CN109584907A - A kind of method and apparatus of abnormal alarm - Google Patents
A kind of method and apparatus of abnormal alarm Download PDFInfo
- Publication number
- CN109584907A CN109584907A CN201811445916.2A CN201811445916A CN109584907A CN 109584907 A CN109584907 A CN 109584907A CN 201811445916 A CN201811445916 A CN 201811445916A CN 109584907 A CN109584907 A CN 109584907A
- Authority
- CN
- China
- Prior art keywords
- mood
- signal
- human
- alarm
- voice signal
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 24
- 230000036651 mood Effects 0.000 claims abstract description 110
- 230000005236 sound signal Effects 0.000 claims abstract description 38
- 238000012806 monitoring device Methods 0.000 claims abstract description 19
- 230000008451 emotion Effects 0.000 claims abstract description 15
- 230000009471 action Effects 0.000 claims description 43
- 238000013528 artificial neural network Methods 0.000 claims description 9
- 238000012544 monitoring process Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 7
- 230000008901 benefit Effects 0.000 description 6
- 206010000117 Abnormal behaviour Diseases 0.000 description 3
- 238000000513 principal component analysis Methods 0.000 description 3
- 230000001960 triggered effect Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 206010016275 Fear Diseases 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000001815 facial effect Effects 0.000 description 1
- 230000008921 facial expression Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
- G10L25/63—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination for estimating an emotional state
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B25/00—Alarm systems in which the location of the alarm condition is signalled to a central station, e.g. fire or police telegraphic systems
- G08B25/01—Alarm systems in which the location of the alarm condition is signalled to a central station, e.g. fire or police telegraphic systems characterised by the transmission medium
- G08B25/012—Alarm systems in which the location of the alarm condition is signalled to a central station, e.g. fire or police telegraphic systems characterised by the transmission medium using recorded signals, e.g. speech
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/27—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique
- G10L25/30—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique using neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Signal Processing (AREA)
- Acoustics & Sound (AREA)
- Human Computer Interaction (AREA)
- Hospice & Palliative Care (AREA)
- Psychiatry (AREA)
- Child & Adolescent Psychology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- General Physics & Mathematics (AREA)
- Alarm Systems (AREA)
Abstract
The present invention provides a kind of method and apparatus of abnormal alarm, this method comprises: receiving the audio signal that monitoring device uploads, identify human voice signal in the audio signal；Content recognition is carried out to the human voice signal, whether is judged in the content of the human voice signal comprising preset keyword；Emotion identification is carried out to the human voice signal, judges whether the corresponding mood of the human voice signal belongs to fierce mood, the fierceness mood includes at least one of: angry, frightened, surprised；When the content of the human voice signal includes preset keyword and/or the corresponding mood of the human voice signal belongs to fierce mood, object is accepted to preset alarm and sends alarm signal.The real time monitoring to sound in environment may be implemented in the present invention, when detecting fierce mood or preset keyword, accepts object to preset alarm in time and sends alarm signal, to form a quickly alarm, meets safety precaution requirement.
Description
Technical field
The present invention relates to security alarm technical fields, more particularly to a kind of method and apparatus of abnormal alarm.
Background technique
Safety problem is all the time by social concerns, in recent years, with increasing for personal safety accident, related person peace
The needs of full property are especially urgent.Currently, needing the injured party itself initiative alarming or the site of the accident when personal safety accident occurs
Other staff assist alarm, but at this moment, often victim fears or be bound the energy for losing and voluntarily alarming due to overstretched
Power, and the other staff of the site of the accident are also because fear to lead to not and alarm, so that the personal safety of victim cannot get
It ensures.
Summary of the invention
The present invention provide a kind of method and apparatus of abnormal alarm to overcome the above problem or at least be partially solved on
State problem.
According to an aspect of the invention, there is provided a kind of method of abnormal alarm, comprising:
The audio signal that monitoring device uploads is received, identifies human voice signal in the audio signal；
Content recognition is carried out to the human voice signal, whether is judged in the content of the human voice signal comprising preset key
Word；
Emotion identification is carried out to the human voice signal, judges whether the corresponding mood of the human voice signal belongs to fierce feelings
Thread, the fierceness mood includes at least one of: angry, frightened, surprised；
When the content of the human voice signal include preset keyword and/or the corresponding mood of the human voice signal belong to it is sharp
When strong mood, object is accepted to preset alarm and sends alarm signal.
It is optionally, described to judge whether the corresponding mood of the human voice signal belongs to fierce mood, comprising:
By the preset fierce mood model of human voice signal input, the voice is judged by the fierce mood model
Whether the corresponding mood of signal belongs to fierce mood；
Wherein, the fierce mood model is using deep neural network algorithm by belonging to the fierce mood to known
Audio be trained.
Optionally, the above method further include:
Receive the vision signal that the monitoring device uploads；
Human action in the vision signal is identified, judges whether the human action belongs to dangerous play；
When the human action belongs to dangerous play, object is accepted to the alarm and sends alarm signal.
Optionally, the human action in the vision signal identifies, judges whether the human action belongs to
In dangerous play, comprising:
Human body contour outline in the vision signal is identified, while the profile for the article that human body is held is known
Not；
When the separating degree of the profile of the profile and trunk of human limb is more than that preset value and/or the human body are held
When the profile of article belongs to the profile of dangerous goods, judge that the human action in the vision signal belongs to the dangerous play.
Optionally, the dangerous goods include at least one of: cutter, club, firearms.
Optionally, after accepting object to the alarm and having sent the alarm signal, further includes:
The audio signal for triggering the alarm signal and/or vision signal are sent to the alarm and accept object.
According to another aspect of the present invention, a kind of device of abnormal alarm is additionally provided, comprising:
First signal receiving module, the audio signal uploaded suitable for receiving monitoring device, is identified in the audio signal
Human voice signal；
First identification judgment module, is suitable for carrying out content recognition to the human voice signal, judges that the human voice signal's is interior
It whether include preset keyword in appearance；
Second identification judgment module, is suitable for carrying out Emotion identification to the human voice signal, judges that the human voice signal is corresponding
Mood whether belong to fierce mood, the fierceness mood includes at least one of: indignation, frightened, surprised；
First alarm module includes that preset keyword and/or the voice are believed suitable for the content as the human voice signal
When number corresponding mood belongs to fierce mood, object is accepted to preset alarm and sends alarm signal.
Optionally, the second identification judgment module is further adapted for:
By the preset fierce mood model of human voice signal input, the voice is judged by the fierce mood model
Whether the corresponding mood of signal belongs to fierce mood；
Wherein, the fierce mood model is using deep neural network algorithm by belonging to the fierce mood to known
Audio be trained.
Optionally, above-mentioned apparatus further include:
Second signal receiving module, the vision signal uploaded suitable for receiving the monitoring device；
Third identifies that judgment module judges the human body suitable for identifying to the human action in the vision signal
Whether movement belongs to dangerous play；
Second alarm module, suitable for accepting object to the alarm and sending when the human action belongs to dangerous play
Alarm signal.
Optionally, the third identification judgment module is further adapted for:
Human body contour outline in the vision signal is identified, while the profile for the article that human body is held is known
Not；
When the separating degree of the profile of the profile and trunk of human limb is more than that preset value and/or the human body are held
When the profile of article belongs to the profile of dangerous goods, judge that the human action in the vision signal belongs to the dangerous play.
Optionally, the dangerous goods include at least one of: cutter, club, firearms.
Optionally, first alarm module is further adapted for for the audio signal for triggering the alarm signal being sent to described
Alarm accepts object；And/or
Second alarm module be further adapted for for the vision signal for triggering the alarm signal being sent to the alarm by
Manage object.
The method and apparatus of abnormal alarm provided by the invention can identify that voice is believed from the audio signal received
Number, and by the content recognition to human voice signal, judge whether comprising preset keyword in the content of human voice signal, and
By the Emotion identification to human voice signal, judge whether the corresponding mood of human voice signal belongs to fierce mood, works as human voice signal
Content when including that preset keyword and/or the corresponding mood of human voice signal belong to fierce mood, accepted to preset alarm
Object sends alarm signal.The real time monitoring to sound in environment may be implemented in technical solution provided by the invention, is detecting
When fierce mood and/or preset keyword, object is accepted to preset alarm in time and sends alarm signal, to form one
Quickly alarm, meets safety precaution requirement.
The above description is only an overview of the technical scheme of the present invention, in order to better understand the technical means of the present invention,
And it can be implemented in accordance with the contents of the specification, and in order to allow above and other objects of the present invention, feature and advantage can
It is clearer and more comprehensible, the followings are specific embodiments of the present invention.
According to the following detailed description of specific embodiments of the present invention in conjunction with the accompanying drawings, those skilled in the art will be brighter
The above and other objects, advantages and features of the present invention.
Detailed description of the invention
By reading the following detailed description of the preferred embodiment, various other advantages and benefits are common for this field
Technical staff will become clear.The drawings are only for the purpose of illustrating a preferred embodiment, and is not considered as to the present invention
Limitation.And throughout the drawings, the same reference numbers will be used to refer to the same parts.In the accompanying drawings:
Fig. 1 shows the method flow schematic diagram of abnormal alarm according to an embodiment of the invention；
Fig. 2 shows the method flow schematic diagrams of abnormal alarm in accordance with another embodiment of the present invention；
Fig. 3 shows the apparatus structure schematic diagram of abnormal alarm according to an embodiment of the invention；
Fig. 4 shows the apparatus structure schematic diagram of abnormal alarm in accordance with another embodiment of the present invention；
Specific embodiment
Exemplary embodiments of the present disclosure are described in more detail below with reference to accompanying drawings.Although showing the disclosure in attached drawing
Exemplary embodiment, it being understood, however, that may be realized in various forms the disclosure without should be by embodiments set forth here
It is limited.On the contrary, these embodiments are provided to facilitate a more thoroughly understanding of the present invention, and can be by the scope of the present disclosure
Completely it is communicated to those skilled in the art.
It should be noted that the feature in the embodiment of the present invention and preferred embodiment can be mutual under the premise of not conflicting
In conjunction with.
Fig. 1 shows the method flow schematic diagram of abnormal alarm according to an embodiment of the invention.As shown in Figure 1, should
Method at least includes the following steps S102 to step S108.
Step S102, the audio signal that monitoring device uploads is received, identifies human voice signal in audio signal.
The speech frequency of the mankind mainly between 500-3000Hz, in an embodiment of the present invention, can preset one
A frequency range carries out analysis to audio signal and obtains the audio signal after the audio signal for receiving monitoring device upload
Frequency, the audio signal for meeting predeterminated frequency range is intercepted out, using the above-mentioned audio signal intercepted out as in audio signal
Human voice signal.
Step S104, content recognition is carried out to human voice signal, whether judged in the content of human voice signal comprising preset pass
Keyword.
In a preferred embodiment of the invention, it can be identified in the text of human voice signal by speech recognition technology
Hold, word extraction operation is carried out to the content of text of human voice signal, obtains at least one content of text vocabulary, by least one text
This content vocabulary is matched with keywords database, if it exists the keyword of matched text content vocabulary, then judges human voice signal's
It include preset keyword in content.Conversely, the keyword of matched text content vocabulary if it does not exist, then can terminate this and sentence
It is disconnected.Here the keyword obtained by data mining mode or human configuration mode is stored in keywords database, such as " rescuing ",
" help ", " fire fighting ", " murder ", " Help " etc..In another preferred embodiment of the invention, expression can also be preset to exhale
The acoustic information rescued, the acoustic information here preset at may include the keyword in keywords database, by human voice signal with it is preset
Acoustic information is compared, and matches if there is at least one preset voice signal with human voice signal, then judges that voice is believed
Number content in include preset keyword., whereas if matching there is no preset voice signal with human voice signal, then tie
Shu Benci judgement.
Step S106, Emotion identification is carried out to human voice signal, judges whether the corresponding mood of human voice signal belongs to fierce feelings
Thread, wherein fierce mood includes at least one of: angry, frightened, surprised.
Emotion identification based on affection computation, is an important component of affection computation, and Emotion identification is studied interior
Holding includes facial expression, voice, heart rate, behavior, text and physiological signal identification etc., and Emotion identification mostly uses facial table
The methods of common scaling method on feelings, body behavior and speech signal analysis method and psychology.Presently used voice mood
Recognition methods mainly has PCA (Principal Component Analysis, Principal Component Analysis), ANN (Artificial
Neural Networks, artificial neural network method), HMM (Hidden Markon Model, Hidden Markov Model method), GMM
(Gaussian Mixture Model, gauss hybrid models method) etc..
Above-mentioned steps S106 can be realized in several ways, in a preferred embodiment of the invention, can be incited somebody to action
The preset fierce mood model of human voice signal's input, judges whether the corresponding mood of human voice signal belongs to by fierce mood model
Fierce mood, fierce mood model here can be using deep neural network algorithm by belonging to the fierce mood to known
Audio be trained to obtain.
In an embodiment of the present invention, the audio of fierce mood can voluntarily record phase using sound pick-up outfit in quiet room
The fierce mood voice answered also can use the phonetic materials such as existing telecine, choose corresponding fierce mood voice, this
Invention to this with no restriction.
Step S108, when the content of human voice signal includes preset keyword and/or the corresponding mood of the human voice signal
When belonging to fierce mood, object is accepted to preset alarm and sends alarm signal.
The method and apparatus of abnormal alarm provided in an embodiment of the present invention can be identified from the audio signal received
Human voice signal, and by the content recognition to human voice signal, whether judge in the content of human voice signal comprising preset key
Word, and by the Emotion identification to human voice signal, judge whether the corresponding mood of human voice signal belongs to fierce mood, work as people
When the content of acoustical signal belongs to fierce mood comprising preset keyword and/or the corresponding mood of human voice signal, to preset report
The alert object that accepts sends alarm signal.The real time monitoring to sound in environment may be implemented in the embodiment of the present invention, is detecting feelings
When the keyword of thread fierceness, object is accepted to preset alarm in time and sends alarm signal, so that a quickly alarm is formed,
Meet safety precaution requirement.Also, the embodiment of the present invention is known by the mood of content recognition and human voice signal to human voice signal
Not, the generation that can effectively reduce wrong report phenomenon, further increases the accuracy rate of alarm.
Fig. 2 shows the flow diagrams of the method for abnormal alarm in accordance with another embodiment of the present invention.Such as Fig. 2 institute
Show, this method at least includes the following steps S202 to step S206.
Step S202, the vision signal that monitoring device uploads is received.
Step S204, the human action in vision signal is identified, judges whether human action belongs to dangerous play.
Step S206, when human action belongs to dangerous play, object is accepted to alarm and sends alarm signal.
Whether the embodiment of the present invention judges human action by identifying to the human action in vision signal
Belong to dangerous play, when determining that human action belongs to dangerous play, accepts object to preset alarm and send alarm signal.This
Inventive embodiments can in time finder abnormal behaviour movement, when determine the abnormal behaviour movement belong to dangerous play, carry out
Timely alarm.
Above-mentioned steps S204 can there are many implementation in a preferred embodiment of the invention can be to video
Human body contour outline in signal is identified, is obtained the profile of human limb and the profile of trunk, is calculated human limb
Profile and trunk profile separating degree, when the separating degree of the profile of the profile and trunk of human limb is more than pre-
If when value, determining that the human action in vision signal belongs to dangerous play.Here separating degree is to judge human limb and human body
One standard of trunk distance, separating degree is larger to mean human limb farther out from trunk, i.e., the movement range of human body compared with
Greatly, the possibility of dangerous movement.In a further advantageous embodiment, the profile for the article that human body is held can be known
Not, when the profile for the article held when human body belongs to the profile of dangerous goods, judge that the human action of vision signal belongs to danger
Movement.Here dangerous goods include but is not limited to cutter, club, firearms, and can be any can be constituted personal safety
The weapon or article of threat.
The article that above-mentioned human body contour outline and above-mentioned human body are held can also be used in combination and carry out profile by the embodiment of the present invention
Identification.For example, can simultaneously in vision signal human body contour outline and the profile of article held of human body identified,
When the separating degree of the profile of the profile and trunk of human limb is more than the profile category for the article that preset value and human body are held
When the profile of dangerous goods, judge that the human action in vision signal belongs to dangerous play.
Further, after accepting object to alarm and having sent alarm signal, alarm signal will can also be triggered
Audio signal and/or vision signal are sent to preset alarm and accept object, and the user for enabling alarm to accept object knows
Dawn triggers the audio signal of alarm signal and/or the particular content of vision signal, and then determines the audio signal and/or video
Whether signal is to need the case where alarming, and further prevents the generation of wrong report phenomenon, improves the accuracy rate of alarm.Also, it will
The audio signal and/or vision signal for triggering alarm signal be sent to it is preset alarm accept object, can also make alarm by
The user for managing object prejudges the mode of processing, is also convenient for subsequent timely evidence obtaining.
Based on the same inventive concept, the embodiment of the invention also provides a kind of devices of abnormal alarm.Fig. 3 is according to this hair
The apparatus structure schematic diagram of the abnormal alarm of bright one embodiment.As shown in figure 3, the device at least may include: the first signal
Receiving module 310, first identifies that judgment module 320, second identifies judgment module 330 and the first alarm module 340, wherein
First signal receiving module 310, the audio signal uploaded suitable for receiving monitoring device, identifies people in audio signal
Acoustical signal；
First identification judgment module 320, couples with the first signal receiving module 310, is suitable for carrying out the human voice signal
Whether content recognition judges in the content of the human voice signal comprising preset keyword；
Second identification judgment module 330, couples with the first signal receiving module 310, is suitable for carrying out mood to human voice signal
Identification, judges whether the corresponding mood of human voice signal belongs to fierce mood, wherein fierce mood includes at least one of: anger
It is anger, fear, surprised；
First alarm module 340 couples with the first identification judgment module 320 and the second identification judgment module 330, is suitable for
When the content of human voice signal includes that preset keyword and/or the corresponding mood of human voice signal belong to fierce mood, to default
Alarm accept object send alarm signal.
In a preferred embodiment, the second identification judgment module 330 is further adapted for: the human voice signal being inputted pre-
If fierce mood model, judge whether the corresponding mood of the human voice signal belongs to fierce feelings by the fierce mood model
Thread；
In a preferred embodiment, cloud can also receive the vision signal of monitoring device upload, such as Fig. 4 institute
Show, which can also include:
Second signal receiving module 410, the vision signal uploaded suitable for receiving the monitoring device；
Third identifies that judgment module 420 judges the people suitable for identifying to the human action in the vision signal
Whether body movement belongs to dangerous play；
Second alarm module 430, suitable for accepting object hair to the alarm when the human action belongs to dangerous play
Send alarm signal.
In a preferred embodiment, third identification judgment module 420 is further adapted for:
Human body contour outline in the vision signal is identified, while the profile for the article that human body is held is known
Not；
When the separating degree of the profile of the profile and trunk of human limb is more than that preset value and/or the human body are held
When the profile of article belongs to the profile of dangerous goods, judge that the human action in the vision signal belongs to the dangerous play.
In a preferred embodiment, dangerous goods include at least one of: cutter, club, firearms.
In a preferred embodiment, the first alarm module 340 is further adapted for that the sound of the alarm signal will be triggered
Frequency signal is sent to the alarm and accepts object；And/or
Second alarm module 430 is further adapted for for the vision signal for triggering the alarm signal being sent to the alarm and accepts
Object.
The method and apparatus of abnormal alarm provided in an embodiment of the present invention can be identified from the audio signal received
Human voice signal, and by the content recognition to human voice signal, whether judge in the content of human voice signal comprising preset key
Word, and by the Emotion identification to human voice signal, judge whether the corresponding mood of human voice signal belongs to fierce mood, work as people
When the content of acoustical signal belongs to fierce mood comprising preset keyword and/or the corresponding mood of human voice signal, to preset report
The alert object that accepts sends alarm signal.The real-time prison to sound in environment may be implemented in technical solution provided in an embodiment of the present invention
Control accepts object to preset alarm in time and sends alarm signal when detecting the keyword having intense feelings, to form one
A quick alarm, meets safety precaution requirement.Also, technical solution provided in an embodiment of the present invention passes through to human voice signal's
The matching of the Emotion identification of content recognition and human voice signal can effectively reduce the generation of wrong report phenomenon, further increase alarm
Accuracy rate.
The method and apparatus of abnormal alarm provided in an embodiment of the present invention, can also be by dynamic to the human body in vision signal
The identification of work, and then judge whether human action belongs to dangerous play, when determining that human action belongs to dangerous play, to pre-
If alarm accept object send alarm signal.Technical solution provided in an embodiment of the present invention can in time finder abnormal row
For movement, belongs to dangerous play when determining that the abnormal behaviour acts, timely alarmed.
Further, technical solution based on the embodiment of the present invention, the user that alarm accepts object energy enough know
The audio signal of alarm signal and/or the particular content of vision signal are triggered, and then determines the audio signal and/or video letter
Number whether it is to need the case where alarming, further prevents the generation of wrong report phenomenon, improve the effective percentage of alarm.Also, it will touching
The audio signal and/or vision signal for having sent out alarm signal are sent to preset alarm and accept object, and alarm can also be made to accept
The user of object prejudges the mode of processing, is also convenient for subsequent timely evidence obtaining.
In the above description of the exemplary embodiment of the present invention, each feature of the invention is grouped together into sometimes
In single embodiment, figure or descriptions thereof.However, the disclosed method should not be interpreted as reflecting the following intention: i.e.
The claimed invention requires features more more than feature expressly recited in each claim.More precisely
It says, as reflected in the following claims, inventive aspect is all less than single embodiment disclosed above
Feature.Therefore, it then follows thus claims of specific embodiment are expressly incorporated in the specific embodiment, wherein each power
Benefit requires itself all as a separate embodiment of the present invention.
Those skilled in the art will understand that can be carried out adaptively to the module in the equipment in embodiment
Change and they are arranged in one or more devices different from this embodiment.It can be the module or list in embodiment
Member or component are combined into a module or unit or component, and furthermore they can be divided into multiple submodule or subelement or
Sub-component.Other than such feature and/or at least some of process or unit exclude each other, it can use any
Combination is to all features disclosed in this specification (including adjoint claim, abstract and attached drawing) and so disclosed
All process or units of what method or apparatus are combined.Unless expressly stated otherwise, this specification is (including adjoint power
Benefit require, abstract and attached drawing) disclosed in each feature can carry out generation with an alternative feature that provides the same, equivalent, or similar purpose
It replaces.
In addition, it will be appreciated by those of skill in the art that although some embodiments described herein include other embodiments
In included certain features rather than other feature, but the combination of the feature of different embodiments mean it is of the invention
Within the scope of and form different embodiments.For example, in detail in the claims, embodiment claimed it is one of any
Can in any combination mode come using.
So far, although those skilled in the art will appreciate that present invention has been shown and described in detail herein multiple shows
Example property embodiment still without departing from the spirit and scope of the present invention, still can according to the present disclosure directly
Determine or deduce out many other variations or modifications consistent with the principles of the invention.Therefore, the scope of the present invention is understood that and recognizes
It is set to and covers all such other variations or modifications.
The embodiment of the invention discloses a kind of methods of abnormal alarm of A1., comprising:
The audio signal that monitoring device uploads is received, identifies human voice signal in the audio signal；
Content recognition is carried out to the human voice signal, whether is judged in the content of the human voice signal comprising preset key
Word；
Emotion identification is carried out to the human voice signal, judges whether the corresponding mood of the human voice signal belongs to fierce feelings
Thread, wherein the fierceness mood includes at least one of: angry, frightened, surprised；
When the content of the human voice signal include preset keyword and/or the corresponding mood of the human voice signal belong to it is sharp
When strong mood, object is accepted to preset alarm and sends alarm signal.
A2. method according to a1, wherein described to judge whether the corresponding mood of the human voice signal belongs to fierce feelings
Thread, comprising:
By the preset fierce mood model of human voice signal input, the voice is judged by the fierce mood model
Whether the corresponding mood of signal belongs to fierce mood；
Wherein, the fierce mood model is using deep neural network algorithm by belonging to the fierce mood to known
Audio be trained.
A3. method according to a1 or a2, wherein further include:
Receive the vision signal that the monitoring device uploads；
Human action in the vision signal is identified, judges whether the human action belongs to dangerous play；
When the human action belongs to dangerous play, object is accepted to the alarm and sends alarm signal.
A4. method according to a3, wherein the human action in the vision signal identifies, judges
Whether the human action belongs to dangerous play, comprising:
Human body contour outline in the vision signal is identified, while the profile for the article that human body is held is known
Not；
When the separating degree of the profile of the profile and trunk of human limb is more than that preset value and/or the human body are held
When the profile of article belongs to the profile of dangerous goods, judge that the human action in the vision signal belongs to the dangerous play.
A5. method according to a4, wherein the dangerous goods include at least one of: cutter, club, firearms.
A6. method according to a3, wherein after accepting object to the alarm and having sent the alarm signal,
Further include:
The audio signal for triggering the alarm signal and/or vision signal are sent to the alarm and accept object.
The present embodiment also discloses a kind of device of abnormal alarm of B7., comprising:
First signal receiving module, the audio signal uploaded suitable for receiving monitoring device, is identified in the audio signal
Human voice signal；
First identification judgment module, is suitable for carrying out content recognition to the human voice signal, judges that the human voice signal's is interior
It whether include preset keyword in appearance；
Second identification judgment module, is suitable for carrying out Emotion identification to the human voice signal, judges that the human voice signal is corresponding
Mood whether belong to fierce mood, the fierceness mood includes at least one of: indignation, frightened, surprised；
First alarm module includes that preset keyword and/or the voice are believed suitable for the content as the human voice signal
When number corresponding mood belongs to fierce mood, object is accepted to preset alarm and sends alarm signal.
B8. the device according to B7, wherein the second identification judgment module is further adapted for:
By the preset fierce mood model of human voice signal input, the voice is judged by the fierce mood model
Whether the corresponding mood of signal belongs to fierce mood；
Wherein, the fierce mood model is using deep neural network algorithm by belonging to the fierce mood to known
Audio be trained.
B9. the device according to B7 or B8, wherein further include:
Second signal receiving module, the vision signal uploaded suitable for receiving the monitoring device；
Third identifies that judgment module judges the human body suitable for identifying to the human action in the vision signal
Whether movement belongs to dangerous play；
Second alarm module, suitable for accepting object to the alarm and sending when the human action belongs to dangerous play
Alarm signal.
B10. the device according to B9, wherein the third identification judgment module is further adapted for:
Human body contour outline in the vision signal is identified, while the profile for the article that human body is held is known
Not；
When the separating degree of the profile of the profile and trunk of human limb is more than that preset value and/or the human body are held
When the profile of article belongs to the profile of dangerous goods, judge that the human action in the vision signal belongs to the dangerous play.
B11. device according to b10, wherein the dangerous goods include at least one of: cutter, club, rifle
Tool.
B12. the device according to B9, wherein
First alarm module be further adapted for for the audio signal for triggering the alarm signal being sent to the alarm by
Manage object；And/or
Second alarm module be further adapted for for the vision signal for triggering the alarm signal being sent to the alarm by
Manage object.
Claims (10)
1. a kind of method of abnormal alarm, comprising:
The audio signal that monitoring device uploads is received, identifies human voice signal in the audio signal；
Content recognition is carried out to the human voice signal, whether is judged in the content of the human voice signal comprising preset keyword；
Emotion identification is carried out to the human voice signal, judges whether the corresponding mood of the human voice signal belongs to fierce mood,
In, the fierceness mood includes at least one of: angry, frightened, surprised；
When the content of the human voice signal includes that preset keyword and/or the corresponding mood of the human voice signal belong to fierce feelings
When thread, object is accepted to preset alarm and sends alarm signal.
2. described to judge whether the corresponding mood of the human voice signal belongs to fierceness according to the method described in claim 1, wherein
Mood, comprising:
By the preset fierce mood model of human voice signal input, the human voice signal is judged by the fierce mood model
Whether corresponding mood belongs to fierce mood；
Wherein, the fierce mood model is to be passed through using deep neural network algorithm to the known sound for belonging to the fierce mood
What frequency was trained.
3. method according to claim 1 or 2, wherein further include:
Receive the vision signal that the monitoring device uploads；
Human action in the vision signal is identified, judges whether the human action belongs to dangerous play；
When the human action belongs to dangerous play, object is accepted to the alarm and sends alarm signal.
4. sentencing according to the method described in claim 3, wherein, the human action in the vision signal identifies
Whether the human action that breaks belongs to dangerous play, comprising:
Human body contour outline in the vision signal is identified, while the profile for the article that human body is held is identified；
When the separating degree of the profile of the profile and trunk of human limb is more than the article that preset value and/or the human body are held
Profile when belonging to the profile of dangerous goods, judge that the human action in the vision signal belongs to the dangerous play.
5. according to the method described in claim 4, wherein, the dangerous goods include at least one of: cutter, club, rifle
Tool.
6. a kind of device of abnormal alarm, comprising:
First signal receiving module, the audio signal uploaded suitable for receiving monitoring device, identifies voice in the audio signal
Signal；
First identification judgment module is suitable for human voice signal progress content recognition, in the content for judging the human voice signal
It whether include preset keyword；
Second identification judgment module, is suitable for carrying out Emotion identification to the human voice signal, judges the corresponding feelings of the human voice signal
Whether thread belongs to fierce mood, and the fierceness mood includes at least one of: angry, frightened, surprised；
First alarm module includes preset keyword and/or the human voice signal couple suitable for the content as the human voice signal
When the mood answered belongs to fierce mood, object is accepted to preset alarm and sends alarm signal.
7. device according to claim 6, wherein the second identification judgment module is further adapted for:
By the preset fierce mood model of human voice signal input, the human voice signal is judged by the fierce mood model
Whether corresponding mood belongs to fierce mood；
Wherein, the fierce mood model is to be passed through using deep neural network algorithm to the known sound for belonging to the fierce mood
What frequency was trained.
8. device according to claim 6 or 7, wherein further include:
Second signal receiving module, the vision signal uploaded suitable for receiving the monitoring device；
Third identifies that judgment module judges the human action suitable for identifying to the human action in the vision signal
Whether dangerous play is belonged to；
Second alarm module, suitable for accepting object to the alarm and sending alarm when the human action belongs to dangerous play
Signal.
9. device according to claim 8, wherein the third identification judgment module is further adapted for:
Human body contour outline in the vision signal is identified, while the profile for the article that human body is held is identified；
When the separating degree of the profile of the profile and trunk of human limb is more than the article that preset value and/or the human body are held
Profile when belonging to the profile of dangerous goods, judge that the human action in the vision signal belongs to the dangerous play.
10. device according to claim 9, wherein the dangerous goods include at least one of: cutter, club, rifle
Tool.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811445916.2A CN109584907A (en) | 2018-11-29 | 2018-11-29 | A kind of method and apparatus of abnormal alarm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811445916.2A CN109584907A (en) | 2018-11-29 | 2018-11-29 | A kind of method and apparatus of abnormal alarm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109584907A true CN109584907A (en) | 2019-04-05 |
Family
ID=65923959
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811445916.2A Pending CN109584907A (en) | 2018-11-29 | 2018-11-29 | A kind of method and apparatus of abnormal alarm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109584907A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111063162A (en) * | 2019-12-05 | 2020-04-24 | 恒大新能源汽车科技(广东)有限公司 | Silent alarm method and device, computer equipment and storage medium |
| CN111210592A (en) * | 2020-01-07 | 2020-05-29 | 珠海爬山虎科技有限公司 | Video identification monitoring method, computer device and computer readable storage medium |
| CN111553192A (en) * | 2020-03-30 | 2020-08-18 | 深圳壹账通智能科技有限公司 | Hierarchical authority unlocking method and device and storage medium |
| CN111986681A (en) * | 2020-08-31 | 2020-11-24 | 贵州乐诚技术有限公司 | Warehouse supervision management system |
| CN112037820A (en) * | 2019-05-16 | 2020-12-04 | 杭州海康威视数字技术股份有限公司 | Security alarm method, device, system and equipment |
| CN112364696A (en) * | 2020-10-13 | 2021-02-12 | 特斯联科技集团有限公司 | Method and system for improving family safety by using family monitoring video |
| CN112634564A (en) * | 2020-12-30 | 2021-04-09 | 重庆赤木科技有限公司 | Community safety monitoring system |
| CN113099283A (en) * | 2021-03-30 | 2021-07-09 | 深圳市冠标科技发展有限公司 | Method for synchronizing monitoring picture and sound and related equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120308971A1 (en) * | 2011-05-31 | 2012-12-06 | Hyun Soon Shin | Emotion recognition-based bodyguard system, emotion recognition device, image and sensor control apparatus, personal protection management apparatus, and control methods thereof |
| CN106128022A (en) * | 2016-07-18 | 2016-11-16 | 四川君逸数码科技股份有限公司 | A kind of wisdom gold eyeball identification violent action alarm method and device |
| CN106203264A (en) * | 2016-06-28 | 2016-12-07 | 深圳市元征科技股份有限公司 | Dangerous goods detection method and device |
| CN107481470A (en) * | 2017-06-30 | 2017-12-15 | 何东波 | A kind of remote monitoring and automatic alarm method and device |
| CN108694958A (en) * | 2018-04-26 | 2018-10-23 | 广州国音科技有限公司 | A kind of security alarm method and device |
-
2018
- 2018-11-29 CN CN201811445916.2A patent/CN109584907A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120308971A1 (en) * | 2011-05-31 | 2012-12-06 | Hyun Soon Shin | Emotion recognition-based bodyguard system, emotion recognition device, image and sensor control apparatus, personal protection management apparatus, and control methods thereof |
| CN106203264A (en) * | 2016-06-28 | 2016-12-07 | 深圳市元征科技股份有限公司 | Dangerous goods detection method and device |
| CN106128022A (en) * | 2016-07-18 | 2016-11-16 | 四川君逸数码科技股份有限公司 | A kind of wisdom gold eyeball identification violent action alarm method and device |
| CN107481470A (en) * | 2017-06-30 | 2017-12-15 | 何东波 | A kind of remote monitoring and automatic alarm method and device |
| CN108694958A (en) * | 2018-04-26 | 2018-10-23 | 广州国音科技有限公司 | A kind of security alarm method and device |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112037820A (en) * | 2019-05-16 | 2020-12-04 | 杭州海康威视数字技术股份有限公司 | Security alarm method, device, system and equipment |
| CN112037820B (en) * | 2019-05-16 | 2023-09-05 | 杭州海康威视数字技术股份有限公司 | Security alarm method, device, system and equipment |
| CN111063162A (en) * | 2019-12-05 | 2020-04-24 | 恒大新能源汽车科技(广东)有限公司 | Silent alarm method and device, computer equipment and storage medium |
| CN111210592A (en) * | 2020-01-07 | 2020-05-29 | 珠海爬山虎科技有限公司 | Video identification monitoring method, computer device and computer readable storage medium |
| CN111553192A (en) * | 2020-03-30 | 2020-08-18 | 深圳壹账通智能科技有限公司 | Hierarchical authority unlocking method and device and storage medium |
| CN111986681A (en) * | 2020-08-31 | 2020-11-24 | 贵州乐诚技术有限公司 | Warehouse supervision management system |
| CN111986681B (en) * | 2020-08-31 | 2023-03-10 | 贵州乐诚技术有限公司 | Warehouse supervision management system |
| CN112364696A (en) * | 2020-10-13 | 2021-02-12 | 特斯联科技集团有限公司 | Method and system for improving family safety by using family monitoring video |
| CN112364696B (en) * | 2020-10-13 | 2024-03-19 | 特斯联科技集团有限公司 | Method and system for improving family safety by utilizing family monitoring video |
| CN112634564A (en) * | 2020-12-30 | 2021-04-09 | 重庆赤木科技有限公司 | Community safety monitoring system |
| CN113099283A (en) * | 2021-03-30 | 2021-07-09 | 深圳市冠标科技发展有限公司 | Method for synchronizing monitoring picture and sound and related equipment |
| CN113099283B (en) * | 2021-03-30 | 2023-02-14 | 深圳市冠标科技发展有限公司 | Method for synchronizing monitoring picture and sound and related equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109584907A (en) | A kind of method and apparatus of abnormal alarm | |
| CN109769099B (en) | Method and device for detecting abnormality of call person | |
| Ntalampiras et al. | On acoustic surveillance of hazardous situations | |
| US9659564B2 (en) | Speaker verification based on acoustic behavioral characteristics of the speaker | |
| EP3618063B1 (en) | Voice interaction system, voice interaction method and corresponding program | |
| CN108694958B (en) | Security alarm method and device | |
| CN111063162A (en) | Silent alarm method and device, computer equipment and storage medium | |
| Fleury et al. | Sound and speech detection and classification in a health smart home | |
| CN101753992A (en) | Multi-mode intelligent monitoring system and method | |
| CN108109331A (en) | Monitoring method and monitoring system | |
| US20070299671A1 (en) | Method and apparatus for analysing sound- converting sound into information | |
| CN109671234A (en) | A kind of alarm method and device of monitoring device | |
| CN112102850B (en) | Emotion recognition processing method and device, medium and electronic equipment | |
| CN108010513B (en) | Voice processing method and device | |
| CN111192574A (en) | Intelligent voice interaction method, mobile terminal and computer readable storage medium | |
| CN110349586B (en) | Telecommunication fraud detection method and device | |
| CN109036393A (en) | Wake-up word training method, device and the household appliance of household appliance | |
| CN112071309B (en) | Network appointment vehicle safety monitoring device and system | |
| Chebbi et al. | On the use of pitch-based features for fear emotion detection from speech | |
| CN111696559A (en) | Providing emotion management assistance | |
| CN108597506A (en) | A kind of intelligent wearable device alarming method for power and intelligent wearable device | |
| CN106504750A (en) | A kind of device and method thereof of utilization Application on Voiceprint Recognition quick detection scene of fire comburant | |
| Gilke et al. | MFCC-based vocal emotion recognition using ANN | |
| CN110719553B (en) | Smart speaker system with cognitive sound analysis and response | |
| Bojanić et al. | Application of neural networks in emotional speech recognition |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190405 |