CN106407993A - Intelligent voice robot system based on image recognition technology and method thereof - Google Patents
Intelligent voice robot system based on image recognition technology and method thereof Download PDFInfo
- Publication number
- CN106407993A CN106407993A CN201610818330.0A CN201610818330A CN106407993A CN 106407993 A CN106407993 A CN 106407993A CN 201610818330 A CN201610818330 A CN 201610818330A CN 106407993 A CN106407993 A CN 106407993A
- Authority
- CN
- China
- Prior art keywords
- image
- signal
- module
- recognition
- robot
- 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.)
- Withdrawn
Links
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000005516 engineering process Methods 0.000 title claims abstract description 32
- 238000012545 processing Methods 0.000 claims abstract description 47
- 238000003709 image segmentation Methods 0.000 claims description 28
- 230000009466 transformation Effects 0.000 claims description 28
- 238000003909 pattern recognition Methods 0.000 claims description 17
- 238000000605 extraction Methods 0.000 claims description 16
- 230000005540 biological transmission Effects 0.000 claims description 14
- 230000006870 function Effects 0.000 claims description 12
- 238000012546 transfer Methods 0.000 claims description 12
- 239000000284 extract Substances 0.000 claims description 10
- 238000006243 chemical reaction Methods 0.000 claims description 8
- 239000011159 matrix material Substances 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 8
- 230000008054 signal transmission Effects 0.000 claims description 8
- 238000005070 sampling Methods 0.000 claims description 7
- 230000011218 segmentation Effects 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 5
- 206010068319 Oropharyngeal pain Diseases 0.000 claims description 4
- 201000007100 Pharyngitis Diseases 0.000 claims description 4
- 230000000052 comparative effect Effects 0.000 claims description 4
- 238000013500 data storage Methods 0.000 claims description 4
- 238000009432 framing Methods 0.000 claims description 4
- 230000005236 sound signal Effects 0.000 claims description 4
- 238000005728 strengthening Methods 0.000 claims description 4
- 238000011161 development Methods 0.000 description 7
- 230000018109 developmental process Effects 0.000 description 7
- 238000004458 analytical method Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
-
- 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
- G10L15/00—Speech recognition
- G10L15/28—Constructional details of speech recognition systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention provides an intelligent voice robot system based on an image recognition technology and a method thereof, and relates to the field of robots. The system is characterized in that the system comprises a cloud server used for providing data update for a robot; the cloud server is in signal connection with a robot processor used for processing the data information of the robot; and the robot processor is in signal connection with a memory used for storing data information, a voice recognition device used for recognizing voice, and an image processing device used for recognizing images. The system and the method have the advantages of intelligence, voice recognition, image recognition, accurate processing result, real-time update, and the like.
Description
Technical field
The present invention relates to robot field, particularly to a kind of intelligent sound robot system based on image recognition technology
And method.
Background technology
With scientific and technical continuous development, human society initially enters the intelligent epoch, and the development of society is to intelligent skill
Art also constantly proposes new demand, roboticses intelligent robot skill till now by original industry mechanical arm development
Art, the productive life that the development of technology gives people class brings new change.Intelligent robot technology have important theory significance and
Using value.
In theory, roboticses have the characteristics that Multidisciplinary Integration and many Integration ofTechnology, are related to numerous ambits
Sophisticated technology, such as balance control design case, intelligent barrier avoiding design etc., which results in its complexity, make intelligent robot perception skill
Art, communication technology, the research of frame for movement are all a kind of challenges, promote feature extraction algorithm, recognizer, path planning algorithm
Develop towards intelligentized direction, promote the development of pattern recognition theory;In addition, whether robot realizes intellectuality and robot
The structure of itself also has very big relation, can will complete task full intellectualizedly, except intellectualized algorithm to be had also will
There are flexible mechanism, the therefore development of the intelligent robot technology development theoretical by being greatly promoted frame for movement, its developing water
The flat comprehensive strength that can reflect a country information and automatic technology..
Content of the invention
In consideration of it, the invention provides a kind of intelligent sound robot system based on image recognition technology and method, basis
Invention has the advantages that intellectuality, possesses speech recognition, possesses image recognition, result accurately and real-time update.
The technical solution used in the present invention is as follows:
A kind of intelligent sound robot system based on image recognition technology is it is characterised in that described system includes:For for machine
Device people provides the cloud server that data updates;Described cloud server signal is connected to for handling machine personal data information
Robot processor;Signal is connected to memorizer for data storage information, is used for identifying described robot processor respectively
The speech recognition equipment of voice and the image processing apparatus for identifying image.
Described speech recognition equipment includes:For obtaining the voice collection device of primary speech signal;Described sound collection
Device signal is connected to the voice recognition chip for being identified to voice;Described voice recognition chip includes:For to adopting
The acoustical signal collecting carries out the sub-frame processing module of sub-frame processing;Signal is connected to for sentencing described sub-frame processing module respectively
Judge module and voice collection device that whether the acoustical signal after disconnected sub-frame processing is coordinated;Described judge module signal connects
In the Fourier transformation module for acoustical signal is carried out with Fourier transformation;Described Fourier transformation module by signal is connected to use
The frequency-domain analysiss module comparing in frequency domain.
Described image identifying device includes:For obtaining the image collecting device of original image signal;Described image gathers
Device signal is connected to the image recognition chip for being identified to image;Described image identification chip includes:For to adopting
The picture signal collecting is identified, and it includes:Image pre-processing module, image segmentation module, characteristic extracting module and image
Identification module;Described image acquisition module signal is connected to image pre-processing module;Described image pretreatment module signal connects
In image segmentation module;Described image segmentation module by signal is connected to characteristic extracting module;Described characteristic extracting module signal is even
It is connected to picture recognition module;Described image identification module signal is connected to dsp processor;It is right that described image pretreatment module is used for
The original image information collecting carries out Image semantic classification, and pretreated image is sent to image segmentation module;Described figure
As segmentation module, for the image after Image semantic classification is carried out image segmentation, the image after Image semantic classification is sent to spy
Levy extraction module;Described characteristic extracting module, for the image after segmentation is carried out feature extraction, by the image after feature extraction
Send to picture recognition module;Described image identification module, for carrying out image recognition to the image after feature extraction.
Described cloud server includes:Cloud database and high in the clouds data transmission device;Described cloud database signal is even
It is connected to high in the clouds data transmission device, for storing template data;Belonging high in the clouds data transmission device, for by cloud database
In template data be sent to the robot processor of robot.
Described robot processor includes:Signal transmission unit and control unit;Described signal transmission unit, for accepting
The data message sending over from high in the clouds data transmission device, the data message receiving is sent and is deposited to memorizer
Storage.
A kind of intelligent sound based on image recognition technology identifies the method for robot system it is characterised in that described side
Method comprises the following steps:
Step 1:For sending sound control order, after speech recognition equipment collects this sound control order, from memorizer
Transfer standard voice signals, carry out speech recognition, recognition result is sent to robot processor;
Step 2:According to the recognition result sending over, judgement should carry out any operation to robot processor, if user requires
Robot carries out image recognition, and robot starts startup pattern recognition device and image is identified;
Step 3:After pattern recognition device collects original picture signal, transfer video standard signal from memorizer, to figure
As signal is identified, recognition result is sent to robot processor;Robot processor according to the recognition result receiving,
Transmitting control commands are to speech recognition equipment;
Step 4:The recognition result that speech recognition equipment sends over according to robot processor, sends acoustical signal.
The recognition methodss of described speech recognition equipment comprise the following steps:
Step 1:After voice collection device collects original sound signal, sub-frame processing module is carried out at framing to acoustical signal
Reason;Each frame acoustical signal after sub-frame processing is sent to judge module;
Step 2:Coordinating property of acoustical signal after judge module is processed to each frame judges;After will determine that result is preserved
Send to Fourier transformation module;
Step 3:Fourier transformation module starts to carry out Fourier transformation to acoustical signal, by conversion after acoustical signal send to
Comparison module;
Step 4:Comparison module carries out frequency domain to the acoustical signal after Fourier transformation and compares, and comparative result is sent to memorizer
In kept in;
Step 5:Sub-frame processing module when processing acoustical signal, can real-time judge process be whether acoustical signal last
Frame, if it is transfers the storage result in flash memory and sends and shown to display device;If not then continuing executing with step
1.
The method that described judge module judges to the coordinating property of acoustical signal after sub-frame processing comprises the following steps:
Step 1:Ask for the short-time energy of the signal after sub-frame processing, the acquiring method of described short-time energy adopts equation below:
, whereinIt is that acoustical signal is believed in the sampling of certain point
Number;
Step 2:Sore throat relieving or Voiced signal are distinguished in short-time energy according to seeking out;
Step 3:If telling is Voiced signal, from harddisk memory, obtain sample, the same acoustical signal that extracts is in this point
Sampling, seek out short-time energy;
Step 4:The short-time energy in this shop for the Voiced signal and sample are contrasted in the short-time energy of this point, is judged both
Difference, and then judge whether the acoustical signal collecting is coordinated.
The recognition methodss of described image identifying device comprise the following steps:
Step 1:Pattern recognition device passes through to obtain original image information;
Step 3:Image is to be carried out Image semantic classification, image segmentation, feature to the original image information receiving successively by device
Extract and image recognition.
The method that described image identifying device carries out image segmentation to image comprises the following steps:
Step 1:Image is mapped to fuzzy characteristics domain from spatial domain, concretely comprises the following steps:Set time that gray level'sFirst graphical representation is a fuzzy set, and in collection, each element is the membership function with respect to certain particular gray level, will
This image is mapped as a fuzzy matrix;WhereinSpecific with respect to certain for the gray level of pixel
Gray levelDegree of membership;
Step 2:SetFor maximum gray scale;Defining membership function is:
;
Step 3:WhereinFor type fuzzy factor reciprocal;For exponential type fuzzy factor, therefore this image becomes fuzzy person in servitude
Genus degree matrix;
Step 4:Nonlinear function is adopted in fuzzy spaceConversion, realizes strengthening the contrast of both sides of edges pixel grey scale;;
Wherein,
RecycleBecome enhanced image of changing commanders and data space is become again by fuzzy space.
Step 5:Defining new boundary operator is:
;
Wherein,Edge is carried out to image information using new boundary operator
Extract.
Using above technical scheme, present invention produces following beneficial effect:
1st, real-time update:The robot system of the present invention, in addition to robot body, additionally provides long-range cloud server
To robot provide more new demand servicing it is ensured that robot can obtain up-to-date data message in real time it is ensured that identification accuracy.
2nd, possesses speech recognition:In the robot system of the present invention, there is speech recognition equipment, can be with the language of identifying user
Message number, user directly just can be with the operation of control machine device people by sound.
3rd, possesses image recognition:In the robot system of the present invention, there is pattern recognition device, can be extraneous with Intelligent Recognition
Image information, and feed back to speech recognition equipment, send voice signal and tell user's identification result.
Using above technical scheme, present invention produces following beneficial effect:
1st, real-time update:The robot system of the present invention, in addition to robot body, additionally provides long-range cloud server
To robot provide more new demand servicing it is ensured that robot can obtain up-to-date data message in real time it is ensured that identification accuracy.
2nd, possesses speech recognition:In the robot system of the present invention, there is speech recognition equipment, can be with the language of identifying user
Message number, user directly just can be with the operation of control machine device people by sound.
3rd, possesses image recognition:In the robot system of the present invention, there is pattern recognition device, can be extraneous with Intelligent Recognition
Image information, and feed back to speech recognition equipment, send voice signal and tell user's identification result.
4th, recognition result is accurate:In the speech recognition of the present invention, except energy being asked for signal in time domain, for this energy
Beyond being judged, the amplitude being also directed to signal in frequency domain is asked for and is judged.Judged result according to both carries out synthesis
Pass judgment on, the identification accuracy of acoustical signal and the accuracy evaluated will be greatly improved.Additionally, the pattern recognition device of the present invention
In, image is transformed into fuzzy characteristics domain and is processed, during ensureing treatment effeciency, improve identification accurately simultaneously
Property.
Brief description
Fig. 1 be the present invention a kind of based on the intelligent sound robot system of image recognition technology and the system structure of method
Schematic diagram.
Specific embodiment
All features disclosed in this specification, or disclosed all methods or during step, except mutually exclusive
Feature and/or step beyond, all can combine by any way.
This specification(Including any accessory claim, summary)Disclosed in any feature, unless specifically stated otherwise,
Replaced by other alternative features equivalent or that there is similar purpose.I.e., unless specifically stated otherwise, each feature is a series of
One of equivalent or similar characteristics example.
A kind of intelligent sound robot system based on image recognition technology is provided, system is tied in the embodiment of the present invention 1
Structure is as shown in Figure 1:
A kind of intelligent sound robot system based on image recognition technology is it is characterised in that described system includes:For for machine
Device people provides the cloud server that data updates;Described cloud server signal is connected to for handling machine personal data information
Robot processor;Signal is connected to memorizer for data storage information, is used for identifying described robot processor respectively
The speech recognition equipment of voice and the image processing apparatus for identifying image.
Described speech recognition equipment includes:For obtaining the voice collection device of primary speech signal;Described sound collection
Device signal is connected to the voice recognition chip for being identified to voice;Described voice recognition chip includes:For to adopting
The acoustical signal collecting carries out the sub-frame processing module of sub-frame processing;Signal is connected to for sentencing described sub-frame processing module respectively
Judge module and voice collection device that whether the acoustical signal after disconnected sub-frame processing is coordinated;Described judge module signal connects
In the Fourier transformation module for acoustical signal is carried out with Fourier transformation;Described Fourier transformation module by signal is connected to use
The frequency-domain analysiss module comparing in frequency domain.
Described image identifying device includes:For obtaining the image collecting device of original image signal;Described image gathers
Device signal is connected to the image recognition chip for being identified to image;Described image identification chip includes:For to adopting
The picture signal collecting is identified, and it includes:Image pre-processing module, image segmentation module, characteristic extracting module and image
Identification module;Described image acquisition module signal is connected to image pre-processing module;Described image pretreatment module signal connects
In image segmentation module;Described image segmentation module by signal is connected to characteristic extracting module;Described characteristic extracting module signal is even
It is connected to picture recognition module;Described image identification module signal is connected to dsp processor;It is right that described image pretreatment module is used for
The original image information collecting carries out Image semantic classification, and pretreated image is sent to image segmentation module;Described figure
As segmentation module, for the image after Image semantic classification is carried out image segmentation, the image after Image semantic classification is sent to spy
Levy extraction module;Described characteristic extracting module, for the image after segmentation is carried out feature extraction, by the image after feature extraction
Send to picture recognition module;Described image identification module, for carrying out image recognition to the image after feature extraction.
Described cloud server includes:Cloud database and high in the clouds data transmission device;Described cloud database signal is even
It is connected to high in the clouds data transmission device, for storing template data;Belonging high in the clouds data transmission device, for by cloud database
In template data be sent to the robot processor of robot.
Described robot processor includes:Signal transmission unit and control unit;Described signal transmission unit, for accepting
The data message sending over from high in the clouds data transmission device, the data message receiving is sent and is deposited to memorizer
Storage.
A kind of method of the intelligent sound robot system based on image recognition technology is provided in the embodiment of the present invention 2:
A kind of intelligent sound based on image recognition technology identifies the method for robot system it is characterised in that methods described bag
Include following steps:
Step 1:For sending sound control order, after speech recognition equipment collects this sound control order, from memorizer
Transfer standard voice signals, carry out speech recognition, recognition result is sent to robot processor;
Step 2:According to the recognition result sending over, judgement should carry out any operation to robot processor, if user requires
Robot carries out image recognition, and robot starts startup pattern recognition device and image is identified;
Step 3:After pattern recognition device collects original picture signal, transfer video standard signal from memorizer, to figure
As signal is identified, recognition result is sent to robot processor;Robot processor according to the recognition result receiving,
Transmitting control commands are to speech recognition equipment;
Step 4:The recognition result that speech recognition equipment sends over according to robot processor, sends acoustical signal.
The recognition methodss of described speech recognition equipment comprise the following steps:
Step 1:After voice collection device collects original sound signal, sub-frame processing module is carried out at framing to acoustical signal
Reason;Each frame acoustical signal after sub-frame processing is sent to judge module;
Step 2:Coordinating property of acoustical signal after judge module is processed to each frame judges;After will determine that result is preserved
Send to Fourier transformation module;
Step 3:Fourier transformation module starts to carry out Fourier transformation to acoustical signal, by conversion after acoustical signal send to
Comparison module;
Step 4:Comparison module carries out frequency domain to the acoustical signal after Fourier transformation and compares, and comparative result is sent to memorizer
In kept in;
Step 5:Sub-frame processing module when processing acoustical signal, can real-time judge process be whether acoustical signal last
Frame, if it is transfers the storage result in flash memory and sends and shown to display device;If not then continuing executing with step
1.
The method that described judge module judges to the coordinating property of acoustical signal after sub-frame processing comprises the following steps:
Step 1:Ask for the short-time energy of the signal after sub-frame processing, the acquiring method of described short-time energy adopts equation below:
, whereinIt is the sampling in certain point for the acoustical signal
Signal;
Step 2:Sore throat relieving or Voiced signal are distinguished in short-time energy according to seeking out;
Step 3:If telling is Voiced signal, from harddisk memory, obtain sample, the same acoustical signal that extracts is in this point
Sampling, seek out short-time energy;
Step 4:The short-time energy in this shop for the Voiced signal and sample are contrasted in the short-time energy of this point, is judged both
Difference, and then judge whether the acoustical signal collecting is coordinated.
The recognition methodss of described image identifying device comprise the following steps:
Step 1:Pattern recognition device passes through to obtain original image information;
Step 3:Image is to be carried out Image semantic classification, image segmentation, feature to the original image information receiving successively by device
Extract and image recognition.
The method that described image identifying device carries out image segmentation to image comprises the following steps:
Step 1:Image is mapped to fuzzy characteristics domain from spatial domain, concretely comprises the following steps:Set time that gray level'sFirst graphical representation is a fuzzy set, and in collection, each element is the membership function with respect to certain particular gray level, will
This image is mapped as a fuzzy matrix;WhereinFor pixel gray level with respect to certain
Particular gray levelDegree of membership;
Step 2:SetFor maximum gray scale;Defining membership function is:
;
Step 3:WhereinFor type fuzzy factor reciprocal;For exponential type fuzzy factor, therefore this image becomes fuzzy person in servitude
Genus degree matrix;
Step 4:Nonlinear function is adopted in fuzzy spaceConversion, realizes strengthening the contrast of both sides of edges pixel grey scale;;
Wherein,
RecycleBecome enhanced image of changing commanders and data space is become again by fuzzy space.
Step 5:Defining new boundary operator is:
;
Wherein,Carry out edge using new boundary operator to image information to carry
Take.
A kind of intelligent sound robot system based on image recognition technology and method is provided in the embodiment of the present invention 3,
System construction drawing is as shown in Figure 1:
A kind of intelligent sound robot system based on image recognition technology is it is characterised in that described system includes:For for machine
Device people provides the cloud server that data updates;Described cloud server signal is connected to for handling machine personal data information
Robot processor;Signal is connected to memorizer for data storage information, is used for identifying described robot processor respectively
The speech recognition equipment of voice and the image processing apparatus for identifying image.
Described speech recognition equipment includes:For obtaining the voice collection device of primary speech signal;Described sound collection
Device signal is connected to the voice recognition chip for being identified to voice;Described voice recognition chip includes:For to adopting
The acoustical signal collecting carries out the sub-frame processing module of sub-frame processing;Signal is connected to for sentencing described sub-frame processing module respectively
Judge module and voice collection device that whether the acoustical signal after disconnected sub-frame processing is coordinated;Described judge module signal connects
In the Fourier transformation module for acoustical signal is carried out with Fourier transformation;Described Fourier transformation module by signal is connected to use
The frequency-domain analysiss module comparing in frequency domain.
Described image identifying device includes:For obtaining the image collecting device of original image signal;Described image gathers
Device signal is connected to the image recognition chip for being identified to image;Described image identification chip includes:For to adopting
The picture signal collecting is identified, and it includes:Image pre-processing module, image segmentation module, characteristic extracting module and image
Identification module;Described image acquisition module signal is connected to image pre-processing module;Described image pretreatment module signal connects
In image segmentation module;Described image segmentation module by signal is connected to characteristic extracting module;Described characteristic extracting module signal is even
It is connected to picture recognition module;Described image identification module signal is connected to dsp processor;It is right that described image pretreatment module is used for
The original image information collecting carries out Image semantic classification, and pretreated image is sent to image segmentation module;Described figure
As segmentation module, for the image after Image semantic classification is carried out image segmentation, the image after Image semantic classification is sent to spy
Levy extraction module;Described characteristic extracting module, for the image after segmentation is carried out feature extraction, by the image after feature extraction
Send to picture recognition module;Described image identification module, for carrying out image recognition to the image after feature extraction.
Described cloud server includes:Cloud database and high in the clouds data transmission device;Described cloud database signal is even
It is connected to high in the clouds data transmission device, for storing template data;Belonging high in the clouds data transmission device, for by cloud database
In template data be sent to the robot processor of robot.
Described robot processor includes:Signal transmission unit and control unit;Described signal transmission unit, for accepting
The data message sending over from high in the clouds data transmission device, the data message receiving is sent and is deposited to memorizer
Storage.
A kind of intelligent sound based on image recognition technology identifies the method for robot system it is characterised in that described side
Method comprises the following steps:
Step 1:For sending sound control order, after speech recognition equipment collects this sound control order, from memorizer
Transfer standard voice signals, carry out speech recognition, recognition result is sent to robot processor;
Step 2:According to the recognition result sending over, judgement should carry out any operation to robot processor, if user requires
Robot carries out image recognition, and robot starts startup pattern recognition device and image is identified;
Step 3:After pattern recognition device collects original picture signal, transfer video standard signal from memorizer, to figure
As signal is identified, recognition result is sent to robot processor;Robot processor according to the recognition result receiving,
Transmitting control commands are to speech recognition equipment;
Step 4:The recognition result that speech recognition equipment sends over according to robot processor, sends acoustical signal.
The recognition methodss of described speech recognition equipment comprise the following steps:
Step 1:After voice collection device collects original sound signal, sub-frame processing module is carried out at framing to acoustical signal
Reason;Each frame acoustical signal after sub-frame processing is sent to judge module;
Step 2:Coordinating property of acoustical signal after judge module is processed to each frame judges;After will determine that result is preserved
Send to Fourier transformation module;
Step 3:Fourier transformation module starts to carry out Fourier transformation to acoustical signal, by conversion after acoustical signal send to
Comparison module;
Step 4:Comparison module carries out frequency domain to the acoustical signal after Fourier transformation and compares, and comparative result is sent to memorizer
In kept in;
Step 5:Sub-frame processing module when processing acoustical signal, can real-time judge process be whether acoustical signal last
Frame, if it is transfers the storage result in flash memory and sends and shown to display device;If not then continuing executing with step
1.
The method that described judge module judges to the coordinating property of acoustical signal after sub-frame processing comprises the following steps:
Step 1:Ask for the short-time energy of the signal after sub-frame processing, the acquiring method of described short-time energy adopts equation below:
, whereinIt is that acoustical signal is believed in the sampling of certain point
Number;
Step 2:Sore throat relieving or Voiced signal are distinguished in short-time energy according to seeking out;
Step 3:If telling is Voiced signal, from harddisk memory, obtain sample, the same acoustical signal that extracts is in this point
Sampling, seek out short-time energy;
Step 4:The short-time energy in this shop for the Voiced signal and sample are contrasted in the short-time energy of this point, is judged both
Difference, and then judge whether the acoustical signal collecting is coordinated.
The recognition methodss of described image identifying device comprise the following steps:
Step 1:Pattern recognition device passes through to obtain original image information;
Step 3:Image is to be carried out Image semantic classification, image segmentation, feature to the original image information receiving successively by device
Extract and image recognition.
The method that described image identifying device carries out image segmentation to image comprises the following steps:
Step 1:Image is mapped to fuzzy characteristics domain from spatial domain, concretely comprises the following steps:Set time that gray level'sFirst graphical representation is a fuzzy set, and in collection, each element is the membership function with respect to certain particular gray level, will
This image is mapped as a fuzzy matrix;WhereinFor pixel gray level with respect to certain
Particular gray levelDegree of membership;
Step 2:SetFor maximum gray scale;Defining membership function is:
;
Step 3:WhereinFor type fuzzy factor reciprocal;For exponential type fuzzy factor, therefore this image becomes fuzzy
Subordinated-degree matrix;
Step 4:Nonlinear function is adopted in fuzzy spaceConversion, realizes strengthening the contrast of both sides of edges pixel grey scale
Degree;;
Wherein,
Recycle QUOTEBecome enhanced image of changing commanders and data space is become again by fuzzy space.
Step 5:Defining new boundary operator is:
;
Wherein,Edge is carried out to image information using new boundary operator
Extract.
The invention is not limited in aforesaid specific embodiment.The present invention expands to and any discloses in this manual
New feature or any new combination, and the arbitrary new method of disclosure or the step of process or any new combination.
Claims (10)
1. a kind of intelligent sound robot system based on image recognition technology is it is characterised in that described system includes:For for
Robot provides the cloud server that data updates;Described cloud server signal is connected to for handling machine personal data information
Robot processor;Signal is connected to memorizer for data storage information, is used for knowing described robot processor respectively
The speech recognition equipment of other voice and the image processing apparatus for identifying image.
2. the intelligent sound robot system based on image recognition technology as claimed in claim 1 is it is characterised in that institute's predicate
Sound identifying device includes:For obtaining the voice collection device of primary speech signal;Described voice collection device signal is connected to
For the voice recognition chip that voice is identified;Described voice recognition chip includes:For to the acoustical signal collecting
Carry out the sub-frame processing module of sub-frame processing;Signal is connected to for judging after sub-frame processing described sub-frame processing module respectively
Judge module and voice collection device that whether acoustical signal is coordinated;Described judge module signal is connected to for believing to sound
Number carry out the Fourier transformation module of Fourier transformation;Described Fourier transformation module by signal is connected to the frequency comparing for frequency domain
Domain analyzing module.
3. the intelligent sound identification robot system based on image recognition technology as claimed in claim 1 is it is characterised in that institute
State pattern recognition device to include:For obtaining the image collecting device of original image signal;Described image harvester signal is even
It is connected to the image recognition chip for being identified to image;Described image identification chip includes:For to the image collecting
Signal is identified, and it includes:Image pre-processing module, image segmentation module, characteristic extracting module and picture recognition module;Institute
State image capture module signal and be connected to image pre-processing module;Described image pretreatment module signal is connected to image segmentation mould
Block;Described image segmentation module by signal is connected to characteristic extracting module;Described characteristic extracting module signal is connected to image recognition
Module;Described image identification module signal is connected to dsp processor;Described image pretreatment module is used for original to collect
Image information carries out Image semantic classification, and pretreated image is sent to image segmentation module;Described image splits module, uses
In the image after Image semantic classification is carried out image segmentation, the image after Image semantic classification is sent to characteristic extracting module;Institute
State characteristic extracting module, for the image after segmentation is carried out feature extraction, the image after feature extraction is sent and knows to image
Other module;Described image identification module, for carrying out image recognition to the image after feature extraction.
4. the intelligent sound identification robot system based on image recognition technology as claimed in claim 3 is it is characterised in that institute
State cloud server to include:Cloud database and high in the clouds data transmission device;Described cloud database signal is connected to high in the clouds number
According to transmitting device, for storing template data;Belonging high in the clouds data transmission device, for by the template number in cloud database
According to the robot processor being sent to robot.
5. the intelligent sound identification robot system based on image recognition technology as claimed in claim 4 is it is characterised in that institute
State robot processor to include:Signal transmission unit and control unit;Described signal transmission unit, for accepting to be derived from high in the clouds number
The data message sending over according to transmitting device, the data message receiving is sent and is stored to memorizer.
6. a kind of based on described in one of claim 1 to 5 based on image recognition technology intelligent sound identify robot system
Method it is characterised in that the method comprising the steps of:
Step 1:For sending sound control order, after speech recognition equipment collects this sound control order, from memorizer
Transfer standard voice signals, carry out speech recognition, recognition result is sent to robot processor;
Step 2:According to the recognition result sending over, judgement should carry out any operation to robot processor, if user requires
Robot carries out image recognition, and robot starts startup pattern recognition device and image is identified;
Step 3:After pattern recognition device collects original picture signal, transfer video standard signal from memorizer, to figure
As signal is identified, recognition result is sent to robot processor;Robot processor according to the recognition result receiving,
Transmitting control commands are to speech recognition equipment;
Step 4:The recognition result that speech recognition equipment sends over according to robot processor, sends acoustical signal.
7. the method that the intelligent sound based on image recognition technology as claimed in claim 6 identifies robot system, its feature
It is, the recognition methodss of described speech recognition equipment comprise the following steps:
Step 1:After voice collection device collects original sound signal, sub-frame processing module is carried out at framing to acoustical signal
Reason;Each frame acoustical signal after sub-frame processing is sent to judge module;
Step 2:Coordinating property of acoustical signal after judge module is processed to each frame judges;After will determine that result is preserved
Send to Fourier transformation module;
Step 3:Fourier transformation module starts to carry out Fourier transformation to acoustical signal, by conversion after acoustical signal send to
Comparison module;
Step 4:Comparison module carries out frequency domain to the acoustical signal after Fourier transformation and compares, and comparative result is sent to memorizer
In kept in;
Step 5:Sub-frame processing module when processing acoustical signal, can real-time judge process be whether acoustical signal last
Frame, if it is transfers the storage result in flash memory and sends and shown to display device;If not then continuing executing with step
1.
8. the method that the intelligent sound based on image recognition technology as claimed in claim 7 identifies robot system, its feature
It is, the method that described judge module judges to the coordinating property of acoustical signal after sub-frame processing comprises the following steps:
Step 1:Ask for the short-time energy of the signal after sub-frame processing, the acquiring method of described short-time energy adopts equation below:
, whereinIt is the sampled signal in certain point for the acoustical signal;
Step 2:Sore throat relieving or Voiced signal are distinguished in short-time energy according to seeking out;
Step 3:If telling is Voiced signal, from harddisk memory, obtain sample, the same acoustical signal that extracts is in this point
Sampling, seek out short-time energy;
Step 4:The short-time energy in this shop for the Voiced signal and sample are contrasted in the short-time energy of this point, is judged both
Difference, and then judge whether the acoustical signal collecting is coordinated.
9. the method that the intelligent sound based on image recognition technology as claimed in claim 6 identifies robot system, its feature
It is, the recognition methodss of described image identifying device comprise the following steps:
Step 1:Pattern recognition device passes through to obtain original image information;
Step 3:Image is to be carried out Image semantic classification, image segmentation, feature to the original image information receiving successively by device
Extract and image recognition.
10. the method that the intelligent sound based on image recognition technology as claimed in claim 8 identifies robot system, its feature
It is, the method that described image identifying device carries out image segmentation to image comprises the following steps:
Step 1:Image is mapped to fuzzy characteristics domain from spatial domain, concretely comprises the following steps:Set time that gray level'sFirst graphical representation is a fuzzy set, and in collection, each element is the membership function with respect to certain particular gray level,
This image is mapped as a fuzzy matrix;WhereinFor pixel gray level with respect to certain
Particular gray levelDegree of membership;
Step 2:SetFor maximum gray scale;Defining membership function is:
;
Step 3:WhereinFor type fuzzy factor reciprocal;For exponential type fuzzy factor, therefore this image becomes fuzzy person in servitude
Genus degree matrix;
Step 4:Nonlinear function is adopted in fuzzy spaceConversion, realizes strengthening the contrast of both sides of edges pixel grey scale;;
Wherein,
RecycleBecome enhanced image of changing commanders and data space is become again by fuzzy space;
Step 5:Defining new boundary operator is:
;
Wherein,Carry out edge using new boundary operator to image information to carry
Take.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610818330.0A CN106407993A (en) | 2016-09-13 | 2016-09-13 | Intelligent voice robot system based on image recognition technology and method thereof |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610818330.0A CN106407993A (en) | 2016-09-13 | 2016-09-13 | Intelligent voice robot system based on image recognition technology and method thereof |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106407993A true CN106407993A (en) | 2017-02-15 |
Family
ID=57999761
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610818330.0A Withdrawn CN106407993A (en) | 2016-09-13 | 2016-09-13 | Intelligent voice robot system based on image recognition technology and method thereof |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106407993A (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107127766A (en) * | 2017-05-24 | 2017-09-05 | 南京华设科技股份有限公司 | Intelligent grid service handling robot |
CN107358949A (en) * | 2017-05-27 | 2017-11-17 | 芜湖星途机器人科技有限公司 | Robot sounding automatic adjustment system |
CN107945803A (en) * | 2017-11-28 | 2018-04-20 | 上海与德科技有限公司 | The assisted learning method and robot of a kind of robot |
CN109087646A (en) * | 2018-10-25 | 2018-12-25 | 武汉拓睿传奇科技有限公司 | A kind of importing artificial intelligence is ultra-deep to be learnt to know method for distinguishing for phonetic image |
CN110091302A (en) * | 2019-05-10 | 2019-08-06 | 东方电子股份有限公司 | A kind of improved intelligent robot of image recognition technology based on deep learning |
CN110163028A (en) * | 2018-01-17 | 2019-08-23 | 黄冈职业技术学院 | A kind of precise image identifying system and image-recognizing method |
WO2019201346A1 (en) * | 2018-04-19 | 2019-10-24 | 苏州宝时得电动工具有限公司 | Self-moving device, server, and automatic working system therefor |
CN111086008A (en) * | 2018-10-24 | 2020-05-01 | 国网河南省电力公司南阳供电公司 | Electric power safety knowledge learning robot and method for preventing electric power operation fault |
CN112530434A (en) * | 2020-12-21 | 2021-03-19 | 云南电网有限责任公司玉溪供电局 | Automatic intelligent robot on duty scheduling system of power station |
CN114025327A (en) * | 2021-10-25 | 2022-02-08 | 天津市普迅电力信息技术有限公司 | Motorcade management robot design method based on data middleboxes |
-
2016
- 2016-09-13 CN CN201610818330.0A patent/CN106407993A/en not_active Withdrawn
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107127766A (en) * | 2017-05-24 | 2017-09-05 | 南京华设科技股份有限公司 | Intelligent grid service handling robot |
CN107358949A (en) * | 2017-05-27 | 2017-11-17 | 芜湖星途机器人科技有限公司 | Robot sounding automatic adjustment system |
CN107945803A (en) * | 2017-11-28 | 2018-04-20 | 上海与德科技有限公司 | The assisted learning method and robot of a kind of robot |
CN110163028A (en) * | 2018-01-17 | 2019-08-23 | 黄冈职业技术学院 | A kind of precise image identifying system and image-recognizing method |
WO2019201346A1 (en) * | 2018-04-19 | 2019-10-24 | 苏州宝时得电动工具有限公司 | Self-moving device, server, and automatic working system therefor |
CN111086008A (en) * | 2018-10-24 | 2020-05-01 | 国网河南省电力公司南阳供电公司 | Electric power safety knowledge learning robot and method for preventing electric power operation fault |
CN109087646A (en) * | 2018-10-25 | 2018-12-25 | 武汉拓睿传奇科技有限公司 | A kind of importing artificial intelligence is ultra-deep to be learnt to know method for distinguishing for phonetic image |
CN109087646B (en) * | 2018-10-25 | 2021-04-06 | 武汉拓睿传奇科技有限公司 | Method for leading-in artificial intelligence ultra-deep learning for voice image recognition |
CN110091302A (en) * | 2019-05-10 | 2019-08-06 | 东方电子股份有限公司 | A kind of improved intelligent robot of image recognition technology based on deep learning |
CN110091302B (en) * | 2019-05-10 | 2021-06-22 | 北京之乐时空科技有限公司 | Intelligent robot with improved image recognition technology based on deep learning |
CN112530434A (en) * | 2020-12-21 | 2021-03-19 | 云南电网有限责任公司玉溪供电局 | Automatic intelligent robot on duty scheduling system of power station |
CN114025327A (en) * | 2021-10-25 | 2022-02-08 | 天津市普迅电力信息技术有限公司 | Motorcade management robot design method based on data middleboxes |
CN114025327B (en) * | 2021-10-25 | 2024-03-26 | 天津市普迅电力信息技术有限公司 | Motorcade management robot design method based on data center platform |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106407993A (en) | Intelligent voice robot system based on image recognition technology and method thereof | |
CN107799126B (en) | Voice endpoint detection method and device based on supervised machine learning | |
US11830479B2 (en) | Voice recognition method and apparatus, and air conditioner | |
CN107133612A (en) | Based on image procossing and the intelligent ward of speech recognition technology and its operation method | |
CN108090458A (en) | Tumble detection method for human body and device | |
CN106157956A (en) | The method and device of speech recognition | |
CN113361636B (en) | Image classification method, system, medium and electronic device | |
Jain et al. | Visual assistance for blind using image processing | |
CN110972112B (en) | Subway running direction determining method, device, terminal and storage medium | |
CN112541529A (en) | Expression and posture fusion bimodal teaching evaluation method, device and storage medium | |
CN101674363A (en) | Mobile equipment and talking method | |
CN104616002A (en) | Facial recognition equipment used for judging age groups | |
CN113936340B (en) | AI model training method and device based on training data acquisition | |
CN109584864B (en) | Image processing apparatus and method | |
CN111383138A (en) | Catering data processing method and device, computer equipment and storage medium | |
CN211512572U (en) | Interactive blind guiding system | |
CN115601818B (en) | Lightweight visible light living body detection method and device | |
CN110738985A (en) | Cross-modal biometric feature recognition method and system based on voice signals | |
CN109785843B (en) | Image processing apparatus and method | |
CN113420783B (en) | Intelligent man-machine interaction method and device based on image-text matching | |
CN115116117A (en) | Learning input data acquisition method based on multi-mode fusion network | |
CN114463667A (en) | Small sample learning method based on video identification | |
CN113276113A (en) | Sight line positioning and voice control system and method for space manipulator on-orbit operation | |
CN111414884A (en) | Facial expression recognition method based on edge calculation | |
Shane et al. | Sign Language Detection Using Faster RCNN Resnet |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WW01 | Invention patent application withdrawn after publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20170215 |