CN111428769A - Artificial intelligence translation system for designing pet behavior language by software - Google Patents
Artificial intelligence translation system for designing pet behavior language by software Download PDFInfo
- Publication number
- CN111428769A CN111428769A CN202010190580.0A CN202010190580A CN111428769A CN 111428769 A CN111428769 A CN 111428769A CN 202010190580 A CN202010190580 A CN 202010190580A CN 111428769 A CN111428769 A CN 111428769A
- Authority
- CN
- China
- Prior art keywords
- model database
- neural network
- information
- artificial intelligence
- pet
- 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
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 20
- 238000013519 translation Methods 0.000 title claims abstract description 11
- 238000012545 processing Methods 0.000 claims abstract description 48
- 238000005516 engineering process Methods 0.000 claims abstract description 18
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 13
- 238000013528 artificial neural network Methods 0.000 claims abstract description 12
- 238000000513 principal component analysis Methods 0.000 claims abstract description 12
- 230000006870 function Effects 0.000 claims abstract description 11
- 230000006399 behavior Effects 0.000 claims description 37
- 210000000056 organ Anatomy 0.000 claims description 22
- 230000014509 gene expression Effects 0.000 claims description 20
- 241000282472 Canis lupus familiaris Species 0.000 claims description 11
- 241000282326 Felis catus Species 0.000 claims description 11
- 230000008859 change Effects 0.000 claims description 11
- 239000012634 fragment Substances 0.000 claims description 9
- 238000007711 solidification Methods 0.000 claims description 8
- 230000008023 solidification Effects 0.000 claims description 8
- 230000009471 action Effects 0.000 claims description 6
- 210000005069 ears Anatomy 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 5
- 230000008094 contradictory effect Effects 0.000 claims description 4
- 230000008909 emotion recognition Effects 0.000 claims description 4
- 238000012847 principal component analysis method Methods 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 4
- 241000894007 species Species 0.000 claims description 4
- 230000003542 behavioural effect Effects 0.000 claims description 3
- 210000001508 eye Anatomy 0.000 claims description 3
- 230000009191 jumping Effects 0.000 claims description 3
- 238000000034 method Methods 0.000 claims description 3
- 210000002105 tongue Anatomy 0.000 claims description 3
- 241001465754 Metazoa Species 0.000 claims description 2
- 210000003128 head Anatomy 0.000 claims description 2
- 230000009467 reduction Effects 0.000 claims description 2
- 238000012958 reprocessing Methods 0.000 claims description 2
- 238000003745 diagnosis Methods 0.000 description 5
- 238000011160 research Methods 0.000 description 5
- 238000011282 treatment Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000002996 emotional effect Effects 0.000 description 4
- 210000001331 nose Anatomy 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 206010063385 Intellectualisation Diseases 0.000 description 2
- 238000012512 characterization method Methods 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 210000000088 lip Anatomy 0.000 description 2
- 238000006748 scratching Methods 0.000 description 2
- 230000002393 scratching effect Effects 0.000 description 2
- 210000000515 tooth Anatomy 0.000 description 2
- 230000036626 alertness Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000003759 clinical diagnosis Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000005236 sound signal Effects 0.000 description 1
- 210000000538 tail Anatomy 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L17/00—Speaker identification or verification techniques
- G10L17/26—Recognition of special voice characteristics, e.g. for use in lie detectors; Recognition of animal voices
-
- 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
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Biophysics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Evolutionary Biology (AREA)
- Biomedical Technology (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Probability & Statistics with Applications (AREA)
- Signal Processing (AREA)
- Child & Adolescent Psychology (AREA)
- Hospice & Palliative Care (AREA)
- Psychiatry (AREA)
- Machine Translation (AREA)
Abstract
The invention relates to an artificial intelligence translation system for designing pet behavior language by software, which belongs to the field of software design and artificial intelligence and comprises the following parts: the system comprises a user side, a data processing side, a classification processing side, an information summarizing side and a result expressing side. The system for intelligently analyzing the pet behavior language by adopting the convolutional neural network technology can realize more efficient and accurate recognition and translation of the pet behavior language by adopting a classification processing technology of a Principal Component Analysis (PCA), a deep belief neural network (DBN) and a SoftMax function.
Description
Technical Field
The invention relates to an artificial intelligence translation system for designing pet behavior language by software, belonging to the field of software design and artificial intelligence.
Background
Aiming at the research of pet behaviours, domestic research starts later than abroad, the research on pet behaviours is started in recent years, and at present, artificial intelligence research patents and software products of pet behavioural languages are not available at home and abroad, so the invention has originality. The research of the pet behavior language has very important significance for deepening communication between a pet owner and a pet, pet training and diagnosis and treatment of pet psychological diseases, but the pet behavior language is developed late in China, has poor theoretical basis and high learning difficulty, and limits the application of the pet behavior language in life and clinical diagnosis and treatment; secondly, the difficulty of applying the computer technology to pet behavior languages is high, and the traditional artificial intelligence technology needs a large amount of video databases or picture databases and is difficult to realize intellectualization, so that artificial intelligence software of relevant pet behavior languages is not available at home and abroad.
The current common neural network identification method at home and abroad is 2DCNN, which mainly cuts a video image into each frame and then uses 2DCNN neural network to identify motion information interaction between frames under the time dimension of a model, and has the defect of low identification accuracy; the convolution layer of the convolutional neural network technology is characterized in that images of three continuous frames are selected for convolutional identification, single-frame image identification of 2DCNN is replaced, connection between each feature map and the previous frame image is guaranteed, motion information is captured, accuracy is high, and the technology is advanced.
Because the audio signals have the characteristics of high dimensionality and high redundancy, in a traditional audio emotion recognition model, audio processing is often required to be performed on audio, appropriate audio features are extracted according to actual needs, and finally model training is performed on the extracted features through modeling, so that the effect of audio emotion recognition is finally achieved, and therefore the recognition accuracy is low, the efficiency is low, and practical application is difficult.
Disclosure of Invention
Based on the above mentioned problems mentioned in the background, the present invention is intended to provide an artificial intelligence translation system for pet behavioral language to solve the above mentioned problems.
The system of the invention is realized by the following steps:
the method comprises the following steps that firstly, a user shoots a video of action behaviors of dogs and cats for a limited time or selects a shot video of the behavior of dogs and cats from an album according to software operation prompts by using an intelligent terminal such as a mobile phone and the like, and submits the video for uploading; in the second step, the data processing end mainly has three functions, and the sequence is not in sequence. Firstly, extracting organ change representations of eyes, a nose, ears, tongues, teeth, lips and tails of uploaded videos according to set organ specific points; secondly, extracting the characteristics of sitting, standing, lying, creeping and jumping changes according to the set special points of the movement; finally, according to the set sound special points, extracting the representation of pitch, tone, time length and time frequency change; and thirdly, the classification processing end is a curing model database which comprises three databases of an organ curing model database, a motion model database and a sound model database. Matching the characterization of organs, behaviors and sounds collected by the data processing end with a curing model database of the classification processing end according to a set idiosyncratic point analysis method to obtain an operation result of the curing model database, wherein the operation result is segment type information; fourthly, the information summarizing end summarizes the large amount of fragment type information, deletes abrupt and contradictory vocabularies, and establishes vocabulary logic to make sentence expressions to obtain processing results; fifthly, the result expression end correspondingly matches the male voice, the female voice and the child voice modes according to the age and sex information of the pet provided by the user registration, and carries out voice expression on the processing result of the information summarizing end.
The technical scheme of the invention is as follows:
an artificial intelligence translation system for designing pet behavior language by software comprises the following parts: user end, data processing end, classification processing end, information summarizing end and result expression end
The user side is as follows: the system is arranged on intelligent terminals such as mobile phones, tablet computers and computers in the form of mobile phone APP, WeChat public numbers and computer software, and a user correspondingly provides information related to the species, age, sex and living environment and living habits of pets according to software registration operation prompts; and the user shoots videos of the behavior of the dog and the cat for a limited time by using an intelligent terminal such as a mobile phone and the like according to the software operation prompt or selects the shot videos of the behavior of the dog and the cat from the photo album, and submits the videos to be uploaded as video information.
The data processing end: carrying out classification processing through an artificial intelligence technology of a convolutional neural network, a principal component analysis method, a deep belief neural network and a SoftMax function, extracting change representations of organs, sounds and behaviors, and establishing a generation analysis object, which is specifically embodied in that: extracting organ change representations of the head, eyes, nose, ears, tongue, teeth, lips and tail of the uploaded video according to set organ specific points; extracting the representations of sitting, standing, lying, creeping and jumping changes of the uploaded video according to set movement special points; and extracting representations of pitch, tone, time length and time frequency change of the uploaded video according to the set sound special points.
The classification processing end is as follows: and the classification processing end is a curing model database and comprises three databases of an organ curing model database, a motion model database and a sound model database, organ, behavior and sound characteristics collected by the data processing end are used as processing objects, and the curing model database of the classification processing end is matched according to a set singularity analysis method to obtain an operation result of the curing model database.
The information summarizing end: the information collecting end is a language processing module, reprocessing the operation result of the solidification model database of the classification processing end, establishing a logic relation and expressing the logic relation through sentence making, the operation result of the solidification model database is fragment type information, the information collecting end collects the large amount of fragment type information, deletes abrupt and contradictory words and words, and establishes word logic to express the sentence making to obtain a processing result.
The result expression end: the result expression end is a language matching module, and correspondingly matches male voice, female voice and child voice modes according to the pet type, age and gender information provided by the user end registration, and carries out voice expression on the processing result of the information summarizing end.
The establishment of the organ solidification model database and the motion model database is realized by the artificial intelligence technology of a convolutional neural network; the acoustic model database is implemented by using Principal Component Analysis (PCA), deep belief neural network (DBN) and SoftMax functions for classification processing.
The artificial intelligence technology of the convolutional neural network specifically comprises the following steps:
(1) firstly, a convolutional layer of a convolutional neural network selects images of three continuous frames to carry out convolutional image recognition, and each characteristic map is ensured to be connected with the image of the previous frame so as to capture motion information;
(2) taking the video continuous frame image with the size of 60 × 40 as an input layer element, and simultaneously performing convolution operation by using a 3D convolution sum with the size of 7 × 3, wherein 7 × 7 identifies a space dimension, and 7 × 3 represents a convolution kernel using three space dimensions;
(3) then adopting 23 × 2max boosting operation to perform down sampling on the data;
(4) and finally, performing convolution operation through a convolution kernel of 7 × 6 × 3, wherein the convolution operation value is very small, so that the data is operated by adopting a down-sampling layer with the size of 7 × 4, and finally, the set organ behaviors and action behaviors of the animal are identified.
The principal component analysis method, the deep belief neural network and the SoftMax function specifically comprise the following steps
(1) Performing audio identification through a short-time average energy algorithm;
(2) thirdly, carrying out dimensionality reduction on redundant noise of the audio data by a Principal Component Analysis (PCA);
(3) inputting the processed audio data into a deep belief neural network (DBN) for training;
(4) and finally, calculating the audio data through a SoftMax function classification processing technology, thereby completing audio emotion recognition.
The pet species that can be analyzed include dogs and cats.
The invention has the beneficial effects that:
(1) the invention adopts the first case at home and abroad to realize a system for intelligently analyzing pet behavior languages by adopting a convolutional neural network technology, and a convolutional layer of the convolutional neural network technology selects images of three continuous frames for convolutional identification to replace single-frame image identification of 2DCNN, so that each characteristic map can be connected with the previous frame image to capture motion information, the accuracy is higher, and the technology is more advanced.
(2) The invention adopts the first case of home and abroad to adopt a Principal Component Analysis (PCA), a deep belief neural network (DBN) and a SoftMax function to carry out classification processing technology to realize a system for intelligently analyzing pet behavior language, gradually and autonomously learns the signal characteristics, continuously optimizes the characteristic of adjusting parameters and can realize more efficient and more accurate voice recognition.
(3) The implementation of artificial intelligence by pet behavior language has five main benefits: firstly, the learning difficulty of pet behaviours is high, most people are difficult to learn and apply, and the intellectualization is realized, so that the application threshold can be reduced; secondly, the user only needs to record a video to the pet by using a multimedia terminal such as a mobile phone, a tablet, a computer and the like, so that the language expressed by the behavior of the pet can be quickly identified, and real artificial intelligence is realized; thirdly, the communication between the breeder and the pet is facilitated, the breeder is helped to know the expression of the pet immediately, and the feelings of the two parties are improved; the pet training work is facilitated, most of ordinary feeders can be helped to train the pet, and the threshold difficulty of pet training is reduced; and because the pet psychological diagnosis and treatment aiming at the pet is less in China and the accuracy is low, the pet psychological diagnosis and treatment and the pet non-positive behavior correction diagnosis and treatment are intelligently facilitated for pet doctors, and the accuracy of the psychological disease diagnosis is facilitated to be improved.
Drawings
FIG. 1 is a schematic diagram of the operation of the system of the present invention.
Fig. 2 is a schematic operation diagram of the data processing terminal according to the present invention.
Fig. 3 is a schematic diagram of the classification processing end of the present invention.
FIG. 4 is a schematic roadmap for convolutional neural network technology
Fig. 5 is a technical route diagram of the sound extraction process.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1
A user uses an intelligent terminal such as a mobile phone and the like to shoot a section of video of 15s for behavior actions of a dog, and submits the video for uploading; the data processing end extracts the organ representation with changes according to the set organ specific points of the uploaded video, for example: squinting, licking the mouth and nose, lying down the ears, lying in a semi-lying posture and scratching the body; the solidification model database at the classification processing end obtains the fragment type information of the change, such as squinting corresponding contradiction, evasion and error identification; licking the mouth and nose, showing love, error and pressure; the groveling ears correspond to fear, fear and obedience; half lying corresponds to comfort, invitation, obedience, contradiction; scratching the body corresponds to contradiction, obedience, weakness and the like; the information collecting end deletes the fragment information and matches the information, deletes the abrupt and contradictory words, retains the similar or same words, establishes the emotional logic to make sentences to express meaning, obtains the dog's emotional conflict, and wants to be punished and displayed wrongly; the result expression end carries out voice expression by matching the expression meaning information with a corresponding voice mode, such as 'owner, I know that you are misled to forgive but cannot beat I'! ".
Example 2
A user uses an intelligent terminal such as a mobile phone and the like to shoot a section of 15s of video for behavior actions of the cat, and submits the video for uploading; the data processing end extracts the organ change characterization with changes, for example: groveling ears, direct vision, standing tail, standing posture, arching waist and back and sharp sound; matching the solidification model database of the classification processing end with changed organ representations such as groveling corresponding fear, fear and obedience; direct vision corresponds to alertness, governance, and the like; tail erection represents governance, attack, threat, etc.; the sound sharp represents fear, threat, scare, expectation, pain and the like, and the fragment type information of the change is obtained; the information collecting end deletes the fragment information and matches the information, and an emotional logic is established to make sentences to express meanings, so that the emotion of fear in the cat is obtained, behavior expression is that sound is threatened, and threatening attack is wanted; the result expression end carries out voice expression by matching the expression meaning information with the corresponding voice mode, such as "I is afraid of, don't go over! Then I will bite you! ".
The user uploads videos of behavior actions of the dog or the cat by using an intelligent terminal such as a mobile phone and the like, and obtains corresponding emotional sound expression and character information through translation processing of an artificial intelligent translation system of the pet behavior language, so that the user can understand the expression of the pet to the owner without any obstacle, and the communication obstacle between the user and the pet is eliminated.
Claims (10)
1. An artificial intelligence translation system for designing pet behavior language by software is characterized by comprising the following parts: the system comprises a user side, a data processing side, a classification processing side, an information summarizing side and a result expressing side.
2. The user end according to claim 1, wherein: the system is arranged on intelligent terminals such as mobile phones, tablet computers and computers in the form of mobile phone APP, WeChat public numbers and computer software, and a user correspondingly provides information related to the species, age, sex and living environment and living habits of pets according to software registration operation prompts; and the user shoots videos of the behavior of the dog and the cat for a limited time by using an intelligent terminal such as a mobile phone and the like according to the software operation prompt or selects the shot videos of the behavior of the dog and the cat from the photo album, and submits the videos to be uploaded as video information.
3. The data processing terminal according to claim 1, wherein: carrying out classification processing through an artificial intelligence technology of a convolutional neural network, a principal component analysis method, a deep belief neural network and a SoftMax function, extracting change representations of organs, sounds and behaviors, and establishing a generation analysis object, which is specifically embodied in that: extracting organ change representations of the head, eyes, nose, ears, tongue, teeth, lips and tail of the uploaded video according to set organ specific points; extracting the representations of sitting, standing, lying, creeping and jumping changes of the uploaded video according to set movement special points; and extracting representations of pitch, tone, time length and time frequency change of the uploaded video according to the set sound special points.
4. The classification processing terminal according to claim 1, wherein: and the classification processing end is a curing model database and comprises three databases of an organ curing model database, a motion model database and a sound model database, organ, behavior and sound characteristics collected by the data processing end are used as processing objects, and the curing model database of the classification processing end is matched according to a set singularity analysis method to obtain an operation result of the curing model database.
5. The information summarizing terminal of claim 1, wherein: the information collecting end is a language processing module, reprocessing the operation result of the solidification model database of the classification processing end, establishing a logic relation and expressing the logic relation through sentence making, the operation result of the solidification model database is fragment type information, the information collecting end collects the large amount of fragment type information, deletes abrupt and contradictory words and words, and establishes word logic to express the sentence making to obtain a processing result.
6. The result expression terminal according to claim 1, wherein: the result expression end is a language matching module, and correspondingly matches male voice, female voice and child voice modes according to the pet type, age and gender information provided by the user end registration, and carries out voice expression on the processing result of the information summarizing end.
7. The establishment of the organ solidification model database and the motion model database according to claim 4 is realized by an artificial intelligence technology of a convolutional neural network; the acoustic model database is implemented by using Principal Component Analysis (PCA), deep belief neural network (DBN) and SoftMax functions for classification processing.
8. An artificial intelligence technique for a convolutional neural network as claimed in claim 3 or claim 7, characterized by the steps of:
(1) firstly, a convolutional layer of a convolutional neural network selects images of three continuous frames to carry out convolutional image recognition, and each characteristic map is ensured to be connected with the image of the previous frame so as to capture motion information;
(2) taking the video continuous frame image with the size of 60 × 40 as an input layer element, and simultaneously performing convolution operation by using a 3D convolution sum with the size of 7 × 3, wherein 7 × 7 identifies a space dimension, and 7 × 3 represents a convolution kernel using three space dimensions;
(3) then adopting 23 × 2max boosting operation to perform down sampling on the data;
(4) and finally, performing convolution operation through a convolution kernel of 7 × 6 × 3, wherein the convolution operation value is very small, so that the data is operated by adopting a down-sampling layer with the size of 7 × 4, and finally, the set organ behaviors and action behaviors of the animal are identified.
9. The principal component analysis method, the deep belief neural network, and the SoftMax function according to claim 3 or claim 7, characterized in that it specifically consists of (1) performing audio recognition by a short-time average energy algorithm;
(2) thirdly, carrying out dimensionality reduction on redundant noise of the audio data by a Principal Component Analysis (PCA);
(3) inputting the processed audio data into a deep belief neural network (DBN) for training;
(4) and finally, calculating the audio data through a SoftMax function classification processing technology, thereby completing audio emotion recognition.
10. Use of the artificial intelligence translation system for a software designed pet behavioral language according to claim 1, wherein: the pet species that can be analyzed include dogs and cats.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010190580.0A CN111428769A (en) | 2020-03-18 | 2020-03-18 | Artificial intelligence translation system for designing pet behavior language by software |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010190580.0A CN111428769A (en) | 2020-03-18 | 2020-03-18 | Artificial intelligence translation system for designing pet behavior language by software |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111428769A true CN111428769A (en) | 2020-07-17 |
Family
ID=71548085
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010190580.0A Pending CN111428769A (en) | 2020-03-18 | 2020-03-18 | Artificial intelligence translation system for designing pet behavior language by software |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111428769A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112507157A (en) * | 2020-11-28 | 2021-03-16 | 爱荔枝科技(北京)有限公司 | Translation method between animals |
| CN113208595A (en) * | 2021-07-10 | 2021-08-06 | 南京蓝洋智能科技有限公司 | Biological expression analysis method, device, storage medium and electronic equipment |
| CN114023337A (en) * | 2021-10-12 | 2022-02-08 | 湖北文理学院 | End-to-end pet language translation method in deep learning |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108600701A (en) * | 2018-05-02 | 2018-09-28 | 广州飞宇智能科技有限公司 | A kind of monitoring system and method judging video behavior based on deep learning |
| CN108766433A (en) * | 2018-05-11 | 2018-11-06 | 深圳双猴科技有限公司 | A kind of body language translation system and method |
| KR20190016365A (en) * | 2017-08-08 | 2019-02-18 | 주식회사 창의산업 | Communication Apparatus between Pets and People |
| US20190213401A1 (en) * | 2018-01-06 | 2019-07-11 | Shaobo Kuang | System and method of social networks for dogs or other pets |
| CN110110707A (en) * | 2019-05-24 | 2019-08-09 | 苏州闪驰数控系统集成有限公司 | Artificial intelligence CNN, LSTM neural network dynamic identifying system |
-
2020
- 2020-03-18 CN CN202010190580.0A patent/CN111428769A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20190016365A (en) * | 2017-08-08 | 2019-02-18 | 주식회사 창의산업 | Communication Apparatus between Pets and People |
| US20190213401A1 (en) * | 2018-01-06 | 2019-07-11 | Shaobo Kuang | System and method of social networks for dogs or other pets |
| CN108600701A (en) * | 2018-05-02 | 2018-09-28 | 广州飞宇智能科技有限公司 | A kind of monitoring system and method judging video behavior based on deep learning |
| CN108766433A (en) * | 2018-05-11 | 2018-11-06 | 深圳双猴科技有限公司 | A kind of body language translation system and method |
| CN110110707A (en) * | 2019-05-24 | 2019-08-09 | 苏州闪驰数控系统集成有限公司 | Artificial intelligence CNN, LSTM neural network dynamic identifying system |
Non-Patent Citations (1)
| Title |
|---|
| 萌宠君: "《人类想听懂动物的语言，AI能帮上忙么？》", pages: 1 - 2 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112507157A (en) * | 2020-11-28 | 2021-03-16 | 爱荔枝科技(北京)有限公司 | Translation method between animals |
| CN113208595A (en) * | 2021-07-10 | 2021-08-06 | 南京蓝洋智能科技有限公司 | Biological expression analysis method, device, storage medium and electronic equipment |
| CN114023337A (en) * | 2021-10-12 | 2022-02-08 | 湖北文理学院 | End-to-end pet language translation method in deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111428769A (en) | Artificial intelligence translation system for designing pet behavior language by software | |
| Zhang et al. | Intelligent facial emotion recognition and semantic-based topic detection for a humanoid robot | |
| US10878818B2 (en) | Methods and apparatus for silent speech interface | |
| WO2019237708A1 (en) | Interpersonal interaction body language automatic generation method and system based on deep learning | |
| CN111496802A (en) | Control method, device, equipment and medium for artificial intelligence equipment | |
| CN112686048B (en) | Emotion recognition method and device based on fusion of voice, semantics and facial expressions | |
| CN113724882B (en) | Method, device, equipment and medium for constructing user portrait based on inquiry session | |
| Kabani et al. | Emotion based music player | |
| CN110210416B (en) | Sign language recognition system optimization method and device based on dynamic pseudo tag decoding | |
| CN104299225A (en) | Method and system for applying facial expression recognition in big data analysis | |
| WO2023226239A1 (en) | Object emotion analysis method and apparatus and electronic device | |
| Verkholyak et al. | Modeling short-term and long-term dependencies of the speech signal for paralinguistic emotion classification | |
| CN114550057A (en) | Video emotion recognition method based on multi-modal representation learning | |
| CN111092798B (en) | Wearable system based on spoken language understanding | |
| da Silva et al. | Recognition of affective and grammatical facial expressions: a study for Brazilian sign language | |
| Gladys et al. | Survey on multimodal approaches to emotion recognition | |
| Zhang et al. | Multimodal Sensing for Depression Risk Detection: Integrating Audio, Video, and Text Data | |
| CN116383360A (en) | Method and system for detecting answer body fitness of psychological consultation chat robot | |
| US11658928B2 (en) | Virtual content creation method | |
| CN118214735A (en) | Dialog generation method and device, computer readable medium and electronic equipment | |
| Liu et al. | A robust multi-modal emotion recognition framework for intelligent tutoring systems | |
| Mishra et al. | Environment descriptor for the visually impaired | |
| CN111145851A (en) | Mental state monitoring and evaluating system based on intelligent bracelet | |
| CN114533063B (en) | Multi-source monitoring combined emotion computing system and method | |
| CN113580166B (en) | Interaction method, device, equipment and storage medium of anthropomorphic robot |
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 |