CN114078273A - Early warning method, device, equipment and storage medium - Google Patents

Early warning method, device, equipment and storage medium Download PDF

Info

Publication number
CN114078273A
CN114078273A CN202111190167.5A CN202111190167A CN114078273A CN 114078273 A CN114078273 A CN 114078273A CN 202111190167 A CN202111190167 A CN 202111190167A CN 114078273 A CN114078273 A CN 114078273A
Authority
CN
China
Prior art keywords
information
early warning
image
target image
current target
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111190167.5A
Other languages
Chinese (zh)
Inventor
邓先付
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Pilot Guards Safety Technology Co ltd
Original Assignee
Shenzhen Pilot Guards Safety Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shenzhen Pilot Guards Safety Technology Co ltd filed Critical Shenzhen Pilot Guards Safety Technology Co ltd
Priority to CN202111190167.5A priority Critical patent/CN114078273A/en
Publication of CN114078273A publication Critical patent/CN114078273A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B13/00—Burglar, theft or intruder alarms
    • G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
    • G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
    • G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
    • G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B13/00—Burglar, theft or intruder alarms
    • G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
    • G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
    • G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
    • G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
    • G08B13/19617—Surveillance camera constructional details
    • G08B13/1963—Arrangements allowing camera rotation to change view, e.g. pivoting camera, pan-tilt and zoom [PTZ]
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
    • G10L17/00—Speaker identification or verification
    • G10L17/22—Interactive procedures; Man-machine interfaces

Abstract

The invention discloses an early warning method, an early warning device, early warning equipment and a storage medium, and belongs to the technical field of Internet of things. The invention can judge whether the door-opening person is a stranger or not by collecting the current target image when the door is opened and extracting the face information from the current target image, can also carry out voiceprint recognition when the comparison fails, further confirms whether illegal invasion behaviors exist or not, and sends out early warning information to a user if the face information comparison and the voiceprint recognition comparison fail. The technical problem that property loss is easily caused because early warning cannot be timely carried out on a user when the user is invaded by an illegal molecule is avoided.

Description

Early warning method, device, equipment and storage medium
Technical Field
The invention relates to the technical field of Internet of things, in particular to an early warning method, an early warning device, early warning equipment and a storage medium.
Background
People in the twenty-first century have increasingly high living standard and pursuit of life, and meanwhile great attention is paid to protecting safety, comfort and convenience in family life. The concept of smart home is becoming mainstream due to the impact of internet technology, and the smart electronic door for protecting the privacy of the home is also produced, but the existing electronic door cannot give an early warning to the user in time when being invaded by illegal molecules, and property loss is easily caused.
The above is only for the purpose of assisting understanding of the technical aspects of the present invention, and does not represent an admission that the above is prior art.
Disclosure of Invention
The invention mainly aims to provide an early warning method, an early warning device, early warning equipment and a storage medium, and aims to solve the technical problem that in the prior art, when an illegal molecule invades, early warning cannot be timely carried out on a user, and property loss is easily caused.
In order to achieve the above object, the present invention provides an early warning method, comprising the steps of:
when a door opening signal is detected, acquiring a current target image;
carrying out face recognition through a preset face recognition model based on the current target image to obtain face information;
when the face information fails to be compared with the prestored target information, reminding a user of voiceprint recognition;
and when the voiceprint signal is not received or the voiceprint identification fails, generating early warning information and sending the early warning information to a user alarm host for early warning, wherein the user alarm host is also used for managing the networking and running functions of each electronic device.
Optionally, the performing face recognition through a preset face recognition model based on the current target image to obtain face information includes:
carrying out human shape detection on the current target image through a preset human shape detection model to obtain a human image;
extracting information of five sense organs of the character image, and determining whether an effective target exists according to the information of the five sense organs;
and if the effective target exists, carrying out face recognition through a preset face recognition model based on the current target image to obtain face information.
Optionally, after extracting information of five sense organs of the person image and determining whether a valid target exists according to the information of the five sense organs, the method further includes:
if no effective target exists, extracting the information of the figure image to obtain the body shape information;
comparing the body shape information with target body shape information in prestored target information;
and when the comparison between the body shape information and the target body shape information fails, reminding the user of carrying out voiceprint recognition.
Optionally, the extracting the information of the five sense organs of the person image and determining whether a valid target exists according to the information of the five sense organs includes:
extracting the information of the five sense organs of the figure image;
when the eye information and the mouth information coexist in the five sense organs information, it is determined that a valid target exists.
Optionally, after the receiving of the voiceprint signal is not performed or the voiceprint recognition fails, generating early warning information and giving an early warning to the user, the method further includes:
carrying out visual angle tracking on the current target image through a preset visual angle tracking model to obtain a visual angle tracking result;
and controlling the camera to rotate according to the visual angle tracking result so that the camera collects a target image.
Optionally, the tracking a view angle of the current target image through a preset view angle tracking model to obtain a view angle tracking result includes:
performing feature extraction on the current target image to obtain feature extraction information;
performing feature annotation on the current target image based on the feature extraction information to obtain a figure image annotation frame;
and tracking the visual angle through a preset visual angle tracking model according to the figure image labeling frame to obtain a visual angle tracking result.
Optionally, the tracking a view angle through a preset view angle tracking model according to the person image labeling frame further includes, before obtaining a view angle tracking result:
acquiring a figure image labeling frame sample and a view tracking result sample corresponding to the figure image labeling frame sample;
and performing model training on the figure image labeling frame sample and the view tracking result through an initial neural network to obtain a preset view tracking model.
In addition, in order to achieve the above object, the present invention further provides an early warning device, including:
the acquisition module is used for acquiring a current target image when a door opening signal is detected;
the detection module is used for carrying out face recognition through a preset face recognition model based on the current target image to obtain face information;
the recognition module is used for reminding a user of voiceprint recognition when the comparison between the face information and the prestored target information fails;
and the early warning module is used for generating early warning information and sending the early warning information to the user alarm host for early warning when the voiceprint signal is not received or voiceprint identification fails, and the user alarm host is also used for managing networking and operating functions of each electronic device.
In addition, in order to achieve the above object, the present invention further provides an early warning device, including: a memory, a processor and a pre-warning program stored on the memory and executable on the processor, the pre-warning program being configured to implement the steps of the pre-warning method as described above.
In addition, to achieve the above object, the present invention further provides a storage medium, on which an early warning program is stored, and the early warning program implements the steps of the early warning method as described above when executed by a processor.
The method comprises the steps of acquiring a current target image when a door opening signal is detected, carrying out face recognition through a preset face recognition model based on the current target image to obtain face information, reminding a user of voiceprint recognition when the face information fails to be compared with prestored target information, and generating early warning information and giving an early warning to the user when the voiceprint signal is not received or the voiceprint recognition fails. Compared with the prior art, the method and the device have the advantages that the current target image when the door is opened is collected, the face information is extracted from the current target image, whether the door-opened person is a stranger or not can be judged by comparing the face information with the pre-stored target information, voiceprint recognition can be carried out when the comparison fails, whether illegal intrusion behaviors exist or not is further confirmed, and if the face information comparison and the voiceprint recognition comparison fail, early warning information is sent to a user. The technical problem that property loss is easily caused because early warning cannot be timely carried out on a user when the user is invaded by an illegal molecule is avoided.
Drawings
Fig. 1 is a schematic structural diagram of an early warning device of a hardware operating environment according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of a first embodiment of the warning method according to the present invention;
FIG. 3 is a flowchart illustrating a second embodiment of an early warning method according to the present invention;
FIG. 4 is a flowchart illustrating a third embodiment of an early warning method according to the present invention;
fig. 5 is a block diagram of the early warning device according to the first embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 1, fig. 1 is a schematic structural diagram of an early warning device in a hardware operating environment according to an embodiment of the present invention.
As shown in fig. 1, the early warning apparatus may include: a processor 1001, such as a Central Processing Unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Wherein a communication bus 1002 is used to enable connective communication between these components. The user interface 1003 may include a Display screen (Display), an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a Wireless interface (e.g., a Wireless-Fidelity (Wi-Fi) interface). The Memory 1005 may be a Random Access Memory (RAM) Memory, or may be a Non-Volatile Memory (NVM), such as a disk Memory. The memory 1005 may alternatively be a storage device separate from the processor 1001.
Those skilled in the art will appreciate that the configuration shown in fig. 1 is not intended to be limiting, and may include more or fewer components than those shown, or some combination of components, or a different arrangement of components.
As shown in fig. 1, a memory 1005, which is a storage medium, may include therein an operating system, a network communication module, a user interface module, and an early warning program.
In the warning device shown in fig. 1, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the warning device of the present invention may be disposed in the warning device, and the warning device calls the warning program stored in the memory 1005 through the processor 1001 and executes the warning method provided by the embodiment of the present invention.
An embodiment of the present invention provides an early warning method, and referring to fig. 2, fig. 2 is a schematic flow diagram of a first embodiment of an early warning method according to the present invention.
In this embodiment, the early warning method includes the following steps:
step S10: and when a door opening signal is detected, acquiring a current target image.
It should be noted that the execution subject of this embodiment is an early warning device, where the early warning device may be a server, a personal computer, or other early warning devices that can implement the same or similar functions.
It will be appreciated that the door open signal may be a signal generated by a sensor mounted at the door lock detecting that the door lock is open.
It should be noted that the current target image may be an image in front of a door acquired by an image acquisition device when the door opening signal is received, and the image acquisition device may be a camera, an image acquisition card, or the like.
In specific implementation, when the door is opened from the outside of the door, the sensor can generate a door opening signal and send the door opening signal to the server, and the server controls the camera to acquire the current image outside the door after receiving the door opening signal.
Step S20: and carrying out face recognition through a preset face recognition model based on the current target image to obtain face information.
It should be noted that, a preset face recognition model is used to recognize whether face information exists in a current target image, extract the face information, and after the current target image passes through the preset face recognition model, a face labeling frame may be obtained, and face information in the current target image is obtained according to the face labeling frame, where the face information may be face feature information, for example: eye information, mouth information, and the like, and may also be a contact relationship of one or more of them.
Further, to obtain the face detection model, before step S20, the method further includes:
acquiring a face image sample and a current target image sample;
and performing model training through an initial neural network model according to the face image sample and the current target image sample to obtain a preset face detection model.
It is understood that the initial neural network model may be a Local Binary Pattern (LBP) model, and may also be other neural network models with a face detection function, and the embodiment is not limited in particular.
Step S30: and when the comparison between the face information and the pre-stored target information fails, reminding a user of voiceprint recognition.
It should be noted that the pre-stored target information may be security information that is preset and stored in the server by the user, the pre-stored target information may be added or deleted by the user, and operations such as prompt for specific targets may also be performed, the security information may be face information, body shape information, voiceprint information, and the like, and the security information may be security information of the user, and may also be a relative, a friend, and the like, which is not limited in this embodiment.
It can be understood that the voiceprint recognition can be to recognize a voice signal sent by a user and determine whether the voice signal conforms to the voiceprint information in the preset target information, and the voiceprint recognition mode includes two types, for example: 1. identifying the voiceprint characteristics; 2. performing voice recognition; in this embodiment, one or two of them may be combined as the voiceprint recognition method of this embodiment.
The voiceprint feature identification is to judge the feature information of the current sound signal through the technologies of frequency analysis, tone analysis, atlas analysis and the like on the current sound signal sent by a user, generally, the voiceprint feature information of each person is different, and whether the user corresponding to the current sound signal is the user corresponding to the pre-stored target information can be judged by comparing the processed voiceprint feature information with the voiceprint information in the preset target information.
Secondly, the voice recognition can be to perform voice recognition on a current voice signal sent by a user, convert the current voice signal into readable characters, and judge whether the user corresponding to the current voice signal is the user corresponding to the pre-stored target information by comparing whether the characters obtained by conversion correspond to the preset characters.
It should be noted that, in the process of performing face recognition on the current target image acquired by the camera, multiple pieces of face information may be recognized in the current target image, that is, multiple persons exist in the current target image, and at this time, as long as one piece of face information is successfully compared with preset target information, the early warning information may not be triggered.
In specific implementation, when a user returns home with a friend, after face recognition is performed on a current target image, a plurality of pieces of face information are obtained, which are respectively: A. b, C, however, the pre-stored target information has face information corresponding to a and does not have safety information corresponding to B, C, and at this time, because the face information of a in the obtained plurality of face information is successfully compared with the pre-stored target information, the pre-warning information is not triggered.
Step S40: and when the voiceprint signal is not received or the voiceprint identification fails, generating early warning information and sending the early warning information to a user alarm host for early warning, wherein the user alarm host is also used for managing the networking and running functions of each electronic device.
It should be noted that the early warning information may be operations with a reminding function, such as text, voice reminding, vibration, etc., or other early warning information with the same or similar function, and the early warning information is sent to a user terminal, which may be a mobile phone, a personal computer, etc., the embodiment is not particularly limited,
it can be understood that when the early warning information is generated, early warning can be performed on the current target to remind a user that the early warning information is received, so that the current target exits from a room, and loss is avoided.
In a specific implementation, since the current user worries about the failure of voiceprint recognition and does not make a sound, an interval time may be set, within which no sound signal is received, i.e. it is considered that voiceprint recognition fails, and the interval time may be set by the user, for example: and 1min, and the like, so that when the voiceprint signal is not received or the voiceprint identification fails, early warning information is generated and sent to the user alarm host for early warning.
In this embodiment, when a door opening signal is detected, a current target image is acquired, face recognition is performed through a preset face recognition model based on the current target image, face information is acquired, when the face information fails to be compared with pre-stored target information, a user is reminded of voiceprint recognition, and when the voiceprint signal is not received or the voiceprint recognition fails, early warning information is generated and early warning is performed on the user. Compared with the prior art, the method and the device have the advantages that the current target image when the door is opened is collected, the face information is extracted from the current target image, whether the door-opening person is a stranger or not can be judged by comparing the face information with the pre-stored target information, voiceprint recognition can be carried out when the comparison fails, whether illegal intrusion behaviors exist or not is further confirmed, and if the face information comparison and the voiceprint recognition comparison fail, early warning information is sent to a user. The technical problem that property loss is easily caused because early warning cannot be timely carried out on a user when the user is invaded by an illegal molecule is avoided.
Referring to fig. 3, fig. 3 is a flowchart illustrating a second embodiment of an early warning method according to the present invention.
Based on the first embodiment, in this embodiment, the step S20 includes:
step S201: and carrying out human shape detection on the current target image through a preset human shape detection model to obtain a human figure image.
It should be noted that the preset human shape detection model is used for extracting the human image in the current target image, so as to subsequently detect the information of the five sense organs. In order to obtain a face detection model, a face image sample and a current target image sample need to be obtained, model training is performed through an initial neural network model according to the face image sample and the current target image sample, and a preset face detection model is obtained.
It is understood that the initial neural network model may be a target detection network (YOLO), and may also be other neural network models with a face detection function, and the embodiment is not particularly limited.
Step S202: and extracting information of five sense organs of the character image, and determining whether a valid target exists according to the information of the five sense organs.
It can be understood that the information of the five sense organs includes eye information, mouth information, ear information, nose information, eyebrow information, etc., and whether the target person in the current target image blocks the face can be determined according to the information of the five sense organs of the person image, and the target person which does not block the face can be marked as an effective target.
Further, in order to obtain the number of effective targets in the current target image, the step S202 includes:
extracting the information of the five sense organs of the figure image;
when the eye information and the mouth information coexist in the five sense organs information, it is determined that a valid target exists.
It should be noted that, when the eye information and the mouth information are detected simultaneously according to the facial feature information of the same person image detected by the preset human shape detection model, it can be determined that the corresponding person image does not block the face, and the person image can be regarded as an effective target.
Step S203: and if the effective target exists, carrying out face recognition through a preset face recognition model based on the current target image to obtain face information.
It should be noted that, a preset face recognition model is used to recognize whether face information exists in a current target image, extract the face information, and after the current target image passes through the preset face recognition model, a face labeling frame may be obtained, and face information in the current target image is obtained according to the face labeling frame, where the face information may be face feature information, for example: eye information, mouth information, and the like, and may also be a contact relationship of one or more of them.
Further, since there may be no valid target in the current target image when performing valid target identification, after step S203, the method further includes:
if no effective target exists, extracting the information of the figure image to obtain the body shape information;
comparing the body shape information with target body shape information in prestored target information;
and when the comparison between the body shape information and the target body shape information fails, reminding the user of carrying out voiceprint recognition.
It should be noted that the shape information may be the shape contour information, the shape height information, etc. of the person image, and since the shape information of each person has difference, when the effective target cannot be identified, the identity of the person image may also be determined through the shape information.
In the specific implementation, because the face detection only detects the face information of the valid target, that is, the identity of the invalid target can be determined by the body shape information, the voiceprint recognition and the like, when both of them pass, no early warning information is sent out, for example: a, B, C, D four persons exist in the current target image detected by the preset human shape detection model, however, A, B is an effective target, C, D is an invalid target, when the four persons are subjected to identity recognition, the identity recognition can be carried out according to face information of A, B, the identity recognition can be carried out according to body shape information of C, D, when the identity recognition of the four persons fails, the four persons can be reminded of carrying out voiceprint recognition, and as long as one person succeeds in the identity recognition, the early warning information can not be generated.
In the embodiment, when a door opening signal is detected, a current target image is obtained, human shape detection is performed on the current target image through a preset human shape detection model, a figure image is obtained, facial features information of the figure image is extracted, whether an effective target exists is determined according to the facial features information, if the effective target exists, face recognition is performed through a preset face recognition model based on the current target image, face information is obtained, when the face information fails to be compared with the preset target information, a user is reminded of voiceprint recognition, and when a voiceprint signal is not received or the voiceprint recognition fails, early warning information is generated and early warning is performed on the user. Compared with the prior art, the method and the device have the advantages that the current target image when the door is opened is collected, the character image is extracted from the current target image, the information of the five sense organs is extracted from the character image, and whether the face is shielded by the current character image or not can be judged from the information of the five sense organs. If the door is not blocked, whether the door-opening person is a stranger or not can be judged by comparing the face information with the prestored target information, voiceprint recognition can be carried out when the comparison fails, whether illegal intrusion behaviors exist or not is further confirmed, and if the face information comparison and the voiceprint recognition comparison fail, early warning information is sent to the user. The technical problem that property loss is easily caused because early warning cannot be timely carried out on a user when the user is invaded by an illegal molecule is avoided.
Referring to fig. 4, fig. 4 is a flowchart illustrating a warning method according to a third embodiment of the present invention.
Based on the second embodiment, in this embodiment, after step S40, the method further includes:
step S50: and tracking the view angle of the current target image through a preset view angle tracking model to obtain a view angle tracking result.
It should be noted that the preset view tracking model is used for performing view tracking on the person image to obtain a view tracking result, where the view tracking result may be a result of predicting movement of the person image, and may also be a result of moving a mark frame of the person image, contour information, and the like, and this embodiment is not limited in particular.
In specific implementation, when voiceprint recognition fails or voiceprint information is not received, in order to prevent a target which is failed in recognition from continuously invading, feature extraction may be performed on the current target image to obtain feature extraction information, feature labeling is performed on the current target image based on the feature extraction information to obtain a character image labeling frame, and perspective tracking is performed through a preset perspective tracking model according to the character image labeling frame to obtain a perspective tracking result.
It should be noted that the feature extraction information may be the shape information of the person image, the clothing information, or other feature information that can distinguish the person, and the identity information of different persons may be determined by the feature information between different persons, so as to perform perspective tracking.
It can be understood that the character image labeling frame may be a labeling frame selected according to the feature information of the character image, in order to label the character image more accurately, an Intersection Over Union (IOU) threshold may be set, the IOU threshold is used to determine whether the information such as the position and size of the labeling frame is too large in deviation from the actual pointer region, if the Intersection ratio of the labeling frame is greater than or equal to the IOU threshold, the labeling frame is considered to be valid, and if the Intersection ratio of the labeling frame is less than the IOU threshold, the labeling frame is considered to be invalid.
It should be noted that the preset view tracking model is used for marking the annotated person image, and the direction of the image capturing device is moved by recording the moving direction of the marking frame and the position of the marking frame, so as to obtain a view tracking result.
Further, in order to obtain the preset view tracking model, before the step S50, the method further includes:
acquiring a figure image labeling frame sample and a view tracking result sample corresponding to the figure image labeling frame sample;
and performing model training on the figure image labeling frame sample and the view tracking result through an initial neural network to obtain a preset view tracking model.
It should be noted that the sample of the character image annotation box may be an image sample input by a user, or may be downloaded from a network, and this embodiment is not limited in particular.
It is understood that the initial neural network may be a neural network based on a mean shift algorithm, or may be other neural networks having the same or similar functions, and the embodiment is not particularly limited.
Step S60: and controlling the camera to rotate according to the visual angle tracking result so that the camera collects a target image.
It should be noted that the target image may be an image of a target person who still enters a room after the authentication fails, and in actual operation, the camera will rotate along with the movement of the target person to clearly acquire the target image.
In the embodiment, when a door opening signal is detected, a current target image is obtained, human shape detection is performed on the current target image through a preset human shape detection model to obtain a character image, information of five sense organs of the character image is extracted, whether an effective target exists or not is determined according to the information of the five sense organs, and if the effective target exists, performing face recognition through a preset face recognition model based on the current target image to obtain face information, when the comparison between the face information and the pre-stored target information fails, the user is reminded to perform voiceprint recognition, when a voiceprint signal is not received or the voiceprint recognition fails, generating early warning information and giving early warning to a user, tracking the view angle of the current target image through a preset view angle tracking model to obtain a view angle tracking result, and controlling the camera to rotate according to the visual angle tracking result so that the camera collects a target image. Compared with the prior art, the method and the device have the advantages that the current target image when the door is opened is collected, the character image is extracted from the current target image, the information of the five sense organs is extracted from the character image, and whether the face is shielded by the current character image or not can be judged from the information of the five sense organs. If the data is not shielded, whether the door-opening person is a stranger or not can be judged by comparing the face information with the prestored target information, voiceprint recognition can be carried out when the comparison fails, whether illegal invasion behaviors exist or not is further confirmed, if the face information comparison fails to compare with the voiceprint recognition, early warning information is sent to a user, visual angle tracking is continuously carried out on the target image, and evidence is left. The technical problem that property loss is easily caused because early warning cannot be timely carried out on a user when the user is invaded by an illegal molecule is avoided.
In addition, an embodiment of the present invention further provides a storage medium, where an early warning program is stored on the storage medium, and the early warning program, when executed by a processor, implements the steps of the early warning method described above.
Since the storage medium adopts all the technical solutions of all the embodiments, at least all the advantages brought by the technical solutions of the embodiments are available, and are not described in detail herein.
Referring to fig. 5, fig. 5 is a block diagram of the early warning device according to the first embodiment of the present invention.
As shown in fig. 5, the early warning apparatus provided in the embodiment of the present invention includes:
the acquiring module 10 is configured to acquire a current target image when a door opening signal is detected.
And the detection module 20 is configured to perform face recognition through a preset face recognition model based on the current target image to obtain face information.
And the recognition module 30 is configured to remind the user of voiceprint recognition when the comparison between the face information and the pre-stored target information fails.
And the early warning module 40 is used for generating early warning information and giving early warning to the user when the voiceprint signal is not received or the voiceprint identification fails.
In this embodiment, when a door opening signal is detected, a current target image is acquired, face recognition is performed through a preset face recognition model based on the current target image, face information is acquired, when the face information fails to be compared with pre-stored target information, a user is reminded of voiceprint recognition, and when the voiceprint signal is not received or the voiceprint recognition fails, early warning information is generated and early warning is performed on the user. Compared with the prior art, the method and the device have the advantages that the current target image when the door is opened is collected, the face information is extracted from the current target image, whether the door-opening person is a stranger or not can be judged by comparing the face information with the pre-stored target information, voiceprint recognition can be carried out when the comparison fails, whether illegal intrusion behaviors exist or not is further confirmed, and if the face information comparison and the voiceprint recognition comparison fail, early warning information is sent to a user. The technical problem that property loss is easily caused because early warning cannot be timely carried out on a user when the user is invaded by an illegal molecule is avoided.
In an embodiment, the detecting module 20 is further configured to perform human shape detection on the current target image through a preset human shape detection model to obtain a human image; extracting information of five sense organs of the character image, and determining whether an effective target exists according to the information of the five sense organs; and if the effective target exists, carrying out face recognition through a preset face recognition model based on the current target image to obtain face information.
In an embodiment, the detection module 20 is further configured to extract information of the person image to obtain body shape information if there is no valid target; comparing the body shape information with target body shape information in prestored target information; and when the comparison between the body shape information and the target body shape information fails, reminding the user of carrying out voiceprint recognition.
In an embodiment, the detection module 20 is further configured to extract information of five sense organs of the person image; when the eye information and the mouth information coexist in the five sense organs information, it is determined that a valid target exists.
In an embodiment, the early warning module 40 is further configured to perform view tracking on the current target image through a preset view tracking model to obtain a view tracking result; and controlling the camera to rotate according to the visual angle tracking result so that the camera collects a target image.
In an embodiment, the early warning module 40 is further configured to perform feature extraction on the current target image to obtain feature extraction information; performing feature annotation on the current target image based on the feature extraction information to obtain a figure image annotation frame; and tracking the visual angle through a preset visual angle tracking model according to the figure image labeling frame to obtain a visual angle tracking result.
In an embodiment, the early warning module 40 is further configured to obtain a person image labeling frame sample and a perspective tracking result sample corresponding to the person image labeling frame sample; and performing model training on the figure image labeling frame sample and the view tracking result through an initial neural network to obtain a preset view tracking model.
It should be understood that the above is only an example, and the technical solution of the present invention is not limited in any way, and in a specific application, a person skilled in the art may set the technical solution as needed, and the present invention is not limited thereto.
It should be noted that the above-described work flows are only exemplary, and do not limit the scope of the present invention, and in practical applications, a person skilled in the art may select some or all of them to achieve the purpose of the solution of the embodiment according to actual needs, and the present invention is not limited herein.
In addition, the technical details that are not described in detail in this embodiment may refer to the warning method provided in any embodiment of the present invention, and are not described herein again.
Further, it is to be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solution of the present invention or portions thereof that contribute to the prior art may be embodied in the form of a software product, where the computer software product is stored in a storage medium (e.g. Read Only Memory (ROM)/RAM, magnetic disk, optical disk), and includes several instructions for enabling a terminal device (e.g. a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.

Claims (10)

1. An early warning method, characterized in that the early warning method comprises:
when a door opening signal is detected, acquiring a current target image;
carrying out face recognition through a preset face recognition model based on the current target image to obtain face information;
when the face information fails to be compared with the prestored target information, reminding a user of voiceprint recognition;
and when the voiceprint signal is not received or the voiceprint identification fails, generating early warning information and sending the early warning information to a user alarm host for early warning, wherein the user alarm host is also used for managing the networking and running functions of each electronic device.
2. The early warning method as claimed in claim 1, wherein the obtaining of the face information by performing face recognition based on the current target image through a preset face recognition model comprises:
carrying out human shape detection on the current target image through a preset human shape detection model to obtain a human image;
extracting information of five sense organs of the character image, and determining whether an effective target exists according to the information of the five sense organs;
and if the effective target exists, carrying out face recognition through a preset face recognition model based on the current target image to obtain face information.
3. The warning method as claimed in claim 2, wherein after extracting the information of the five sense organs of the human figure and determining whether a valid target exists according to the information of the five sense organs, the method further comprises:
if no effective target exists, extracting the information of the figure image to obtain the body shape information;
comparing the body shape information with target body shape information in prestored target information;
and when the comparison between the body shape information and the target body shape information fails, reminding the user of carrying out voiceprint recognition.
4. The warning method as claimed in claim 2, wherein the extracting of the information on the five sense organs of the human figure and the determining of whether there is a valid target according to the information on the five sense organs comprises:
extracting the information of the five sense organs of the figure image;
when the eye information and the mouth information coexist in the five sense organs information, it is determined that a valid target exists.
5. The warning method according to any one of claims 1 to 4, wherein after the warning information is generated and sent to the user alarm host for warning when the voiceprint signal is not received or when voiceprint recognition fails, the method further comprises:
carrying out visual angle tracking on the current target image through a preset visual angle tracking model to obtain a visual angle tracking result;
and controlling the camera to rotate according to the visual angle tracking result so that the camera collects a target image.
6. The early warning method as claimed in claim 5, wherein the step of tracking the current target image through a preset view tracking model to obtain a view tracking result comprises:
performing feature extraction on the current target image to obtain feature extraction information;
performing feature annotation on the current target image based on the feature extraction information to obtain a figure image annotation frame;
and tracking the visual angle through a preset visual angle tracking model according to the figure image labeling frame to obtain a visual angle tracking result.
7. The warning method as claimed in claim 5, wherein before the tracking of the angle of view through a preset angle of view tracking model according to the character image labeling frame and obtaining the tracking result of the angle of view, the method further comprises:
acquiring a figure image labeling frame sample and a view tracking result sample corresponding to the figure image labeling frame sample;
and performing model training on the figure image labeling frame sample and the view tracking result through an initial neural network to obtain a preset view tracking model.
8. An early warning device, characterized in that the early warning device comprises:
the acquisition module is used for acquiring a current target image when a door opening signal is detected;
the detection module is used for carrying out face recognition through a preset face recognition model based on the current target image to obtain face information;
the recognition module is used for reminding a user of voiceprint recognition when the comparison between the face information and the prestored target information fails;
and the early warning module is used for generating early warning information and sending the early warning information to the user alarm host for early warning when the voiceprint signal is not received or voiceprint identification fails, and the user alarm host is also used for managing networking and operating functions of each electronic device.
9. An early warning device, characterized in that the early warning device comprises: a memory, a processor, and a warning program stored on the memory and executable on the processor, the warning program configured to implement the warning method of any one of claims 1 to 7.
10. A storage medium having stored thereon a warning program which, when executed by a processor, implements a warning method according to any one of claims 1 to 7.
CN202111190167.5A 2021-10-12 2021-10-12 Early warning method, device, equipment and storage medium Pending CN114078273A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111190167.5A CN114078273A (en) 2021-10-12 2021-10-12 Early warning method, device, equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111190167.5A CN114078273A (en) 2021-10-12 2021-10-12 Early warning method, device, equipment and storage medium

Publications (1)

Publication Number Publication Date
CN114078273A true CN114078273A (en) 2022-02-22

Family

ID=80283349

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111190167.5A Pending CN114078273A (en) 2021-10-12 2021-10-12 Early warning method, device, equipment and storage medium

Country Status (1)

Country Link
CN (1) CN114078273A (en)

Similar Documents

Publication Publication Date Title
CN113366487A (en) Operation determination method and device based on expression group and electronic equipment
Jain et al. An introduction to biometric recognition
KR101291899B1 (en) Procedure for identifying a person by eyelash analysis and acquisition device using thereof
JP3469031B2 (en) Face image registration apparatus and method
CN107992739A (en) User authentication method, apparatus and system
CN107844748A (en) Auth method, device, storage medium and computer equipment
CN112328999B (en) Double-recording quality inspection method and device, server and storage medium
CN108875341A (en) A kind of face unlocking method, device, system and computer storage medium
CN107133608A (en) Identity authorization system based on In vivo detection and face verification
JP2010128938A (en) Authentication device and authentication method
US20020176610A1 (en) Face image recording system
KR20070062599A (en) Security alarm notification using iris detection systems
WO2020167155A1 (en) Method and system for detecting troubling events during interaction with a self-service device
Alrahawe et al. A Biometric Technology‐Based Framework for Tackling and Preventing Crimes
EP4057237A1 (en) Reference image enrollment and evolution for security systems
JP2002304651A (en) Device and method for managing entering/leaving room, program for executing the same method and recording medium with the same execution program recorded thereon
JP2002279466A (en) Device and method for admission control
CN109448193A (en) Identity information recognition methods and device
CN112183167B (en) Attendance checking method, authentication method, living body detection method, device and equipment
CN109345186B (en) Service handling method based on Internet of things and terminal equipment
CN111932755A (en) Personnel passage verification method and device, computer equipment and storage medium
KR100888824B1 (en) Portable fingerprint Recognition and identificaion apparatus and method using the same
CN114078273A (en) Early warning method, device, equipment and storage medium
CN107491759B (en) Mixed mode registration method and device
Pravinthraja et al. Multimodal biometrics for improving automatic teller machine security

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination