as auxiliary judgment, and if the face of the visitor is recognized to have blinking and/or mouth opening and closing actions during voice password recognition, the face of the visitor is judged to be a living body and can be identified through identity; and if the face of the visitor is identified to have no action during voice recognition, judging that the identified face of the visitor is a non-living body and does not pass identity recognition.

The control module 115 controls the corresponding recognition unit 1140 in the recognition module 114 and the corresponding recognition module in the information collector 13 to perform the identification operation according to the parameters, such as the type, the number, and the execution sequence of the identification operation set in the preset recognition mode, and then comprehensively determines whether the visitor is the target object according to the recognition result of the relevant identification operation and the execution sequence of the different identification operations in the preset recognition mode, and performs subsequent operations according to the determination result, for example: controls the opening and closing of the door lock 12.

It is understood that, when the identification device 1 is applied to the access device 3, the control module 115 may further set a second countdown, and if it is detected that the door 10 is not pushed after the door lock 12 is opened, the door lock 12 is locked again. The second countdown may be set to half a minute to one minute.

In the present embodiment, the recognition mode includes a voice password recognition operation and a face recognition operation. According to the set recognition mode, the control module 115 controls the voice password recognition unit 1142 and the voice acquisition module 134 to perform voice password recognition operation, controls the face recognition unit 1141 and the image acquisition module 132 to perform face recognition operation after the voice password recognition is performed, and completes identity recognition after the face recognition is performed. Because the power consumption of the voice acquisition module 134 for acquiring voice and the voice password recognition unit 1142 for performing semantic recognition is relatively low, and the power consumption generated by the operation of transmitting the structured light and subsequently analyzing the structured light spot pattern to acquire the three-dimensional data when acquiring the three-dimensional face data is significantly higher than that of semantic analysis, the voice password recognition operation with low power consumption is performed first, and then the face recognition operation with high power is started, so that the overall power consumption of the identity recognition device 1 can be effectively reduced.

In a variant, the recognition modes comprise a voice password recognition operation, a voiceprint recognition operation and a face recognition operation. According to the set identification mode, the control module 115 controls the voice password identification unit 1142 and the voice acquisition module 134 to execute the voice password identification operation, controls the voiceprint identification unit 1143 to execute the voiceprint identification operation after the voice password identification is performed, controls the face identification unit 1141 and the image acquisition module 132 to execute the face identification operation after the voiceprint identification operation is performed, and completes the identity identification after the face identification is performed. Similar to the above theory, the power consumption of the voiceprint recognition operation is lower than that of the face recognition operation, and the voiceprint feature has higher identity recognition degree than the voice password, so the addition of the voiceprint recognition operation between the voice password recognition operation and the face recognition operation is an implementation mode which improves the recognition security level and considers the lower power consumption.

In a variant, the recognition modes comprise a voice password recognition operation, a voiceprint recognition operation and a face recognition operation. According to the set recognition mode, the control module 115 controls the voice password recognition unit 1142 and the voice acquisition module 134 to perform voice password recognition operation, controls the voiceprint recognition unit 1143 to perform voiceprint recognition operation after the voice password recognition is performed, and controls the face recognition unit 1143 and the image acquisition module 132 to perform face recognition operation. The control module 115 completes the identification after both voiceprint recognition and face recognition pass. The power consumption can be reduced by executing the voice password identification first, and the speed of the identification can be increased by simultaneously executing the voiceprint identification and the face identification subsequently.

It is to be appreciated that in other embodiments, the recognition mode may include any one or more of a combination of the voice password recognition operation, a face recognition operation, a voiceprint recognition operation, a gesture recognition operation, a stature recognition operation, a body pose recognition operation, and an iris recognition operation. All possible execution sequences of the different types of identification operations in the identification mode are within the scope of the present application.

Referring to fig. 1 and fig. 6 together, the present application further provides an identification method for identifying the identity of a visitor by using the identification apparatus 1 of the first embodiment. The identification device 1 includes a sensor 14, an interactive device 15, an image capturing module 132, a sound capturing module 134, and a storage medium 18. The storage medium 18 is preset with an identity feature template of the target object for identifying the identity of the target object, including but not limited to three-dimensional face data of the target object, voiceprint features of the target object, and a password preset by the target object. The identity recognition method comprises the following steps:

step S101, sensing whether there is a visitor, and controlling the sensor 14 to sense whether there is a visitor who needs to perform identification. The sensor 14 may be, but not limited to, a sound sensor, a proximity sensor, and a touch sensor, and senses whether there is a visitor in front of the door 10 at a preset frequency. It is understood that this step S101 may also be omitted.

Step S102, reminding the visitor of identity recognition, and controlling the interaction device 15 to remind the visitor of identity recognition. The interaction device 15 includes, but is not limited to, a touch screen, a doorbell, and/or a speaker. For example: the touch screen and/or the loudspeaker remind the visitor of speaking the voice password, making a gesture, or looking at the appointed position by eyes, and the like, and the visitor can also be informed of reading out preset content for voiceprint recognition. Or the doorbell is provided with a lamp to flash and then matched with the display content of the touch screen to remind the visitor to press the doorbell to trigger and identify the connection with the target object. It is understood that step S102 may be omitted.

Step S103, acquiring a plurality of image characteristics and voice characteristics related to the identity of the visitor. As used herein, "plurality" means two or more. The image features and the voice features include, but are not limited to, three-dimensional face data, two-dimensional face data, voiceprint features, voice passwords, gestures, stature, body posture and irises, and are respectively acquired by corresponding collectors. For example: the image acquisition module 132 can be used to acquire the image information of the visitor and then extract the image features such as two-dimensional face data, three-dimensional face data, gestures, stature, body posture, iris and the like related to the identity of the visitor, or the voice acquisition module 134 can be used to acquire the voice information spoken by the visitor and then extract the voice password content and voiceprint features related to the identity of the visitor.

It is understood that the acquisition of different image features and voice features may be performed synchronously, or may be performed separately according to a preset sequence, depending on the specific setting content of the preset recognition mode.

In this embodiment, the voice information spoken by the visitor can be acquired by the voice collecting module 134. The voice information of the visitor is analyzed by utilizing relevant voice recognition and semantic analysis technology to obtain the voice password content spoken by the visitor.

The image information of the visitor can be acquired by the image capturing module 132. The image information can be used for analyzing and obtaining three-dimensional face data related to the identity of the visitor. According to the structured light pattern sensing principle, the three-dimensional face data of the visitor is obtained by projecting a preset infrared light spot dot matrix pattern on the face of the visitor and analyzing the change of the infrared light spot dot matrix pattern along with the three-dimensional shape of the face.

Referring to fig. 6 and 7, in the present embodiment, the step S103 includes the following sub-steps:

and step S1031, acquiring the voice information of the visitor and carrying out voice password identification.

Step S1032 acquires image information of the visitor to perform face recognition.

Wherein, the step S1031 is executed first, and the step S1032 is executed later.

Referring to fig. 6 and 8, in another embodiment, the step S103 further includes the sub-steps of:

step S1033, voice information of the visitor is acquired to perform voiceprint recognition. The voice data of the visitor is acquired through the voice acquisition module 134 and analyzed to extract the voiceprint characteristics of the visitor. The voiceprint features include, but are not limited to, the spectrum, timbre, prosody, etc. of the sound.

Step S1031 is executed first, and step S1032 and step S1033 may be executed in a sequential order or may be executed synchronously.

Referring to fig. 7 and 9 together, in another embodiment, the step S1032 is performed first, and the step S1031 is performed later.

And step S104, carrying out combined recognition on various image features and voice features of the visitor according to a preset execution sequence. As used herein, "plurality" means two or more. The relationship between the identity recognition operations corresponding to the different types of image features and the voice features may be caused sequentially, for example: the first identification operation is passed first and then the second identification operation is carried out. The identification operations corresponding to the different types of image features and the voice features may also be synchronously performed, for example: the first identification operation and the second identification operation are carried out synchronously, and the next step can be carried out only after the first identification operation and the second identification operation pass.

Referring to fig. 6 and 7, in the present embodiment, the step S104 includes the following sub-steps:

step S1041, comparing the content of the voice password spoken by the visitor with a password preset by the target object, and if the matching degree between the content of the voice password spoken by the visitor and the password preset by the target object meets a preset threshold, determining that the visitor passes the voice password identification. If the matching degree of the voice password content spoken by the visitor and the password preset by the target object is lower than the preset threshold value, the voice password identification is judged not to pass.

It will be appreciated that the visitor may also be prompted by the interactive device 15 that the voice password is incorrect and asked to re-perform voice password recognition when the contents of the voice password spoken by the visitor do not match the password preset for the target object.

It will be appreciated that a plurality of target object passwords may be preset, such as: please open the door, sesame open the door, or a string of digital codes "33143", etc. Different matching thresholds may also be set for different target object passwords, respectively, for example: for the target object password of "please open the door", the target object password can be considered to be matched as long as the "open the door" is said; the target object password for the numeric password requires that the numbers spoken by the visitor match exactly the same as the preset numeric password.

Step S1042, performing face recognition according to the two-dimensional face data or the three-dimensional face data of the visitor. Specifically, the acquired two-dimensional face data or three-dimensional face data of the visitor is compared with a preset two-dimensional face data template or three-dimensional face data template of the target object, if the matching degree meets a preset threshold value, it is judged that the visitor passes face recognition, and if the matching degree is lower than the preset threshold value, it is judged that the visitor does not pass face recognition.

It will be appreciated that the visitor may also be prompted by the interactive device 15 that face recognition is not passed when there is no match and the visitor may be asked to re-face recognition.

Wherein, the step S1041 is executed first, and the step S1042 is executed later.

Referring to fig. 6 and 8, in another embodiment, the step S104 may further include the sub-steps of:

in step S1403, voiceprint recognition is performed on the visitor. The voiceprint recognition unit 1143 compares the obtained voiceprint feature of the visitor with a preset voiceprint feature of the target object, and determines that the visitor passes voiceprint recognition when the matching degree of the voiceprint feature of the visitor and the preset voiceprint feature of the target object meets a preset threshold.

The step S1041 is executed first, and the step S1042 and the step S1043 may be executed in sequence or may be executed synchronously.

Referring to fig. 5 and fig. 8, in another embodiment, the step S1042 is performed first, and the step S1041 is performed later.

Step S105, if the visitor is identified by the combination of the multiple image features and the voice features, determining the identity of the visitor as a target object, and then performing an operation corresponding to the determined identity authority, for example: open the door, start the car, etc.

It is understood that the control module 115 may send a notification message to the target object if the visitor does not recognize the combination. Under the condition that the network connection is not disconnected, the target object can also check the image information of the visitor acquired by the image acquisition module 132 through the connection module 113, and can also directly open the door through the control module 115 after confirming the identity of the visitor.

Compared with the existing identity recognition device, the identity recognition device 1 and the identity recognition method thereof provided by the application open the door by collecting various image characteristics and voice characteristics of a visitor to carry out identity recognition, and a target object does not need to carry a key or an access card. And different image characteristics and voice characteristics are set to perform combined recognition according to a reasonable execution sequence, so that safety and convenience of identity recognition can be considered.

As shown in fig. 10, the second embodiment of the present application further provides an identification apparatus 2, which has a structure substantially the same as that of the identification apparatus 1 of the first embodiment, and the main difference is that the identification apparatus 2 further includes an alarm module 216.

The setting module 211 may also be configured to set an alarm voice password, alarm information, and alarm object data. The alarm voice password may be a sentence or a string of numbers. The alarm voice password is used for giving an alarm in a real-time and sound-color manner when a target object is stressed by a malicious third party, and simultaneously executing operations of corresponding authorities after identity confirmation, such as: opening the door, and the like, so as to temporarily stabilize a malicious third person and avoid stimulating the malicious third person to make a behavior of damaging a target object. The alarm information may include, but is not limited to, address information of the identification device 2, personal identification data of the target object, pre-edited alarm text or voice information, and the like. The alarm object data is a receiving terminal of the sending object of the alarm information, and includes but is not limited to an alarm terminal of a community property, an alarm terminal of a local police station, a mobile terminal of an emergency contact person and the like.

The voice password recognition unit 2142 compares the analyzed voice password content spoken by the visitor with a preset alarm voice password, and determines that the visitor is under duress when the matching degree of the voice password content spoken by the visitor and the preset alarm voice password satisfies a preset threshold.

The alarm module 216 is configured to send the alarm information to corresponding alarm objects according to preset alarm object data after the voice password recognition unit 2142 recognizes the alarm voice password.

The control module 215 executes the operation of the corresponding authority after the identity is confirmed after the phonetic password recognition unit 2142 recognizes the alarm phonetic password, such as: and opening the door, and the like, and simultaneously controlling the image acquisition module 232 and the sound acquisition module 234 to record the field condition as evidence, and controlling the connection module 213 to transmit the field condition recorded by the image acquisition module 232 and the sound acquisition module 234 to the receiving terminal of the alarm object through network connection.

Referring to fig. 10 and fig. 11, the present application further provides an identification method for identifying the identity of a visitor by using the identification apparatus 2 of the second embodiment. The identification device 2 includes a sensor 24, an interactive device 25, an image capturing module 232, a sound capturing module 234, and a storage medium 28. The storage medium 28 is preset with an alarm voice password, alarm information and alarm object data. The identity recognition method comprises the following steps:

step S201, controlling the sensor 24 to sense whether a visitor approaches. The sensor 24 may be, but is not limited to, a proximity sensor and a touch sensor, and senses whether there is a visitor at a preset frequency. It is understood that step S201 may be omitted.

Step S202, controlling the interactive device 15 to remind the visitor of identity recognition. The interactive device 25 includes, but is not limited to, a touch screen and/or a speaker. And displaying and reminding a visitor to identify the identity through the touch screen. For example: the touch screen and/or the loudspeaker remind the visitor of speaking the voice password, making a gesture, or looking at the appointed position by eyes, and the like, and the visitor can also be informed of reading out preset content for voiceprint recognition. It is understood that step S202 may be omitted.

Step S203, the voice password content spoken by the visitor is acquired. Specifically, the voice information spoken by the visitor is obtained through the voice collecting module 234. The voice information of the visitor collected by the voice collection module 234 is analyzed by using the relevant voice recognition and semantic analysis techniques to obtain the content of the voice password spoken by the visitor.

Step S204, identifying whether the voice password content spoken by the visitor is an alarm voice password. And comparing the obtained voice password content spoken by the visitor with a preset alarm voice password, and judging that the visitor is forced when the matching degree of the voice password content spoken by the visitor and the preset alarm voice password meets a preset threshold value.

And S205, after recognizing that the visitor speaks as the alarm voice password, respectively sending the alarm information to corresponding alarm objects according to preset alarm object data.

In step S206, after recognizing that the visitor said the alarm voice password, the image capturing module 232 and the sound capturing module 234 are activated to record the scene conditions, including but not limited to recording the scene image information through the image capturing module 232 and recording the scene sound information through the sound capturing module 234.

Step S207, after recognizing that the visitor said as the alarm voice password, executing the operation of the corresponding authority after the identity recognition, such as: open the door, start the car, etc.

And step S208, after recognizing that the visitor speaks as an alarm voice password, sending the field conditions recorded by the image acquisition module 232 and the sound acquisition module 234 to a receiving terminal of an alarm object. And if the network is disconnected, reestablishing network connection and then sending the recorded field condition to an alarm object.

In the description herein, references to the description of the terms "one embodiment," "certain embodiments," "an illustrative embodiment," "an example," "a specific example," or "some examples" or the like mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present application. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and the scope of the preferred embodiments of the present application includes other implementations in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present application.

The logic and/or steps represented in the flowcharts or otherwise described herein, such as an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processing module-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a random access storage media (RAM), a read-only storage media (ROM), an erasable programmable read-only storage media (EPROM or flash storage media), an optical fiber device, and a portable compact disc read-only memory (CDROM). Further, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer storage medium.

It should be understood that portions of the embodiments of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in the storage medium 18 and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.

The above description is only for the purpose of illustrating the preferred embodiments of the present application and is not to be construed as limiting the present application, and any modifications, equivalents and improvements made within the spirit and principle of the present application are intended to be included within the scope of the present application.

Claims

1. An identity recognition method, characterized in that: the identity recognition method is used for recognizing the identity of a visitor according to a preset voice password template, a two-dimensional face data template and a three-dimensional face data template which are related to the identity of a target object, and comprises the following steps:

acquiring voice information of a visitor;

extracting the voice password spoken by the visitor from the acquired voice information of the visitor, and comparing the voice password with a voice password template of a preset target object to carry out password identification;

acquiring image information of a visitor after the password identification is passed;

extracting two-dimensional face data or three-dimensional face data of the visitor from the acquired image information of the visitor, and comparing the two-dimensional face data or the three-dimensional face data with a preset two-dimensional face data template or a preset three-dimensional face data template of a target object to perform face recognition; and

and after the face recognition is passed, the identity of the visitor is confirmed as a target object.

2. The identity recognition method of claim 1, wherein: before the voice information of the visitor is obtained, whether the visitor approaches is sensed, and after the visitor approaches is sensed, the voice information of the visitor is obtained.

3. The identity recognition method of claim 2, wherein: the method also comprises the following steps after the approach of a visitor is sensed: and reminding the visitor to identify the identity by displaying a reminder or making a sound.

4. The identity recognition method of claim 1, wherein: an alarm password, alarm information and alarm object data are preset, and the identity identification method further comprises the following steps:

5. The identity recognition method of claim 4, wherein: after the voice password of the visitor is identified as the alarm password in the password identification, the method also comprises the following steps: and recording the image information and the sound information of the scene.

6. The identity recognition method of claim 1, wherein: when password recognition is performed, a two-dimensional and/or three-dimensional image of the face of a visitor is acquired to perform face motion recognition, whether the recognized visitor is a living body is determined, and password recognition is allowed only when the recognized visitor is determined to be a living body.

7. The identity recognition method of claim 1, wherein: three-dimensional face data of the visitor is acquired according to a structured light pattern sensing principle, a binocular vision sensing principle and/or a time-of-flight sensing principle.

8. The identity recognition method of claim 1, wherein: and after confirming that the visitor identity is the target object, granting the identity authority corresponding to the target object to execute corresponding operation.

9. An identification method as claimed in any one of claims 1 to 8, characterized in that: a voiceprint feature template of the target object is also preset, and the identity recognition method further comprises the following steps:

10. A storage medium storing program code executable by one or more processors, characterized in that: the program is executable by the processor to implement the identification method as claimed in claims 1-9.