CN111950721A - Flavor identification method based on smell-taste joint perception model - Google Patents
Flavor identification method based on smell-taste joint perception model Download PDFInfo
- Publication number
- CN111950721A CN111950721A CN202011070975.3A CN202011070975A CN111950721A CN 111950721 A CN111950721 A CN 111950721A CN 202011070975 A CN202011070975 A CN 202011070975A CN 111950721 A CN111950721 A CN 111950721A
- Authority
- CN
- China
- Prior art keywords
- module
- node
- nodes
- taste
- olfactory
- 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.)
- Granted
Links
- 230000008447 perception Effects 0.000 title claims abstract description 79
- 239000000796 flavoring agent Substances 0.000 title claims abstract description 58
- 235000019634 flavors Nutrition 0.000 title claims abstract description 56
- 238000000034 method Methods 0.000 title claims abstract description 47
- 210000001103 thalamus Anatomy 0.000 claims abstract description 25
- 230000007830 nerve conduction Effects 0.000 claims abstract description 9
- 210000001679 solitary nucleus Anatomy 0.000 claims abstract description 5
- 238000011426 transformation method Methods 0.000 claims abstract description 4
- 230000014860 sensory perception of taste Effects 0.000 claims description 26
- 235000019640 taste Nutrition 0.000 claims description 25
- 230000001339 gustatory effect Effects 0.000 claims description 16
- 210000003901 trigeminal nerve Anatomy 0.000 claims description 11
- 210000000956 olfactory bulb Anatomy 0.000 claims description 9
- 230000037361 pathway Effects 0.000 claims description 9
- 210000003792 cranial nerve Anatomy 0.000 claims description 7
- 210000000256 facial nerve Anatomy 0.000 claims description 6
- 210000001932 glossopharyngeal nerve Anatomy 0.000 claims description 6
- 210000000653 nervous system Anatomy 0.000 claims description 6
- 210000001186 vagus nerve Anatomy 0.000 claims description 6
- 208000001308 Fasciculation Diseases 0.000 claims description 5
- 206010028293 Muscle contractions involuntary Diseases 0.000 claims description 5
- 210000003668 pericyte Anatomy 0.000 claims description 5
- 230000026683 transduction Effects 0.000 claims description 4
- 238000010361 transduction Methods 0.000 claims description 4
- 206010001497 Agitation Diseases 0.000 claims description 3
- 230000002964 excitative effect Effects 0.000 claims description 3
- 239000000835 fiber Substances 0.000 claims description 3
- 230000002401 inhibitory effect Effects 0.000 claims description 3
- 230000000946 synaptic effect Effects 0.000 claims 4
- 208000035239 Synesthesia Diseases 0.000 abstract description 4
- 238000010276 construction Methods 0.000 abstract 4
- 238000004451 qualitative analysis Methods 0.000 abstract 1
- 235000013405 beer Nutrition 0.000 description 24
- 239000000243 solution Substances 0.000 description 11
- 238000004140 cleaning Methods 0.000 description 9
- 230000001953 sensory effect Effects 0.000 description 8
- 239000012088 reference solution Substances 0.000 description 7
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 7
- 235000013305 food Nutrition 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- LFQSCWFLJHTTHZ-UHFFFAOYSA-N Ethanol Chemical compound CCO LFQSCWFLJHTTHZ-UHFFFAOYSA-N 0.000 description 5
- GXCLVBGFBYZDAG-UHFFFAOYSA-N N-[2-(1H-indol-3-yl)ethyl]-N-methylprop-2-en-1-amine Chemical compound CN(CCC1=CNC2=C1C=CC=C2)CC=C GXCLVBGFBYZDAG-UHFFFAOYSA-N 0.000 description 5
- 238000001514 detection method Methods 0.000 description 5
- 238000002474 experimental method Methods 0.000 description 5
- 230000004044 response Effects 0.000 description 5
- 241000209094 Oryza Species 0.000 description 4
- 235000007164 Oryza sativa Nutrition 0.000 description 4
- WCUXLLCKKVVCTQ-UHFFFAOYSA-M Potassium chloride Chemical compound [Cl-].[K+] WCUXLLCKKVVCTQ-UHFFFAOYSA-M 0.000 description 4
- 230000008901 benefit Effects 0.000 description 4
- 230000007246 mechanism Effects 0.000 description 4
- 235000009566 rice Nutrition 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 235000013361 beverage Nutrition 0.000 description 3
- 235000019658 bitter taste Nutrition 0.000 description 3
- 238000002790 cross-validation Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 210000000196 olfactory nerve Anatomy 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000000638 stimulation Effects 0.000 description 3
- 230000003461 thalamocortical effect Effects 0.000 description 3
- 206010013911 Dysgeusia Diseases 0.000 description 2
- VEXZGXHMUGYJMC-UHFFFAOYSA-N Hydrochloric acid Chemical compound Cl VEXZGXHMUGYJMC-UHFFFAOYSA-N 0.000 description 2
- 241000209140 Triticum Species 0.000 description 2
- 235000021307 Triticum Nutrition 0.000 description 2
- 210000001015 abdomen Anatomy 0.000 description 2
- 230000036528 appetite Effects 0.000 description 2
- 235000019789 appetite Nutrition 0.000 description 2
- 235000019606 astringent taste Nutrition 0.000 description 2
- 210000004027 cell Anatomy 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 239000012153 distilled water Substances 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000002349 favourable effect Effects 0.000 description 2
- 239000007788 liquid Substances 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 229910044991 metal oxide Inorganic materials 0.000 description 2
- 150000004706 metal oxides Chemical class 0.000 description 2
- 238000002156 mixing Methods 0.000 description 2
- 235000001968 nicotinic acid Nutrition 0.000 description 2
- 239000001103 potassium chloride Substances 0.000 description 2
- 235000011164 potassium chloride Nutrition 0.000 description 2
- 238000002360 preparation method Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 235000014101 wine Nutrition 0.000 description 2
- 101100054292 Arabidopsis thaliana ABCG36 gene Proteins 0.000 description 1
- 101100351526 Arabidopsis thaliana PEN3 gene Proteins 0.000 description 1
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 208000012898 Olfaction disease Diseases 0.000 description 1
- KWYUFKZDYYNOTN-UHFFFAOYSA-M Potassium hydroxide Chemical compound [OH-].[K+] KWYUFKZDYYNOTN-UHFFFAOYSA-M 0.000 description 1
- 235000013334 alcoholic beverage Nutrition 0.000 description 1
- 210000003484 anatomy Anatomy 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 210000003710 cerebral cortex Anatomy 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 235000011389 fruit/vegetable juice Nutrition 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 235000003642 hunger Nutrition 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- FEWJPZIEWOKRBE-LWMBPPNESA-N levotartaric acid Chemical compound OC(=O)[C@@H](O)[C@H](O)C(O)=O FEWJPZIEWOKRBE-LWMBPPNESA-N 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002610 neuroimaging Methods 0.000 description 1
- 210000002475 olfactory pathway Anatomy 0.000 description 1
- 238000012567 pattern recognition method Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 235000019643 salty taste Nutrition 0.000 description 1
- 229920006395 saturated elastomer Polymers 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
- 230000035807 sensation Effects 0.000 description 1
- 235000019615 sensations Nutrition 0.000 description 1
- 238000011896 sensitive detection Methods 0.000 description 1
- 230000008786 sensory perception of smell Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000003756 stirring Methods 0.000 description 1
- 230000002195 synergetic effect Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 235000019583 umami taste Nutrition 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- 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
- G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
- G06N3/063—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
- G06N3/065—Analogue means
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/02—Food
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
- G06F17/13—Differential equations
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Optimization (AREA)
- Data Mining & Analysis (AREA)
- Pure & Applied Mathematics (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Food Science & Technology (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Biophysics (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Operations Research (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Artificial Intelligence (AREA)
- Neurology (AREA)
- Medicinal Chemistry (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a flavor recognition method based on a smell-taste joint perception model, which relates to the technical field of flavor recognition methods and comprises the steps of establishing the smell-taste joint perception model, inputting flavor information of different samples acquired by a smell-taste sensor into the smell-taste joint perception model, extracting the characteristics of the smell-taste joint perception model under the condition of inputting the flavor information by a wavelet packet transformation method, and inputting the characteristics into a GS-SVM model to realize qualitative classification of the different samples. The construction of the olfactory-gustatory synesthesia perception model is based on olfactory-gustatory synesthesia nerve conduction path of human body, which comprises construction of olfactory-gustatory synaptitus solitary nucleus module, construction of olfactory-gustatory synaptitus thalamus ventral posterior medial nucleus module and construction of olfactory-gustatory connection structure module. The flavor identification method can be used for carrying out qualitative analysis on different flavor information more accurately.
Description
Technical Field
The invention relates to the technical field of flavor recognition methods, in particular to a flavor recognition method based on a smell-taste joint perception model.
Background
At present, research aiming at olfactory-gustatory nerve conduction mechanisms is mostly connected with human sensory tests, relevant scholars have researched olfactory-gustatory synaesthesia mechanisms of a cerebral cortex central nervous system through means of physiological anatomy, electrophysiological experiments, an electromagnetic imaging technology, a brain imaging technology and the like, and documents disclose that the smell of food exposed in the air can have a remarkable influence on the appetite of a testee, and things which make people feel pleasurable can improve the hunger of the testee so as to increase the appetite; it is stated in the literature that food is a product category in which olfactory stimuli greatly contribute to the overall taste experience, and most of the food-related sensory pleasure and evaluation of taste is dependent on olfactory perception, as evidenced by studies on the taste perception of patients with olfactory impairment; the information synaesthesia ability of the olfactory and gustatory systems is found by people to form food perception and favorable feeling, and is verified by experiments, and the result shows that the primary olfactory cortex is identified as a possible site playing an important role in forming the food perception and favorable feeling. There are papers and patents which propose smell and taste perception models respectively, but the sensory experience and sensory evaluation of foods are usually the sensory mixing phenomenon after the synergistic action of human taste and smell, and both models start from the perspective of a single sensory channel and do not consider the sensory mixing phenomenon between smell and taste.
The above research aiming at the olfactory-olfactory synesthesia mechanism is mainly the research result obtained by physiological experiments or a single olfactory/gustatory system, but the overall olfactory-olfactory synesthesia mechanism and the perception capability thereof are not researched, and high bionics cannot be realized.
Disclosure of Invention
The invention mainly aims to provide a flavor identification method based on a smell-taste joint perception model.
The technical scheme adopted by the invention is as follows: a flavor identification method based on a smell-taste joint perception model comprises the following steps:
s1: establishing a topological structure and dynamic characteristic description model of an orphan beam kernel module in a smell-taste joint perception model;
s2: establishing a topological structure and dynamic characteristic description model of a thalamus ventral posterior medial nucleus module in the olfactory-olfactory synaptism perception model;
s3: establishing a topological structure and dynamic characteristic description model of a connection structure module in a smell-taste joint perception model;
s4: constructing an olfactory-gustatory synaesthesia perception model by combining the modules in the steps S1-S3 and the cranial nerve module, the trigeminal nerve module, the islet cortex module, the pericyte module, the olfactory bulb module, the anterior olfactory nucleus module, the anterior piriformis cortex module, the exocortical layer module and the olfactory-gustatory feedback module;
s5: acquiring flavor information of different samples through a smell-taste sensor and inputting the flavor information into the smell-taste joint perception model in the step S4;
s6: the characteristic of the smell-taste joint perception model under the condition of inputting flavor information of different samples is extracted through a wavelet packet transformation method, and the characteristic is input into a GS-SVM model, so that qualitative classification of the different samples is realized.
Furthermore, the topological structure of the solitary bundle nucleus module is in a multi-input multi-output structure mode, the solitary bundle nucleus module receives the outputs of the facial nerve module, the glossopharyngeal nerve module, the vagus nerve module and the trigeminal nerve module and the feedback information of the islet cortex module, and outputs taste information to the thalamus retroperitoneal inner nucleus module and the islet cortex module, wherein the topological structure of the trigeminal nerve module is in a single-output structure mode; the trigeminal module outputs information to the solitary bundle nucleus module, and is provided with a node J5;
The isolated beam nuclear module is provided with four nodes J1、J2、K1And K2The dynamics description models are differential equations of formula (1) to formula (4), respectively:
(1)
(2)
(3)
(4)
wherein, a and b1All are differential equation coefficients, and n is the number of channels of the taste perception model;
J1i(t)、J2i(t)、K1i(t)、K2i(t) are respectively the ith passage J1、J2、K1、K2Potential state of the node (i =1, …, n);
Q(J1i(t)),Q(J2i(t)),Q(K1i(t)),Q(K2i(t)),Q(E1i(t)),Q(E2i(t)),Q(E3i(t)) and Q (J)5i(t)) is the ith channel J1, J2,K1,K2,E1,E2,E3,J5An output of the node;
E1,E2,E3respectively are nodes of a facial nerve module, a glossopharyngeal nerve module and a vagus nerve module in the cranial nerve module in the taste perception model;
Q(J1j(t)) is the jth channel J1The output of the node (j =1, …, n); q (K)1j(t)) is the jth channel K1An output of the node;
D6(t) is the potential state of the feedback link from the islet cortex to the solitary nucleus in the gustatory nerve conduction pathway;
Wjjfor all channels J1The connection coefficient between the nodes;
Wj1j2,Wj2j1is the same channel J1And J2Connection coefficient of nodes, all channels J1And J2The connection coefficient values of the nodes are consistent;
Wj1k1,Wk1j1is the same channel J1And K1Connection coefficient of nodes, all channels J1And K1The connection coefficient values of the nodes are consistent;
Wj1k2,Wk2j1is the same channel J1And K2Connection coefficient of nodes, all channels J1And K2The connection coefficient values of the nodes are consistent;
Wj1e1is the same channel J1And E1Connection coefficient of nodes, all channels J1And E1The connection coefficient values of the nodes are consistent;
Wj1e2is the same channel J1And E2Connection coefficient of nodes, all channels J1And E2The connection coefficient values of the nodes are consistent;
Wj1e3is the same channel J1And E3Connection coefficient of nodes, all channels J1And E3The connection coefficient values of the nodes are consistent;
Wj1j5is the same channel J1And J5Connection coefficient of nodes, all channels J1And J5The connection coefficient values of the nodes are consistent;
Wj2k1,Wk1j2is the same channel J2And K1Connection coefficient of nodes, all channels J2And K1The connection coefficient values of the nodes are consistent;
Wkkfor all channels K1The connection coefficient between the nodes;
Wk1k2,Wk2k1is the same channel K1And K2Connection coefficient of nodes, all channels K1And K2The connection coefficient values of the nodes are consistent;
Kd6is J1Node and feedback link D6Is connected withA coefficient; the values of i and j are all 1 to n, i is not equal to j, and n is the number of channels of the taste perception model.
Furthermore, the topological structure of the thalamus ventral posterior medial nucleus module is a multi-input single-output structure mode, namely the thalamus ventral posterior medial nucleus module receives the gustatory information of the solitary fasciculus nucleus module and the insular cortex module and the olfactory information of the front piriformis cortex module and outputs the gustatory information to the insular cortex module, and the thalamus ventral posterior medial nucleus module is provided with four nodes which are J nodes respectively3、J4、K3、K4The dynamics description models are differential equations of equations (8) to (11), respectively:
(8)
(9)
(10)
(11)
wherein, a and b1All are differential equation coefficients, and n is the number of channels of the taste perception model;
J3(t),J4(t),K3(t), K4(t) is J3,J4,K3,K4The potential state of the node;
Q(J3(t)),Q(J4(t)),Q(K3(t)),Q(K4(t)),Q(B3(t)) is J3,J4,K3,K4,B3An output of the node;
B3the nodes are front piriform cortex module nodes;
Q(J1j(t)) is the jth channel J1The output of the node, wherein j takes the value of 1 to n;
D5(t) is the potential state of the feedback link from the islet cortex to the retrothalamic medial nucleus in the taste transduction pathway;
Wj3j1for all channels J1Node and J3The connection coefficient of the node;
Wj3j4,Wj4j3is J3And J4The connection coefficient of the node;
Wj3k3,Wk3j3is J3And K3The connection coefficient of the node;
Wj3k4,Wk4j3is J3And K4The connection coefficient of the node;
Wj4k3,Wk3j4is J4And K3The connection coefficient of the node;
Wk3k4,Wk4k3is K3And K4The connection coefficient of the node;
Kd5is J3Node and feedback link D5The connection coefficient between;is the potential state of the central noise of the taste perception model.
Furthermore, the topological structure of the linkage structure module is a multi-input structure mode, the linkage structure module receives the taste information of the island cortex module of the taste nervous system and the posterior medial nucleus module of the thalamus abdomen and the smell information of the outer cortex module of the smell nervous system, and the dynamic characteristic description model of the N node of the smell-taste linkage structure module is a differential equation of a formula (22):
wherein, a and b1All are differential equation coefficients, and s is the number of channels of the taste perception model minus 1;
Nl(t) is the potential state of the nth node of the l channel, l =1,2, …, N; n is the number of channels of the taste perception model;
Q(K3(t)), Q (M (t)), Q (C (t)) is K3The output of node M, C;
Q(Nj(t)) is the output of the N node of the j channel, wherein l takes on the value from 1 to s;
Wnnconnecting coefficients among N nodes of all channels;
Wnk3is K3And the connection coefficient of the N node;
Wnmthe connection coefficient of the N node and the M node is obtained;
Wncthe connection coefficient of the N node and the C node is obtained;
q (C (t)) is the output of the skin layer module, which is represented by the C node;
q (M (t)) is the output of the island cortex module, which is represented by the M node.
Further, in the olfactory-olfactory joint perception model, trigeminal module J5Node and isolated beam nucleus module J1Node association, J5The node outputs the excitability information to J1A node;
b of front piriform cortex module3J for outputting inhibitory olfactory information to posterior medial nucleus module of thalamus abdomen3A node, the fiber link connecting the sniff-gustatory perception model;
olfactory information from the cortex module and gustatory information from the islet cortex module are input to the N nodes of the olfactory-gustatory linkage module in a linear superposition, and furthermore, the N nodes receive information from the thalamus ventral medial nucleus module K3An output of the node;
lateral connection exists among N nodes of all channels of the olfactory-olfactory joint perception model, and each channel N node receives excitatory output from other channel N nodes.
The invention has the advantages that:
the flavor recognition method provided by the invention describes the processing process of human flavor information from the aspect of the olfaction-flavor nerve conduction dynamic characteristic, is close to the olfaction-flavor joint perception experience of human, and improves the bionics degree of a flavor recognition system.
The flavor identification method of the invention can accurately identify the flavor information of wines or other liquid beverages, and make qualitative classification, and can be widely applied to the production and identification of wines and other beverages.
In addition to the objects, features and advantages described above, other objects, features and advantages of the present invention are also provided. The present invention will be described in further detail below with reference to the drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the invention and, together with the description, serve to explain the invention and not to limit the invention.
FIG. 1 is a topological structure of an orphan bundle kernel module of a flavor identification method based on a smell-taste joint perception model of the invention;
FIG. 2 is a topological structure of a trigeminal nerve module of a flavor identification method based on a smell-taste joint perception model of the invention;
FIG. 3 is a topological structure of a thalamus retroventral medial nucleus module of a flavor recognition method based on a smell-taste joint perception model of the invention;
FIG. 4 is a topological structure of a front piriformis layer module of the flavor identification method based on a smell-taste joint perception model;
FIG. 5 is a topological structure of a connection structure module of a flavor identification method based on a smell-taste joint perception model of the invention;
FIG. 6 is a topological structure of an outer skin layer module of a flavor identification method based on a smell-taste joint perception model of the invention;
FIG. 7 is a topological structure of an island cortex module of a flavor identification method based on a smell-taste joint perception model of the invention;
FIG. 8 is a topological structure of an overall olfactory-olfactory simultaneous perception model of the olfactory-olfactory simultaneous perception model-based flavor recognition method of the present invention;
FIG. 9 is a flow chart of a flavor recognition method based on a smell-taste joint perception model according to the present invention;
FIG. 10 is a radar chart of flavor information of beer of different brands based on a flavor recognition method of a smell-taste joint perception model of the invention;
FIG. 11 is a parameter search process of GS-SVM in the flavor recognition method based on the olfactory-olfactory joint perception model of the present invention;
fig. 12 shows the classification result of the GS-SVM in the flavor recognition method based on the olfactory-olfactory joint perception model according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 1 to 12, a flavor recognition method based on a smell-taste joint perception model includes the following steps:
s1: establishing a topological structure and dynamic characteristic description model of an orphan beam kernel module in a smell-taste joint perception model;
s2: establishing a topological structure and dynamic characteristic description model of a thalamus ventral posterior medial nucleus module in the olfactory-olfactory synaptism perception model;
s3: establishing a topological structure and dynamic characteristic description model of a connection structure module in a smell-taste joint perception model;
s4: constructing an olfactory-gustatory synaesthesia perception model by combining the modules in the steps S1-S3 and the cranial nerve module, the trigeminal nerve module, the islet cortex module, the pericyte module, the olfactory bulb module, the anterior olfactory nucleus module, the anterior piriformis cortex module, the exocortical layer module and the olfactory-gustatory feedback module;
s5: acquiring flavor information of different samples through a smell-taste sensor and inputting the flavor information into the smell-taste joint perception model in the step S4;
s6: the characteristic of the smell-taste joint perception model under the condition of inputting flavor information of different samples is extracted through a wavelet packet transformation method, and the characteristic is input into a GS-SVM model, so that qualitative classification of the different samples is realized.
In this embodiment, the topological structure of the solitary bundle nucleus module is in a multi-input multi-output structural mode, the solitary bundle nucleus module receives the outputs of the facial nerve module, the glossopharyngeal nerve module, the vagus nerve module and the trigeminal nerve module and the feedback information of the islet cortex module, and outputs taste information to the thalamus retroperitoneal inner nucleus module and the islet cortex module, wherein the topological structure of the trigeminal nerve module is in a single-output structural mode; as shown in FIG. 2, the trigeminal module outputs information to the solitary bundle nucleus module, which is provided with a node J5;
The isolated beam nuclear module is provided with four nodes J1、J2、K1And K2The dynamics description models are differential equations of formula (1) to formula (4), respectively:
(1)
(2)
(3)
(4)
wherein, a and b1Are coefficients of differential equations, n is taste perceptionThe number of channels of the model;
J1i(t)、J2i(t)、K1i(t)、K2i(t) are respectively the ith passage J1、J2、K1、K2Potential state of the node (i =1, …, n);
Q(J1i(t)),Q(J2i(t)),Q(K1i(t)),Q(K2i(t)),Q(E1i(t)),Q(E2i(t)),Q(E3i(t)) and Q (J)5i(t)) is the ith channel J1, J2,K1,K2,E1,E2,E3,J5An output of the node;
E1,E2,E3respectively are nodes of a facial nerve module, a glossopharyngeal nerve module and a vagus nerve module in the cranial nerve module in the taste perception model;
Q(J1j(t)) is the jth channel J1The output of the node (j =1, …, n); q (K)1j(t)) is the jth channel K1An output of the node;
D6(t) is the potential state of the feedback link from the islet cortex to the solitary nucleus in the gustatory nerve conduction pathway;
Wjjfor all channels J1The connection coefficient between the nodes;
Wj1j2,Wj2j1is the same channel J1And J2Connection coefficient of nodes, all channels J1And J2The connection coefficient values of the nodes are consistent;
Wj1k1,Wk1j1is the same channel J1And K1Connection coefficient of nodes, all channels J1And K1The connection coefficient values of the nodes are consistent;
Wj1k2,Wk2j1is the same channel J1And K2Connection coefficient of nodes, all channels J1And K2The connection coefficient values of the nodes are consistent;
Wj1e1is the same channel J1And E1Connection coefficient of nodes, all channels J1And E1The connection coefficient values of the nodes are consistent;
Wj1e2is the same channel J1And E2Connection coefficient of nodes, all channels J1And E2The connection coefficient values of the nodes are consistent;
Wj1e3is the same channel J1And E3Connection coefficient of nodes, all channels J1And E3The connection coefficient values of the nodes are consistent;
Wj1j5is the same channel J1And J5Connection coefficient of nodes, all channels J1And J5The connection coefficient values of the nodes are consistent;
Wj2k1,Wk1j2is the same channel J2And K1Connection coefficient of nodes, all channels J2And K1The connection coefficient values of the nodes are consistent;
Wkkfor all channels K1The connection coefficient between the nodes;
Wk1k2,Wk2k1is the same channel K1And K2Connection coefficient of nodes, all channels K1And K2The connection coefficient values of the nodes are consistent;
Kd6is J1Node and feedback link D6The connection coefficient between; the values of i and j are all 1 to n, i is not equal to j, and n is the number of channels of the taste perception model.
In this example, the cranial nerve model in the taste perception model (E)1,E2,E3) The dynamic characterization model of (1):
(5)
(6)
(7)
E1i(t)、E2i(t)、E3i(t) are respectively the ith channel E1、E2、E3The potential state of the node; i =1,2, …, n, n is the number of channels of the taste perception model;
Wr1,Wr2,Wr3is the same channel E1,E2And E3The connection coefficient of the nodes and the RG is consistent with the connection coefficient of all channels;
We3e2is the same channel E2And E3The connection coefficient between the nodes is consistent with the connection coefficient values of all the channels;
NGp(t) is the potential state of the peripheral noise in the taste perception model; a and b1Is the differential equation coefficient; RG is the external stimulus in the taste perception model; RG (route group)iAnd (t) is the potential state of the external stimulus of the ith channel taste perception model.
In this embodiment, the topology of the thalamocortical nucleus module is a multi-input single-output structure mode, as shown in fig. 3, that is, the thalamocortical nucleus module receives the gustatory information of the solitary nucleus module, the islet cortex module, and the olfactory information of the anterior piriformis cortex module, and outputs the received information to the islet cortex module, and the thalamocortical nucleus module is provided with four nodes, J being respectively3、J4、K3、K4The dynamics description models are differential equations of equations (8) to (11), respectively:
(8)
(9)
(10)
(11)
wherein, a and b1All are differential equation coefficients, and n is the number of channels of the taste perception model;
J3(t),J4(t),K3(t), K4(t) is J3,J4,K3,K4The potential state of the node;
Q(J3(t)),Q(J4(t)),Q(K3(t)),Q(K4(t)),Q(B3(t)) is J3,J4,K3,K4,B3An output of the node;
Q(J1j(t)) is the jth channel J1The output of the node, wherein j takes the value of 1 to n;
D5(t) is the potential state of the feedback link from the islet cortex to the retrothalamic medial nucleus in the taste transduction pathway;
Wj3j1for all channels J1Node and J3The connection coefficient of the node;
Wj3j4,Wj4j3is J3And J4The connection coefficient of the node;
Wj3k3,Wk3j3is J3And K3The connection coefficient of the node;
Wj3k4,Wk4j3is J3And K4The connection coefficient of the node;
Wj4k3,Wk3j4is J4And K3The connection coefficient of the node;
Wk3k4,Wk4k3is K3And K4The connection coefficient of the node;
Kd5is J3Node and feedback link D5The connection coefficient between;is the potential state of the central noise of the taste perception model.
In this example, Q (B)3(t)) is B3The output of the node is the output of a front piriformis layer module in the olfactory perception model; the topological structure of the front piriformis layer module is in a multi-input and multi-output structure mode, as shown in fig. 4, the front piriformis layer module receives information of the olfactory bulb module and the exocortical layer module and outputs the information to the exocortical layer module, the thalamus abdomen posterior medial nucleus module and the front olfactory nucleus module;
the front piriform cortex module is provided with four nodes A1、A2、B3、B4The dynamics description models are differential equations of equations (12) to (15), respectively:
(12)
(13)
(14)
(15)
wherein, a and b2All are differential equation coefficients, and m is the number of channels of the olfactory perception model; a. the1(t)、A2(t)、B3(t)、B4(t) are each A1、A2、B3、B4The potential state of the node;
Q(A1(t)),Q(A2(t)), Q(B3(t)),Q(B4(t)), Q (C (t)) is A1,A2,B3,B4Outputting a C node, wherein the C node is a topological form of an exodermis layer module in the olfactory perception model;
Q(M1k(t)) is the k-th channel M1Output of node, M1The nodes are key nodes of the olfactory bulb in the olfactory transduction model (k =1, …, m);
Wb3cis B3And the connection coefficient between the C nodes;
Wa1mfor all channels M1Node A and node B1The connection coefficient between the nodes;
Wa1a2,Wa2a1is A1And A2The connection coefficient of the node;
Wa1b3,Wb3a1is A1And B3The connection coefficient of the node;
Wa1b4,Wb4a1is A1And B4The connection coefficient of the node;
Wa2b3and Wb3a2Is A2And B3The connection coefficient of the node;
Wb3b4,Wb4b3is B3And B4The connection coefficient of the node.
In this embodiment, the olfactory perception model includes an olfactory bulb module (M)1,M2,G1,G2) The dynamic characteristic description model of (1) is as follows:
(16)
(17)
(18)
(19)
wherein M is1l(t)、M2l(t)、G1l(t)、G2l(t) are the first channels M, respectively1、M2、G1、G2The potential state of the node (l =1, …, m), where m is the number of channels of the olfactory perception model; a and b2Are all differential equation coefficients;
Q(M1l(t)),Q(M2l(t)),Q(G1l(t)),Q(G2l(t)),Q(P1l(t)) is the l-th channel M1,M2,G1,G2,P1Output of node, where P1The nodes are pericentral cell module nodes in the olfactory perception model;
Q(M1j(t)) is the jth channel M1The output of the node (j =1, …, m); q (G)1j(t)) is the jth channel G1An output of the node;
D1(t) is the potential state of the feedback link from the anterior olfactory nucleus to the olfactory bulb in the olfactory nerve conduction pathway;
D4(t) is the potential state of the feedback link from the exodermis to the olfactory bulb in the olfactory nerve conduction pathway;
Wm1g1,Wg1m1is the same channel M1And G1Connection of nodesConnecting coefficients, all channels M1And G1The connection coefficient values of the nodes are consistent;
Wm1g2,Wg2m1is the same channel M1And G2Connection coefficient of nodes, all channels M1And G2The connection coefficient values of the nodes are consistent;
Wm1p1is the same channel M1And P1Connection coefficient of nodes, all channels M1And P1The connection coefficient values of the nodes are consistent;
Wm1ris the same channel M1And ROlAll channels M1And ROlThe values of the connection coefficients are consistent; ROlThe external stimulation of the first channel is consistent with that of all channels; ROl(t) is the potential state of the first channel external stimulation;
Wm2g1,Wg1m2is the same channel M2And G1Connection coefficient of nodes, all channels M2And G1The connection coefficient values of the nodes are consistent;
Wg1g2is the same channel G1And G2Connection coefficient of nodes, all channels G1And G2The connection coefficient values of the nodes are consistent;
Wm1m2is the same channel M1And M2Connection coefficient of nodes, all channels M1And M2The connection coefficient values of the nodes are consistent.
In this example, the pericyte module (P) in the olfactory perception model1,P2) The dynamic characteristic description model of (1) is as follows:
(20)
(21)
wherein, P1l(t)、P2l(t) is the first channel P1、P2The potential state of the node (l =1, …, m), where m is the number of channels of the olfactory perception model; a and b2Are all differential equation coefficients;
Q(P1l(t)) is the first channel P1An output of the node;
Q(P1j(t)) is the jth channel P1The output of the node (j =1, …, m);
D2(t) is the potential state of the feedback link from the anterior olfactory nucleus to the pericyte in the olfactory nerve conduction pathway;
Wppfor all channels P1The connection coefficient between the nodes;
Wp1p2is the same channel P1And P2Connection coefficient of nodes, all channels P1And P2The connection coefficient values of the nodes are consistent.
In this embodiment, the topology of the connection structure module is a multi-input structure mode, as shown in fig. 5, the connection structure module receives gustatory information of the gustatory nervous system island cortex module and the thalamus ventral posterior medial nucleus module and olfactory information of the olfactory nervous system cortex module, and the dynamics characteristic description model of the node N of the olfactory-gustatory connection structure module is a differential equation of formula (22):
wherein, a and b1All are differential equation coefficients, and s is the number of channels of the taste perception model minus 1;
Nl(t) is the potential state of the nth node of the l channel, l =1,2, …, N; n is the number of channels of the taste perception model;
Q(K3(t)), Q (M (t)), Q (C (t)) is K3The output of node M, C;
Q(Nj(t)) is the output of the N node of the j channel, wherein l takes on the value from 1 to s;
Wnnconnecting coefficients among N nodes of all channels;
Wnk3is K3And the connection coefficient of the N node;
Wnmthe connection coefficient of the N node and the M node is obtained;
Wncthe connection coefficient of the N node and the C node is obtained;
q (c (t)) is the output of the cortex module, the topological structure of the cortex module is a single-input multiple-output structural mode, as shown in fig. 6, the cortex module receives olfactory information of the front piriformis module and outputs the information to the olfactory bulb module, the front piriformis module and the connection structural module; the cortex module is represented by a C node;
q (m (t)) is an output of the island cortex module, and the topological structure of the island cortex module is a multi-input multi-output structural mode, as shown in fig. 7, the island cortex module receives information of the solitary fasciculation nucleus module and the thalamus ventral posterior medial nucleus module, and outputs information to the solitary fasciculation nucleus module, the thalamus ventral posterior medial nucleus module and the connection structural module; the island cortex modules are represented by M nodes.
In this embodiment, in the olfactory-olfactory joint perception model, as shown in fig. 8, the trigeminal module J5Node and isolated beam nucleus module J1Node association, J5The node outputs the excitability information to J1A node;
J1also received is module E from cranial nerves1、E2、E3Excitability output information of the node;
front piriform cortex module A1The node accepts all M's from the sniffer ball module1Olfactory processing information of the node and outputting the information to I of the pre-olfactory nucleus module1A node;
b of front piriform cortex module3J for outputting inhibitory olfactory information to posterior medial nucleus module of thalamus abdomen3A node, the fiber link connecting the sniff-gustatory perception model;
the olfactory information from the cortex module and the gustatory information from the islet cortex module are input to the N node of the olfactory-gustatory linkage module in a linear superposition manner, and furthermore, the N node receives the olfactory information from the posterior-medial ventral thalamusNuclear Module K3An output of the node;
lateral connection exists among N nodes of all channels of the olfactory-olfactory joint perception model, and each channel N node receives excitatory output from other channel N nodes.
And (3) experimental verification:
in the present invention, the flavor information recognition process and qualitative classification of the beer of different brands by the flavor recognition method of the present invention will be described in detail by taking beer of different brands as an example, but the present invention is not limited to beer in practical use, and flavor information of other alcoholic beverages and liquid beverages may be accurately and qualitatively classified.
(I) Experimental Equipment and Material
The different samples can be 5 kinds of beer with different brands, and the specific parameters are shown in the table 1:
serial number | Brand | Alcohol content (% vol) | Concentration of raw wheat juice (1)。P) | Raw | Producing area | |
1 | Blue belt | ≥4.3 | 11 | Water, malt, rice, hop | Guangdong (Chinese character of Guangdong) |
|
2 | Snow flake | ≥3.3 | 9 | Water, malt, rice, | Jilin Changchun | |
3 | Baiwei medicine | ≥3.6 | 9.7 | Water, malt, wheat, hop | Tangshan mountain of Hebei river | |
4 | Harbin | ≥3.6 | 9.1 | Water, malt, rice, | Wuhan Hubei | |
5 | Qingdao (Qingdao) | ≥4.3 | 11 | Water, malt, rice, hop | Shandong Qingdao |
TABLE 1 parameter information for different brands of beer
The olfactory sensor can be a PEN3 electronic nose, and a sensor array of the electronic nose comprises 10 metal oxide sensors, so that cross-sensitive detection of odor information can be realized.
The taste sensor can be an SA-402B electronic tongue, the sensor array of which consists of 2 reference electrodes and 5 basic taste sensors; the basic taste sensor can realize the sensory information detection of 5 basic tastes including sour, fresh, salty, bitter and astringent tastes of a sample to be detected.
(II) a specific process for acquiring flavor information of beer of different brands:
1. olfactory information acquisition of beer of different brands
(1) An experimenter puts a 5ml beer sample into a 50ml sample bottle, screws down a bottle cap and keeps for 10 minutes to ensure that the gas at the top of the sealed bottle reaches a saturated state;
(2) prior to gas detection, the experimenter needs to clean and calibrate the sensor gas cell. The specific operation is as follows: introducing clean gas dried by active carbon with the flow rate of 300ml/min into the air chamber, and keeping the flow rate for 60 s;
(3) after the correction is completed, the detection is started. Each set of samples was tested for 100s to allow the sensor response to reach steady state. The sensor response value is G/G0(G0G), where G is the conductance of the sensor when the gas to be measured enters the chamber, G0Is the conductance of the sensor when pure air enters the air chamber.
(4) For each beer 18 parallel samples were prepared and for 5 beers 90 sets of data were obtained. The conductivity value at 60s of the sensor response curve was taken as a characteristic value for data analysis. The beer smell-taste information acquisition experiment conditions are consistent: the room temperature is 20 +/-0.5 ℃, and the relative humidity is 65 +/-2% RH.
2. Taste information acquisition for different brands of beer
(1) Placing a beer sample, a reference solution and a positive and negative electrode cleaning solution;
the reference solution is a solution containing 0.3 mmol/L tartaric acid and 30mmol/L potassium chloride; the preparation process of the anode cleaning solution comprises the following steps: adding 300mL of 95% ethanol into about 500mL of distilled water, fully stirring, adding 100mL of 1M hydrochloric acid solution, transferring the solution into a 1000mL volumetric flask for constant volume to obtain a positive electrode cleaning solution; the preparation process of the negative electrode cleaning solution comprises the following steps: after 7.46g of potassium chloride and 500mL of 95% ethanol were added to about 500mL of distilled water and stirred uniformly, 10mL of 1M potassium hydroxide solution was added and the mixture was transferred to a 1000mL volumetric flask for constant volume to obtain a negative electrode cleaning solution.
(2) Before the test starts, the positive electrode sensor array is placed into positive electrode cleaning solution, the negative electrode sensor array is placed into negative electrode cleaning solution and cleaned for 90s, after the test is finished, the positive electrode sensor array and the negative electrode sensor array are respectively placed into two containers containing reference solution and cleaned for 120s, the reference solution is replaced and cleaned continuously for 120s, and then the reference solution is replaced to enable the balance of the sensors to return to zero for 30s so as to ensure that output signals are stable;
(3) after the response output of the sensor reaches balance, starting to acquire taste information of the beer, wherein the detection time of each type of beer is 30s, after the measurement is finished, quickly cleaning the beer in a reference solution for 2 times, detecting the aftertaste value of the basic taste information in the replaced reference solution, finishing the measurement once, repeating the step (2), and cleaning and calibrating the sensor;
(4) 3 parallel samples are prepared for the beer of different brands, each group of samples are repeatedly detected for 6 times by setting system parameters, namely 18 groups of taste information data are intelligently obtained for the beer of each brand, 90 groups of taste data are obtained after the experiment is finished, and the voltage value of the 30 th s of the sensor response curve is taken as a characteristic value for data analysis.
And performing characteristic fusion and selection on the olfactory information and the gustatory information of the beer to finally obtain the flavor information of the beer. FIG. 10 shows radar charts of flavor information of beer of different brands (wherein C00: bitterness sensor; AE 1: astringency sensor; CA 0: acidity sensor; Cpa (COO): bitterness sensor aftertaste value detection; AAE: umami sensor; W1C, W3S, W3C, W5C, W1W, W2S, W1S are electronic nose metal oxide sensors).
Parameter value setting of differential equations of different modules of (II) smell-taste joint perception model
Setting parameter values according to a computer numerical simulation result of the smell-taste joint perception model:
TABLE 2 parameter values of the differential equations of all modules of the olfactory-olfactory Joint perception model
The method comprises the steps of extracting features of an olfactory-olfactory joint perception model under the condition that flavor information of different samples is input through a Wavelet Packet Transform (WPT) method, and inputting the features into a GS-SVM (grid search-support vector machine) model to realize qualitative classification of the different samples.
Referring to fig. 9 to 12, 5 different brands of beer flavor data are input into the olfactory-olfactory simultaneous perception model, and wavelet packet variances of N node output sequences of 5 channels OFC layers of the olfactory-olfactory simultaneous perception model are obtained by a WPT method.
The method divides 90 groups of characteristic data into 60 groups of training set data and 30 groups of test set data, selects a Radial Basis Function (RBF) as a kernel function, and optimizes a penalty factor in a support vector machine model by utilizing a grid search algorithmAnd kernel parameters. In this connection, it is possible to use,in the range of [0.25, 16 ]],In the range of [0.0625, 16]. In the parameter searching process, the cross validation classification error rate is adopted as a fitness function, and the parameter with the maximum cross validation accuracy rateAndand returning and establishing the model.
The method is used for detecting the smell-taste synaesthesia perception modelThe wave packet variance feature data set is processed, the parameter searching process is shown in FIG. 11, and the optimal parameter is found by searching the curve chartAndand the GS-SVM cross validation accuracy is 93.33%, the classification accuracy is 96.67%, and the numerical result shows that the pattern recognition method provided by the invention can realize qualitative classification of beer samples of different brands.
The invention relates to a flavor identification method based on a smell-taste joint perception model, which has the working principle that: by using a smell-taste sensor, which can be an electronic nose-electronic tongue, the flavor information of different samples is obtained as external stimulationInputting the olfactory sensation to the olfactory-olfactory simultaneous perception model constructed by the invention in combination with external noise; and extracting the wavelet packet variance characteristics of the olfactory-olfactory joint perception model under the input stimulation, inputting the characteristics into a GS-SVM model, and verifying the effectiveness of the mode recognition method according to the qualitative classification result of the GS-SVM on flavor information of different samples.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (5)
1. A flavor recognition method based on a smell-taste joint perception model is characterized by comprising the following steps of
The following steps:
s1: establishing a topological structure and dynamic characteristic description model of an orphan beam kernel module in a smell-taste joint perception model;
s2: establishing a topological structure and dynamic characteristic description model of a thalamus ventral posterior medial nucleus module in the olfactory-olfactory synaptism perception model;
s3: establishing a topological structure and dynamic characteristic description model of a connection structure module in a smell-taste joint perception model;
s4: constructing an olfactory-gustatory synaesthesia perception model by combining the modules in the steps S1-S3 and the cranial nerve module, the trigeminal nerve module, the islet cortex module, the pericyte module, the olfactory bulb module, the anterior olfactory nucleus module, the anterior piriformis cortex module, the exocortical layer module and the olfactory-gustatory feedback module;
s5: acquiring flavor information of different samples through a smell-taste sensor and inputting the flavor information into the smell-taste joint perception model in the step S4;
s6: the characteristic of the smell-taste joint perception model under the condition of inputting flavor information of different samples is extracted through a wavelet packet transformation method, and the characteristic is input into a GS-SVM model, so that qualitative classification of the different samples is realized.
2. The method for flavor recognition based on olfactory-taste synaptic perception model according to claim 1, wherein the method comprises
Is characterized in that the topological structure of the solitary fasciculation nucleus module is a multi-input multi-output structure mode, the solitary fasciculation nucleus module receives the output of the facial nerve module, the glossopharyngeal nerve module, the vagus nerve module and the trigeminal nerve module and the feedback information of the islet cortex module and outputs taste information to the thalamus retroperitoneal inner side nucleus module and the islet cortex module, wherein the topological structure of the trigeminal nerve module is a single-output structure mode; the trigeminal module outputs information to the solitary bundle nucleus module, and is provided with a node J5;
The isolated beam nuclear module is provided with four nodes J1、J2、K1And K2The dynamics description models are differential equations of formula (1) to formula (4), respectively:
(1)
(2)
(3)
(4)
wherein, a and b1All are differential equation coefficients, and n is the number of channels of the taste perception model;
J1i(t)、J2i(t)、K1i(t)、K2i(t) are respectively the ith passage J1、J2、K1、K2Potential state of the node (i =1, …, n);
Q(J1i(t)),Q(J2i(t)),Q(K1i(t)),Q(K2i(t)),Q(E1i(t)),Q(E2i(t)),Q(E3i(t)) and Q (J)5i(t)) is the ith channel J1, J2,K1,K2,E1,E2,E3,J5An output of the node;
E1,E2,E3respectively are nodes of a facial nerve module, a glossopharyngeal nerve module and a vagus nerve module in the cranial nerve module in the taste perception model;
Q(J1j(t)) is the jth channel J1The output of the node (j =1, …, n); q (K)1j(t)) is the jth channel K1An output of the node;
D6(t) is the potential state of the feedback link from the islet cortex to the solitary nucleus in the gustatory nerve conduction pathway;
Wjjfor all channels J1The connection coefficient between the nodes;
Wj1j2,Wj2j1is the same channel J1And J2Connection coefficient of nodes, all channels J1And J2The connection coefficient values of the nodes are consistent;
Wj1k1,Wk1j1is the same channel J1And K1Connection coefficient of nodes, all channels J1And K1The connection coefficient values of the nodes are consistent;
Wj1k2,Wk2j1is the same channel J1And K2Connection coefficient of nodes, all channels J1And K2The connection coefficient values of the nodes are consistent;
Wj1e1is the same channel J1And E1Connection coefficient of nodes, all channels J1And E1The connection coefficient values of the nodes are consistent;
Wj1e2is the same channel J1And E2Connection coefficient of nodes, all channels J1And E2The connection coefficient values of the nodes are consistent;
Wj1e3is the same channel J1And E3Connection coefficient of nodes, all channels J1And E3The connection coefficient values of the nodes are consistent;
Wj1j5is the same channel J1And J5Connection coefficient of nodes, all channels J1And J5The connection coefficient values of the nodes are consistent;
Wj2k1,Wk1j2is the same channel J2And K1Connection coefficient of nodes, all channels J2And K1The connection coefficient values of the nodes are consistent;
Wkkfor all channels K1The connection coefficient between the nodes;
Wk1k2,Wk2k1is the same channel K1And K2Connection coefficient of nodes, all channels K1And K2The connection coefficient values of the nodes are consistent;
Kd6is J1Node and feedback link D6The connection coefficient between; the values of i and j are all 1 to n, i is not equal to j, and n is taste senseThe number of model channels is known.
3. The method for flavor recognition based on olfactory-taste synaptic perception model according to claim 1, wherein the method comprises
Is characterized in that the topological structure of the thalamus ventral posterior medial nucleus module is a multi-input single-output structure mode, namely the thalamus ventral posterior medial nucleus module receives the gustatory information of the solitary fasciculation nucleus module and the islet cortex module and the olfactory information of the front piriformis cortex module and outputs the gustatory information to the islet cortex module, the thalamus ventral posterior medial nucleus module is provided with four nodes which are J nodes respectively3、J4、K3、K4The dynamics description models are differential equations of equations (8) to (11), respectively:
(8)
(9)
(10)
(11)
wherein, a and b1All are differential equation coefficients, and n is the number of channels of the taste perception model;
J3(t),J4(t),K3(t), K4(t) is J3,J4,K3,K4The potential state of the node;
Q(J3(t)),Q(J4(t)),Q(K3(t)),Q(K4(t)),Q(B3(t)) is J3,J4,K3,K4,B3An output of the node;
B3the nodes are front piriform cortex module nodes;
Q(J1j(t)) is the jth channel J1The output of the node, wherein j takes the value of 1 to n;
D5(t) is the potential state of the feedback link from the islet cortex to the retrothalamic medial nucleus in the taste transduction pathway;
Wj3j1for all channels J1Node and J3The connection coefficient of the node;
Wj3j4,Wj4j3is J3And J4The connection coefficient of the node;
Wj3k3,Wk3j3is J3And K3The connection coefficient of the node;
Wj3k4,Wk4j3is J3And K4The connection coefficient of the node;
Wj4k3,Wk3j4is J4And K3The connection coefficient of the node;
Wk3k4,Wk4k3is K3And K4The connection coefficient of the node;
4. The method for flavor recognition based on olfactory-taste synaptic perception model according to claim 1, wherein the method comprises
Characterized in that the topological structure of the connection structure module is a multi-input structure mode, the connection structure module receives taste information of an island cortex module of a taste nervous system and an island cortex module of a thalamus ventral posterior medial nucleus module and smell information of an outer cortex module of an olfactory nervous system, and a dynamic characteristic description model of an N node of the smell-taste connection structure module is a differential equation of a formula (22):
wherein, a and b1All are differential equation coefficients, and s is the number of channels of the taste perception model minus 1;
Nl(t) is the potential state of the nth node of the l channel, l =1,2, …, N; n is the number of channels of the taste perception model;
Q(K3(t)), Q (M (t)), Q (C (t)) is K3The output of node M, C;
Q(Nj(t)) is the output of the N node of the j channel, wherein l takes on the value from 1 to s;
Wnnconnecting coefficients among N nodes of all channels;
Wnk3is K3And the connection coefficient of the N node;
Wnmthe connection coefficient of the N node and the M node is obtained;
Wncthe connection coefficient of the N node and the C node is obtained;
q (C (t)) is the output of the skin layer module, which is represented by the C node;
q (M (t)) is the output of the island cortex module, which is represented by the M node.
5. The method for flavor recognition based on olfactory-taste synaptic perception model according to claim 1, wherein the method comprises
Characterized in that in the olfactory-olfactory synaesthesia perception model, the trigeminal nerve module J5Node and isolated beam nucleus module J1Node association, J5The node outputs the excitability information to J1A node;
b of front piriform cortex module3J for outputting inhibitory olfactory information to posterior medial nucleus module of thalamus abdomen3A node, the fiber link connecting the sniff-gustatory perception model;
fromOlfactory information of the cortex module and gustatory information of the islet cortex module are input to the N node of the olfactory-gustatory linkage structure module in a linear superposition manner, and the N node receives olfactory information from the thalamus retroventral nucleus module K3An output of the node;
lateral connection exists among N nodes of all channels of the olfactory-olfactory joint perception model, and each channel N node receives excitatory output from other channel N nodes.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011070975.3A CN111950721B (en) | 2020-10-09 | 2020-10-09 | Flavor identification method based on smell-taste perception model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011070975.3A CN111950721B (en) | 2020-10-09 | 2020-10-09 | Flavor identification method based on smell-taste perception model |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111950721A true CN111950721A (en) | 2020-11-17 |
CN111950721B CN111950721B (en) | 2024-05-31 |
Family
ID=73357005
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011070975.3A Active CN111950721B (en) | 2020-10-09 | 2020-10-09 | Flavor identification method based on smell-taste perception model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111950721B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112598116A (en) * | 2020-12-22 | 2021-04-02 | 王槐林 | Pet appetite evaluation method, device, equipment and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108981800A (en) * | 2018-06-25 | 2018-12-11 | 东北电力大学 | It is a kind of to smell-gustatism effect visualization method using neurodynamics system model progress machine |
WO2019202532A2 (en) * | 2018-04-18 | 2019-10-24 | Johnson & Johnson Consumer Inc. | Multifunctional benefits of fragrance |
CN111475936A (en) * | 2020-04-03 | 2020-07-31 | 东北电力大学 | Taste perception model-based taste recognition method |
-
2020
- 2020-10-09 CN CN202011070975.3A patent/CN111950721B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2019202532A2 (en) * | 2018-04-18 | 2019-10-24 | Johnson & Johnson Consumer Inc. | Multifunctional benefits of fragrance |
CN108981800A (en) * | 2018-06-25 | 2018-12-11 | 东北电力大学 | It is a kind of to smell-gustatism effect visualization method using neurodynamics system model progress machine |
CN111475936A (en) * | 2020-04-03 | 2020-07-31 | 东北电力大学 | Taste perception model-based taste recognition method |
Non-Patent Citations (2)
Title |
---|
包晓敏;曹作宝;汪亚明;周砚江;朱寒宇;: "基于嗅觉神经网络的织物组织识别", 纺织学报, vol. 32, no. 04, 15 April 2011 (2011-04-15) * |
李光;傅均;张佳;: "基于嗅觉模型的电子鼻仿生信息处理技术研究进展", 科学通报, vol. 53, no. 22, 30 November 2008 (2008-11-30) * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112598116A (en) * | 2020-12-22 | 2021-04-02 | 王槐林 | Pet appetite evaluation method, device, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN111950721B (en) | 2024-05-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Preedy | Electronic noses and tongues in food science | |
JP6682975B2 (en) | Gas analyzer and gas analysis method | |
Rudnitskaya et al. | Instrumental measurement of beer taste attributes using an electronic tongue | |
Eklöv et al. | Monitoring sausage fermentation using an electronic nose | |
CN108981800B (en) | Method for visualizing machine smell-taste perception effect by using neuro-dynamics system model | |
CN111950721A (en) | Flavor identification method based on smell-taste joint perception model | |
Lu et al. | Grade classification model tandem BpNN method with multi-metal sensor for rice eating quality evaluation | |
Bhattacharyya et al. | Electronic nose and electronic tongue | |
Zheng et al. | Olfactory-taste synesthesia model: An integrated method for flavor responses of electronic nose and electronic tongue | |
Zheng et al. | Computational model of taste pathways: A biomimetic algorithm for electronic tongue based on nerve conduction mechanism | |
CN111475936B (en) | Taste recognition method based on taste perception model | |
Morin et al. | Use of experienced panelists and the projective mapping task in comparison to trained panelists and naïve consumers | |
Liu et al. | Using spike-based bio-inspired olfactory model for data processing in electronic noses | |
Zheng et al. | A data processing method for electronic tongue based on computational model of taste pathways and convolutional neural network | |
Xiao et al. | Discrimination of Xihulongjing tea grade using an electronic tongue | |
Tudu et al. | Normalization techniques for gas sensor array as applied to classification for black tea | |
Nagamalla et al. | Detection of adulteration in food using recurrent neural network with internet of things | |
Zheng et al. | Taste sensation evaluation for an electronic tongue based on an optimized computational model of taste pathways | |
CN116662736A (en) | Human body state assessment method based on deep learning hybrid model | |
CN113609775B (en) | Quantitative forecasting method for delicious salivation sense sensory evaluation score of solid compound seasoning | |
Zheng et al. | An olfactory-taste synesthesia model combined with electronic nose and electronic tongue to identify flavor substances | |
CN113380332B (en) | Characterization method and application of taste characteristics of mixed components | |
Ribeiro et al. | Effects of training and taper on neuromuscular fatigue profile on 100-m swimming performance | |
Guthrie | Machine olfaction | |
CN113504316B (en) | Method for evaluating main fragrance of Maillard reaction product by utilizing functional groups |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |