CN103235087B - A kind of discrimination method of the oolong tea original producton location based on sense of smell and taste sensor information fusion - Google Patents
A kind of discrimination method of the oolong tea original producton location based on sense of smell and taste sensor information fusion Download PDFInfo
- Publication number
- CN103235087B CN103235087B CN201310124290.6A CN201310124290A CN103235087B CN 103235087 B CN103235087 B CN 103235087B CN 201310124290 A CN201310124290 A CN 201310124290A CN 103235087 B CN103235087 B CN 103235087B
- Authority
- CN
- China
- Prior art keywords
- information
- oolong tea
- sensor
- taste
- electrode
- 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.)
- Expired - Fee Related
Links
Landscapes
- Investigating Or Analyzing Materials By The Use Of Fluid Adsorption Or Reactions (AREA)
- Investigating Or Analyzing Materials By The Use Of Electric Means (AREA)
Abstract
The invention discloses a kind of oolong tea original producton location discrimination method based on multi-sensor information fusion.Concrete detection comprises the following steps: the oolong tea sample choosing a collection of different original producton location; Look quick type olfactory sensor system (be fixed on above hydrophobic membrane by 9 kinds of sensitive materials such as Porphyrin and its derivatives and make sensor array) is utilized to obtain the odiferous information of sample; The taste sensor be made up of in conjunction with contrast electrode again four kinds of working electrodes such as gold electrode, platinum electrode, copper electrode, glass-carbon electrodes is utilized to obtain the flavour information of sample; Then the two groups of Information data fusions will collected, set up the model of relevant information, can realize the discriminating (concrete steps and thinking are as Figure of abstract) of the oolong tea sample to different original producton location.The present invention utilizes sense of smell and the discriminating of sense of taste multi-sensor fusion technology realization to different original producton locations oolong tea sample, simultaneously easy and simple to handle, quick, intellectuality, message complementary sense, reliable results.
Description
Technical field
The present invention relates to a kind of quick, intelligent discrimination method of oolong tea original producton location of multi-sensor information fusion, namely utilize look quick type olfactory sensor and multi-electrode taste sensor information fusion technology to differentiate the oolong tea in different original producton location.
Background technology
Oolong tea is Chinese distinctive teas, mainly originates in the north of Fujian Province in Fujian, the south of Fujian Province and Guangdong, three, Taiwan province, and therefore oolong tea is different according to the place of production, can be divided into the south of Fujian Province oolong tea, the north of Fujian Province oolong tea, Guangdong oolong tea and Taiwan oolong tea.Oolong tea is semi-fermented tea, its fresh leaf forms through the green grass or young crops-fried green that withers-do-knead-drying, the original producton location of fresh leaf is different, owing to being subject to growing environment as temperature condition, the impact of the extraneous factors such as soil environment, there is certain difference in internal component, the flavor qualities such as the fragrance of the oolong tea that processing obtains and flavour also have certain difference, but this species diversity is very trickle.
Although Electronic Nose, electronic tongues technology are widely used in the flavor quality of tealeaves detects in recent years, but research is all adopt single-sensor technology mostly, because its quality discrepancy of oolong tea of Different sources is very trickle, single detection means often can not describe this species diversity all sidedly, has certain limitation.Current multi-sensor fusion technology obtains related application at food and agricultural products detection field.As authorized patent of invention " the quality of famous tea intelligent equipment based on multi-sensor data fusion evaluates method " (ZL200910232916.9) and open patent of invention " leaf vegetables blade pesticide residue detection device and method based on multi-sensor data fusion " (application number 201210265737.7), but literature search shows, current integration technology is all the Electronic Nose and the electronic tongue system that adopt comparatively perfect, they are all business-like universal instruments, and testing cost is high; Secondly, information fusion itself that is to say that different classes of sensing data simply superposes, and sensory judgments's behavior of the mankind differs greatly.The invention provides a kind of different original producton locations oolong tea discrimination method based on sense of smell and sense of taste multi-sensor information fusion.For oolong tea smell and flavor characteristics, build look quick type olfactory sensor and multi-electrode taste sensor system, obtain fragrance and the flavor characteristics information of oolong tea respectively, line number of going forward side by side Data preprocess and characteristic variable are extracted, then olfactory characteristic variable and taste characteristics variable will be extracted, a high-dimensional feature space is mapped to by unified approach, some dummy variables are taken out according to non-supervisory mode identification method, complete sense of smell and taste characteristics information fusion, build finally by suitable mode identification method and differentiate model based on Multi-information acquisition, can realize carrying out fast to different original producton locations oolong tea, intelligent discriminating.
Summary of the invention
Method of the present invention is as follows:
(1) oolong tea in different original producton location is chosen as sample;
(2) each sample utilizes 50mL boiling water to brew 5min, by the gas collection of enrichment to look quick type olfactory sensor system (being fixed on the sensor array that hydrophobic membrane forms by 9 kinds of Porphyrin and its derivatives), for the odiferous information of collecting sample; After tea grounds is filtered, obtain millet paste and after being cooled to normal temperature, utilize multi-electrode taste sensor system (being made up of in conjunction with contrast electrode again four kinds of working electrodes such as gold electrode, platinum electrode, copper electrode, glass-carbon electrodes) for the flavour information of collecting sample.
(3) the two groups of information collected are carried out data fusion, extract corresponding characteristic variable, by unified approach, these characteristic informations are mapped to a high-dimensional feature space, some dummy variables are taken out according to non-supervisory mode identification method, complete sense of smell and taste characteristics information fusion, build finally by Artificial Neural Network and differentiate model based on Multi-information acquisition.
(4) for a unknown sample, utilize olfactory sensor to gather odiferous information respectively, utilize taste sensor to gather flavour information, and carry out corresponding feature extraction and information fusion, input the discrimination model established, the discriminating treating this original producton location of test sample ownership can be realized.
Look quick type olfactory sensor has the quick material of the look of color reaction (9 kinds of Porphyrin and its derivatives: 1. 5 by preferred to oolong tea smell, 10, 15, 20-tetraphenyl-21H, 23H-porphines, 2. 5, 10, 15, 20-tetraphenyl-21H, 23H-porphines manganese chloride (III), 3. 2, 3, 7, 8, 12, 13, 17, 18-octaethyl-21H, 23H-porphines manganese (III) chloride, 4. 5, 10, 15, 20-tetra-(4-anisyl)-21H, 23H-porphines iron chloride (III), 5. 5, 10, 15, 20-tetraphenyl-21H, 23H-porphines iron chloride (III), 6. 5, 10, 15, 20-tetraphenyl-21H, 23H-porphines copper (II), 7. 5, 10, 15, 20-tetra-(pentafluorophenyl group)-21H, 23H-porphyrin iron chloride (III), 8. 5, 10, 15, 20-tetra-(4-anisyl)-21H, 23H-porphines cobalt (II), 9. 5, 10, 15, 20-tetraphenyl-21H, 23H-porphines (zinc)), and be fixed in the sensor array that hydrophobic membrane makes above, this sensor can overcome INVENTIONConventional metal-oxide smell sensor and easily be subject to humidity, temperature impact produces the shortcoming of baseline wander, multi-electrode taste sensor is by the preferred four kinds of working electrodes such as gold electrode, platinum electrode, copper electrode, glass-carbon electrode oolong tea flavour being had to characteristic response, again in conjunction with contrast electrode composition taste sensor array, this does not need to carry out pre-treatment, the integrated information of flavour can be obtained rapidly, exactly, highly sensitive, repeatability, good reliability.
The present invention utilizes nose and the tongue of look quick type olfactory sensor technology and multi-electrode taste sensor technical modelling people, realizes differentiating the quick, intelligent of different original producton locations oolong tea.Length consuming time, efficiency are low because of complex steps to the method overcome traditional Physico-chemical tests method, and single-sensor technology is because the acquisition of information shortcoming such as the accuracy of detection caused is low comprehensively.
Accompanying drawing explanation
Fig. 1 differentiates step and thinking based on the oolong tea original producton location of sense of smell and taste sensor information fusion.
The PCA classification results of Fig. 2 sense of smell and taste sensor information fusion.
Fig. 3 is based on the PCA classification results of single olfactory sensor information.
Fig. 4 based on four kinds of working electrodes such as PCA classification results gold electrode, platinum electrode, copper electrode, glass-carbon electrode of single taste sensor, then forms taste sensor array in conjunction with contrast electrode.
Embodiment
(1) sample of embodiment is respectively from the south of Fujian Province Anxi Tieguanyin Tea, Wuyi, Fujian rock clovershrub, single four original producton locations such as fir and Taiwan Dongding Oolong Tea of Guangdong phoenix.Each classification has 15 samples, totally 60 samples.
(2) each sample utilizes 50mL boiling water to brew 5min, the gas collection of enrichment (is fixed on by 9 kinds of Porphyrin and its derivatives the sensor array made above hydrophobic membrane to form: 1. 5 to look quick type olfactory sensor system, 10, 15, 20-tetraphenyl-21H, 23H-porphines, 2. 5, 10, 15, 20-tetraphenyl-21H, 23H-porphines manganese chloride (III), 3. 2, 3, 7, 8, 12, 13, 17, 18-octaethyl-21H, 23H-porphines manganese (III) chloride, 4. 5, 10, 15, 20-tetra-(4-anisyl)-21H, 23H-porphines iron chloride (III), 5. 5, 10, 15, 20-tetraphenyl-21H, 23H-porphines iron chloride (III), 6. 5, 10, 15, 20-tetraphenyl-21H, 23H-porphines copper (II), 7. 5, 10, 15, 20-tetra-(pentafluorophenyl group)-21H, 23H-porphyrin iron chloride (III), 8. 5, 10, 15, 20-tetra-(4-anisyl)-21H, 23H-porphines cobalt (II), 9. 5, 10, 15, 20-tetraphenyl-21H, 23H-porphines (zinc)), for the odiferous information of collecting sample, after tea grounds is filtered, obtain millet paste and after being cooled to normal temperature, utilize multi-electrode taste sensor (being made up of in conjunction with contrast electrode again four kinds of working electrodes such as gold electrode, platinum electrode, copper electrode, glass-carbon electrodes) for the flavour information of collecting sample.
(3) to the olfactory sensor information collected and taste sensor information; Carry out data prediction and characteristic variable extraction, olfactory characteristic variable and taste characteristics variable will be extracted, a high-dimensional feature space is mapped to by unified approach, then take out some dummy variables according to non-supervisory mode identification method, complete sense of smell and taste characteristics information fusion (concrete steps and thinking are as Fig. 1).By principal component analysis (PCA) (PCA) method, we can see, utilize the method PCA classifying quality (as Fig. 2) of sense of smell and taste sensor information fusion will significantly better than use single olfactory sensor technology (as Fig. 3) or single taste sensor technology (as Fig. 4).
(4) 40 independent samples (each classification has 10 samples) are utilized to verify the method, result shows the discrimination of the method to all 40 independent samples and reaches 100%, show that the method can be carried out intellectualityization to the oolong tea in different original producton location and be differentiated, and identification result to be much better than single olfactory sensor technology or single taste sensor technology.
Claims (2)
1., based on an oolong tea original producton location discrimination method for sense of smell and taste sensor information fusion, carry out according to following concrete steps:
(1) different original producton locations oolong tea is chosen as sample;
(2) each sample utilizes 50mL boiling water to brew 5min, by the gas collection of enrichment to look quick type olfactory sensor system, and the odiferous information of collecting sample; After tea grounds is filtered, obtain millet paste and after being cooled to normal temperature, utilize the flavour information of multi-electrode taste sensor collecting sample;
(3) the two groups of information collected are carried out data fusion, extract corresponding characteristic variable, set up the discrimination model of sense of smell and sense of taste Multi-information acquisition; Specifically data prediction and characteristic variable extraction are carried out to the olfactory sensor information collected and taste sensor information, the olfactory characteristic variable of extraction and taste characteristics variable are mapped to a high-dimensional feature space by unified approach, then some dummy variables are taken out according to non-supervisory mode identification method, complete sense of smell and taste characteristics information fusion, last Artificial Neural Network builds based on Multi-information acquisition discrimination model;
(4) for a unknown sample, utilize olfactory sensor to gather odiferous information respectively, utilize taste sensor to gather flavour information, and carry out corresponding feature extraction and information fusion, input the discrimination model established, can realize differentiating the oolong tea in different original producton location.
2. method according to claim 1, is characterized in that, look quick type olfactory sensor makes sensor array by there being the quick material of the look of color reaction to be fixed on above hydrophobic membrane to oolong tea smell, multi-electrode taste sensor be by oolong tea flavour by the gold electrode of characteristic response, platinum electrode, copper electrode and glass-carbon electrode four kinds of working electrodes again in conjunction with the taste sensor array that contrast electrode forms, described have the quick material of the look of color reaction to be selected from following 9 kinds of Porphyrin and its derivatives to oolong tea smell: 1. 5, 10, 15, 20-tetraphenyl-21H, 23H-porphines, 2. 5, 10, 15, 20-tetraphenyl-21H, 23H-porphines manganese chloride (III), 3. 2, 3, 7, 8, 12, 13, 17, 18-octaethyl-21H, 23H-porphines manganese (III) chloride, 4. 5, 10, 15, 20-tetra-(4-anisyl)-21H, 23H-porphines iron chloride (III), 5. 5, 10, 15, 20-tetraphenyl-21H, 23H-porphines iron chloride (III), 6. 5, 10, 15, 20-tetraphenyl-21H, 23H-porphines copper (II), 7. 5, 10, 15, 20-tetra-(pentafluorophenyl group)-21H, 23H-porphyrin iron chloride (III), 8. 5, 10, 15, 20-tetra-(4-anisyl)-21H, 23H-porphines cobalt (II), 9. 5, 10, 15, 20-tetraphenyl-21H, 23H-porphines zinc.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310124290.6A CN103235087B (en) | 2013-04-10 | 2013-04-10 | A kind of discrimination method of the oolong tea original producton location based on sense of smell and taste sensor information fusion |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310124290.6A CN103235087B (en) | 2013-04-10 | 2013-04-10 | A kind of discrimination method of the oolong tea original producton location based on sense of smell and taste sensor information fusion |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103235087A CN103235087A (en) | 2013-08-07 |
CN103235087B true CN103235087B (en) | 2015-10-28 |
Family
ID=48883138
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310124290.6A Expired - Fee Related CN103235087B (en) | 2013-04-10 | 2013-04-10 | A kind of discrimination method of the oolong tea original producton location based on sense of smell and taste sensor information fusion |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103235087B (en) |
Families Citing this family (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103472197B (en) * | 2013-09-10 | 2015-03-04 | 江苏大学 | Cross-perception information interaction sensing fusion method in intelligent bionic evaluation for food |
CN104132968A (en) * | 2014-08-15 | 2014-11-05 | 黑龙江八一农垦大学 | Identification method of rice geographical indications and application of identification method of rice geographical indications |
CN107886095A (en) * | 2016-09-29 | 2018-04-06 | 河南农业大学 | A kind of classifying identification method merged based on machine vision and olfactory characteristic |
CN106560699A (en) * | 2016-10-20 | 2017-04-12 | 中国计量大学 | Combined detection method used for identification of producing area of Wuyi rock tea |
CN106560701A (en) * | 2016-10-20 | 2017-04-12 | 中国计量大学 | Wuyi rock tea production place deep studying system based on five-hiding layer |
CN106560700A (en) * | 2016-10-20 | 2017-04-12 | 中国计量大学 | Machine learning method for identifying origin of Wuyi rock tea automatically |
CN108960315B (en) * | 2018-06-27 | 2021-08-06 | 河南农业大学 | Intelligent evaluation system and method for quality of cooked meat product |
CN109959653B (en) * | 2018-12-26 | 2022-03-18 | 云南中烟工业有限责任公司 | Bionic array sensor-based plant extract taste measurement method |
CN109781949A (en) * | 2019-01-15 | 2019-05-21 | 江苏大学 | A kind of method of discrimination in the tealeaves source area of view-based access control model, smell and sense of taste sensor information fusion |
CN110426389B (en) * | 2019-08-13 | 2022-02-01 | 宿州学院 | Method for quickly identifying adulterated pork in beef based on visual olfaction technology |
CN111766212B (en) * | 2020-07-26 | 2023-05-12 | 中南民族大学 | Method for identifying green tea with different names by using porphyrin ultraviolet probe |
CN113970546A (en) * | 2021-09-03 | 2022-01-25 | 江苏大学 | Visual sensing and distinguishing method for green tea quality based on olfactory-gustatory interaction |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101419166A (en) * | 2008-11-18 | 2009-04-29 | 江苏大学 | Tea quality nondestructive detecting method and device based on near-infrared spectrum and machine vision technology |
CN101692053A (en) * | 2009-10-09 | 2010-04-07 | 江苏大学 | Multi-sensing information fusion based instrumental intelligent evaluation method for quality of famous tea |
CN102539326A (en) * | 2012-01-13 | 2012-07-04 | 江苏大学 | Method for carrying out quantitative evaluation on soup hue quality of tea |
CN102721793A (en) * | 2012-06-11 | 2012-10-10 | 江苏大学 | Method and device for digitally detecting quality of edible vinegar |
CN102879445A (en) * | 2012-09-26 | 2013-01-16 | 江苏大学 | Method for identifying millet wine storage time based on multi-electrode sensing technology |
CN102967597A (en) * | 2012-09-26 | 2013-03-13 | 江苏大学 | Olfactory imaging sensing technology based yellow wine storage time identification method and identification system |
-
2013
- 2013-04-10 CN CN201310124290.6A patent/CN103235087B/en not_active Expired - Fee Related
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101419166A (en) * | 2008-11-18 | 2009-04-29 | 江苏大学 | Tea quality nondestructive detecting method and device based on near-infrared spectrum and machine vision technology |
CN101692053A (en) * | 2009-10-09 | 2010-04-07 | 江苏大学 | Multi-sensing information fusion based instrumental intelligent evaluation method for quality of famous tea |
CN102539326A (en) * | 2012-01-13 | 2012-07-04 | 江苏大学 | Method for carrying out quantitative evaluation on soup hue quality of tea |
CN102721793A (en) * | 2012-06-11 | 2012-10-10 | 江苏大学 | Method and device for digitally detecting quality of edible vinegar |
CN102879445A (en) * | 2012-09-26 | 2013-01-16 | 江苏大学 | Method for identifying millet wine storage time based on multi-electrode sensing technology |
CN102967597A (en) * | 2012-09-26 | 2013-03-13 | 江苏大学 | Olfactory imaging sensing technology based yellow wine storage time identification method and identification system |
Also Published As
Publication number | Publication date |
---|---|
CN103235087A (en) | 2013-08-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103235087B (en) | A kind of discrimination method of the oolong tea original producton location based on sense of smell and taste sensor information fusion | |
CN103196954B (en) | A kind of zhenjiang vinegar storage time discrimination method based on sense of smell and taste sensor information fusion | |
CN110133050B (en) | Method for qualitatively and quantitatively detecting tea quality based on multi-sensor fingerprint spectrum | |
CN101692053B (en) | Multi-sensing information fusion based instrumental intelligent evaluation method for quality of famous tea | |
Zhuang et al. | Identification of green tea origins by near-infrared (NIR) spectroscopy and different regression tools | |
CN104111274A (en) | Method for identifying producing area of red bayberry juice by using gas sensor array type electronic nose fingerprint analysis system | |
Neumann et al. | The colors of heath flowering–quantifying spatial patterns of phenology in Calluna life‐cycle phases using high‐resolution drone imagery | |
CN109164187A (en) | A method of distinguishing same type different sources tealeaves | |
CN109164188A (en) | A method of distinguishing different producing area and fermented type oolong tea | |
CN104678019A (en) | Panax notoginseng root tuber production place identification method based on stable isotope fingerprints | |
CN103674638A (en) | Method for rapidly identifying years of production of lycium barbarum by gustation finger-prints | |
CN103558311A (en) | Bitter and astringent green tea flavors judging method based on tea leaf biochemical components | |
CN104132968A (en) | Identification method of rice geographical indications and application of identification method of rice geographical indications | |
CN105158424A (en) | Method for rapidly identifying authenticity and quality of Cordyceps sinensis | |
CN103389323A (en) | Method for evaluating ages of precious medicinal materials quickly and losslessly | |
CN101158657B (en) | Tea-leaf producing area identification method based on X-ray fluorescence technology | |
Soh et al. | Development of neural network-based electronic nose for herbs recognition | |
CN103399050A (en) | Method for rapidly evaluating ginseng-adulterated American ginseng based on mouth feel information | |
CN107578050A (en) | The automatic classifying identification method of rice basal part of stem On Planthopperss and its worm state | |
CN103376282B (en) | Taste information based method for rapid evaluation of ginsengs of different ages | |
Ariyama et al. | The determination technique of the geographic origin of Welsh onions by mineral composition and perspectives for the future | |
CN106324127B (en) | The pueraria root powder true and false identifies and the method for maca assay | |
CN108241016B (en) | Method and device for rapidly detecting theaflavin content in black tea | |
Heaney et al. | Tea and flavoured tea | |
Sarkar et al. | Taste recognizer by multi sensor electronic tongue: a case study with tea quality classification |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20151028 Termination date: 20200410 |