CN104483460A - Sensor array optimization method for meat detection - Google Patents
Sensor array optimization method for meat detection Download PDFInfo
- Publication number
- CN104483460A CN104483460A CN201410523077.7A CN201410523077A CN104483460A CN 104483460 A CN104483460 A CN 104483460A CN 201410523077 A CN201410523077 A CN 201410523077A CN 104483460 A CN104483460 A CN 104483460A
- Authority
- CN
- China
- Prior art keywords
- sensor array
- data
- meat
- sensor
- analysis
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Abstract
The invention provides a sensor array optimization method for meat detection. The method comprises the following steps: S1, quality index data of a detected meat sample is acquired; S2, odour of the detected meat sample is gathered through a sensor array so as to obtain response data of the sensor array; S3, the response data undergoes feature extraction to obtain feature data of the sensor array; S4, the feature data undergoes principal component analysis to obtain a storage days and principal component model of the detected meat sample, and the feature data undergoes load analysis according to the storage days and principal component model so as to obtain contribution rate of each sensor in the sensor array; S5, correlation analysis between the feature data of the sensor array and the quality index data is carried out; and S6, the sensor array is optimized according to the contribution rate and the result of the correlation analysis. According to the invention, effects of sensor quantity simplification and identification and meat freshness rapid determination can be achieved.
Description
Technical field
The present invention relates to field of food detection, particularly relate to a kind of sensor array optimization method detected for meat.
Background technology
The smell of meat is the volatile ingredient that muscle from-inner-to-outer distributes, and the change of external smell reflects the change of muscle interior tissue, and consumer usually adopts and smells news method to pass judgment on the quality of measurement techniques for quality detection of meat.But not only subjectivity is strong for the quality evaluating meat by artificial sensory evaluation method, reappearance is also poor; Conventional physico-chemical analysis and the poisonous and harmful reagent involved by microorganism detection many, complex operation step, time and effort consuming; The demand that quick nondestructive passes judgment on quality can not be met.By adopting Artificial Olfactory (smell sensors array, also Electronic Nose is claimed) can the change of cardinal principle that occurs of objective quick identification meat odour component, and then express-analysis and assessment are carried out to quality controls such as its freshness, adulterated composition, microorganism, Local Geographical Indications.The response sensor array that selection is suitable for is the important prerequisite of pattern-recognition, but be equipped with 6 ~ 18 gas sensor arrays do not waited in existing electric nasus system, not only increase equipment development cost, and make the output signal of unnecessary sensor array disturb overall pattern recognition result.
Summary of the invention
(1) technical matters that will solve
The technical problem to be solved in the present invention is to provide a kind of sensor array optimization method detected for meat, can optimize advantage identification sensor array, reaches the effect of simplifying identification sensor quantity and freshness of meat and judging fast.
(2) technical scheme
For solving the problems of the technologies described above, the invention provides a kind of sensor array optimization method detected for meat, comprising:
S1: the index of quality data obtaining tested meat sample;
S2: gathered by the smell of sensor array to described tested meat sample, obtain the response data of sensor array;
S3: carry out feature extraction to described response data, obtains the characteristic of described sensor array;
S4: the storage number of days principal component model that principal component analysis (PCA) obtains described tested meat sample is carried out to described characteristic, according to described storage number of days principal component model, the contribution rate that load on analysis obtains each sensor in described sensor array is carried out to described characteristic;
S5: carry out correlation analysis between the characteristic of described sensor array and described index of quality data;
S6: the result according to described contribution rate and described correlation analysis is optimized described sensor array.
Further, also comprise after step S6:
Carry out back appraising to the sensor array after described optimization and analyse.
Further, the index of quality data of described tested meat sample comprise at least one in sensory evaluation data, physico-chemical analysis data and microorganism detection data.
Further, described physico-chemical analysis data are TVBN data.
Further, described tested meat sample is pork.
(3) beneficial effect
The present invention adopts the contribution rate of load on analyte sensors array under existing pattern-recognition, and by setting up the correlativity of sensor array response and the index of quality, screening and optimization advantage identification sensor array, not only reach and simplify identification sensor quantity, also reach the effect that pork freshness judges fast simultaneously.
Accompanying drawing explanation
Fig. 1 is the schematic diagram of a kind of sensor array optimization method for meat detection that embodiment of the present invention provides;
Fig. 2 is the schematic diagram of the change of storage pork sense organ overall acceptance score value in embodiment of the present invention;
Fig. 3 is the schematic diagram of the change of TVBN in the storage pork in embodiment of the present invention;
Fig. 4 is the schematic diagram of the change of storage pork total plate count in embodiment of the present invention;
Fig. 5 is the schematic diagram that changes of sensor array of the advantage response in embodiment of the present invention;
Fig. 6 is the schematic diagram that 10 sensor arraies in embodiment of the present invention carry out PCA pattern-recognition;
Fig. 7 is the schematic diagram that 10 sensors in embodiment of the present invention carry out the sensor contribution rate of pattern-recognition;
Fig. 8 is the schematic diagram of rear 6 the sensor array recognition results of the optimization in embodiment of the present invention.
Embodiment
Below in conjunction with drawings and Examples, the specific embodiment of the present invention is described in further detail.Following examples for illustration of the present invention, but are not used for limiting the scope of the invention.
Fig. 1 is the schematic diagram of a kind of sensor array optimization method for meat detection that embodiment of the present invention provides, and comprising:
S1: the index of quality data obtaining tested meat sample;
S2: gathered by the smell of sensor array to described tested meat sample, obtain the response data of sensor array;
S3: carry out feature extraction to described response data, obtains the characteristic of described sensor array;
S4: the storage number of days principal component model that principal component analysis (PCA) obtains described tested meat sample is carried out to described characteristic, according to described storage number of days principal component model, the contribution rate that load on analysis obtains each sensor in described sensor array is carried out to described characteristic;
S5: carry out correlation analysis between the characteristic of described sensor array and described index of quality data;
S6: the result according to described contribution rate and described correlation analysis is optimized described sensor array.
Preferably, also comprise after step S6:
Carry out back appraising to the sensor array after described optimization and analyse.
Wherein, the index of quality data of described tested meat sample comprise at least one in sensory evaluation data, physico-chemical analysis data and microorganism detection data.
Wherein, described physico-chemical analysis data are TVBN data.
Such as, the sensor array for pork detection can be optimized, specifically comprise the following steps:
Steps A: obtain tested pork sample
The unified back leg lean meat getting hog on hook carries out aerobic pallet packing, transports analysis room in 0 ~ 4 DEG C of ice temperature.By Sample Storage in 4 DEG C of High Precision Low Temperature incubators, under duration of storage gets each temperature at set intervals at random, 3 box porks carry out the attributional analysises such as sense organ, physics and chemistry and microorganism.
Step B: meat quality analysis
Step B1: obtain sensory evaluation data
Sensory evaluation data can be obtained with reference to prior art, individual event and overall acceptance evaluation can be carried out to the index such as color, quality, elasticity, smell of tested meat sample respectively.The higher freshness of score value is better, and 10 is best, and 0 is very corrupt, and 6.5 is sense organ acceptable limits.As shown in Figure 2, through the comprehensive evaluation of sense organ personnel to the index such as color, quality, elasticity, smell of pork, obtain the overall acceptance of pork, namely whether can accept it as consumable products.Known through evaluation analysis, before the 3rd day, pork is in high quality life, and within the 7th day, Organoleptic acceptability just reaches 6.55 points, namely can connect restricted edge, and after 7 days, pork is in the sense organ unacceptable phase.
Step B2: physico-chemical analysis
In storage, under endogenous enzymes in pork and microorganism acting in conjunction, muscle protein decomposes alkaline nitrogen substance such as generation ammonia and amine etc. and is called total volatile basic nitrogen (total volatilebasic nitrogen, TVBN), it is the putrid and deteriorated common counter of meat protein, is listed in the evaluation criterion of freshness of meat.TVBN content is relevant with microbial growth and breaks down proteins.TVBN content carries out semimicro by the filtrate after the extraction of employing perchloric acid and determines nitrogen analysis mensuration, concrete steps are as follows: accurately take 10g mixing meat sample, add 90mL perchloric acid, homogeneous, dipping 30min, in the centrifugal 15min of 3000rpm, filter, get 5mL filtrate and distill 6min in semi-automatic kjeldahl apparatus, the HCl solution of receiving liquid 0.01mol/L is titrated to neutrality, by HCl solution usage conversion TVBN content (mg/100g).
As can be seen from Figure 3, the initial TVBN value about 12mg/100g of pork, before 7 days, TVBN value is all no more than 15mg/100g, and within the 7th day, arrive 20.17mg/100g, TVBN value increases sharply subsequently, reaches storage maximal value 62.47mg/100g in latter stage; Show that the spoilage rate of pork increases.According to NYT632-2002 regulation, the TVBN of chilled pork should be no more than 15 mg/100g.Namely 7 days is the unacceptable phase later.
Step B3: microorganism detection
Microorganism detection is with reference to GB4789.20-2003 microbiological test of food hygiene method, all operations carries out all under aseptic conditions, get 25g pork sample and be placed in 225mL stroke-physiological saline solution, shaken well, do 10 times of serial dilutions, at least choose three dilute concentrations to inoculate, each dilutability do two parallel.Psychrotrophs adopts 20 DEG C, CVT agar to cultivate 72h, calculates colonies typical quantity, is converted into logarithm value lg (CFU/g).As shown in Figure 4, under 4 DEG C of holding conditions, along with storage time extends, total plate count presents the exponential growth curve of slight fluctuations.About 3 days of its lag phase, total plate count presents exponential phase fast subsequently, tapers off to 8.45lgCFU/g after reaching maximal value 9.43lgCFU/g.7th day, total plate count quantity reached 7.10, and microbial safety reaches restriction, and therefore, after 7 days, microorganism is in putrefaction stage.
Showing from the interpretation of result of sensory evaluation, physico-chemical analysis and microorganism detection ,≤3 days is high quality life, within 4 ~ 6 days, is time fresh phase, within >=7 days, is incipient spoilage or corrupt phase.
Step C: sensor array system identification meat quality changes
Step C1: smell collection and detection
In logistics progress, the detection method of pork smell can not, in lab analysis, therefore need to adopt portable odor and acquisition method.Each each 3 pieces of pork getting different storage time under different reserve temperature at random, is cut into meat cubelets.Accurately take meat cubelets 15g meat cubelets, be laid in 150mL beaker respectively, seal immediately with double-deck preservative film, room temperature starts after leaving standstill 10min to measure.Each sample repeats 3 times.The workflow of sensor array system is: sample sealing balances after a period of time until smell, headspace gas is pumped in detection system through sampling channel, sensor is because having adsorbed a certain amount of volatile matter, conductivity changes, and this signal is obtained by data acquisition system (DAS) and is stored in computing machine.After sampling, the pure air after activated carbon filtration is pumped to Electronic Nose, cleans and make it return to original state to sensor, for measurement is next time prepared.Wherein, when Electronic Nose is measured, detection time is 60s, and scavenging period is 60s.60s scavenging period fully can ensure that each sensor of Electronic Nose recovers its original state.
Step C2: sensor array response feature extraction
When sensor array system measures pork odor characteristics, obtain the response collection of illustrative plates of each sensor array, in collection of illustrative plates, each curve represents a sensor to the response of smell, when point on curve represents odour component by sensor passage, relative resistance rate (G/G0) is with the situation of change of storage time.For each response curve, from zero initial gas to the stationary process of final sample gas, relative resistance rate in the early stage (within 20 seconds) increases fast, and then tend to be steady (after 35 seconds).By the response curve of each sensor array under the different reserve temperature of Integrated comparative, after 35 seconds, each advantage sensor response tends towards stability and metastable state, therefore the signal of under steady state (SS) 40 ~ 45 seconds can be adopted as the feature extraction of this model construction, obtain the characteristic of sensor array.
4 DEG C of duration of storage, sensor array system is adopted to gather the head space air of pork, the curve synoptic diagram of sensor array to the characteristic response of the smell that difference storage number of days produces can be obtained, wherein, to preserve 1 day, 5 days, it within 13 days, is example, advantage sensor array response is respectively: < 1.5, < 2.5, > 4.0, show the prolongation of sensor array response along with storage time, odorousness constantly increases and increases, using 40 ~ 45 seconds curve stationary phases for feature extraction data are as further analysis, as Fig. 5 result shows, along with the prolongation of storage time, the sensor array of advantage response changes.Time fresh, pork smell is light, nearly all sensor has good response, along with the prolongation of storage time, there is response in various degree successively in No. 8, No. 6, No. 4, No. 2 sensors, but the kind of sensor of advantage response almost power that is consistent and response is followed successively by: No. 8, No. 6, No. 4, No. 2, the response of other sensors to smell is lower.Because sensor array has good response to different storage time meat quality deterioration process, for follow-up quality judging provides good data basis.
Step C3: pattern-recognition
In order to intactly reflect the quality grade of pork, first the present invention carries out principal component analysis (PCA) to the characteristic of each sensor, obtain the storage number of days principal component model of tested meat, in most certificate of comforming, find out two major components that can reflect quality grade and data correlation.Principal component analysis (PCA) (Principal component analysis, PCA) be a kind of statistical method of dimensionality reduction, when the information without any relevant sample can provide, PCA can browse rapidly all data, pass through dimensionality reduction, explain the variable that sample room is potential and factor by little index, find out the feature be associated between data, summarize rapidly the model information that can explain.This method is chosen the first two major component and is analyzed, PC1 and PC2 contains the contribution rate of first principal component and the Second principal component, obtained in PCA conversion.Contribution rate is larger, illustrates that this composition more can reflect the information of original multi objective preferably.If the contribution rate of accumulative total of the first two major component is greater than 80%, then can analyze for the coefficient in the first two major component.
In order to intactly reflect the quality grade of pork, first the present invention adopts the characteristic of 10 sensors to carry out principal component analysis (PCA), finds out two major components that can reflect quality grade and data correlation in most certificate of comforming.PCA is a kind of statistical method of dimensionality reduction, when the information without any relevant sample can provide, PCA can browse rapidly all data, pass through dimensionality reduction, the variable that sample room is potential and factor is explained by little index, find out the feature be associated between data, summarize rapidly the model information that can explain.In Fig. 6, each ellipse represents the data acquisition of different storage time.Look at from PC1 and PC2, although obvious linear distribution does not appear in each set of data points, the sample area calibration of each different times is good, and the close set of data points of quality is more close, and data acquisition Clustering Effect as each in Fig. 6 is good.When adopting PCA to analyze, the ratio of PC1 or PC2 is larger, shows that the contribution of this major component to model is larger.From Fig. 6 storage time PCA figure in can find out, the total contribution rate of diaxon is respectively 95.93%, shows that the method can distinguish the different quality grade of pork.
Step C4: sensor array optimization
After forming types model of cognition of the present invention, analyzed the contribution rate of each sensor under this pattern-recognition by load on, finally set up the correlativity between the index of quality and each sensor, for optimal screening advantage response sensor array.
Load on analysis is the related coefficient of major component and corresponding original index variable, for reflecting the level of intimate between each variable of Summing Factor.The present invention adopts load on analysis to weigh the importance of sensor in pattern-recognition or contribution rate.By observing the response of sensor in coordinate axis, its positional distance (0,0) point is far away, and namely load parameter value is larger, illustrates that the effect that sensor plays in assess sample quality is larger.If certain sensor is load parameter near-zero in pattern-recognition, the recognition capability of this sensor is negligible; If response is higher, this sensor is exactly identification sensor.As Fig. 7, load on analysis shows, maximum sensor is contributed to be respectively 8,6,4,2,9, No. 7 under different storage time pattern-recognition condition, sensitive to ethanol, methane, hydrogen, oxynitrides, organic sulfide, sulfide respectively, closely, therefore these identification sensors are comparatively sensitive to the peculiar smell response in meat quality deterioration process for the characterization of adsorption of sensor self and the corrupt smell component (aldehyde, ketone, alcohols etc. of amine, ammonia, sulfuretted hydrogen, short chain) of pork.
Correlation analysis is level of intimate or the relation of interdependence of the mutual relation of data of description variable, and related coefficient, for reflecting the simple correlation between two variablees, comprises the rank correlation between relevant and two grade variables between two continuous variables; Determine to optimize sensor array by the size of the related coefficient between sensor array and the index of quality, namely related coefficient is higher, and more stable with the correlativity of each index, can be considered as advantage identification sensor array.Because the response of sensor changes along with the kind of smell and the change of concentration, therefore, should be more reliable with screening advantage response sensor array in conjunction with other index of quality comprehensive evaluations.The internal association between each sensor and index of quality can be specified by correlation analysis.Table 1 is the correlation analysis result of each sensor and the index of quality, according to the change of the index of quality and sensor response in meat quality deterioration process, consider and storage time, Organoleptic acceptability, TVBN, the sensor array that total plate count correlativity is higher is followed successively by: 2>9>7>6GreatT. GreaT.GT4>8, 2>9>7>6GreatT. GreaT.GT4>8, 2>6, 2>7>9>6GreatT. GreaT.GT4>8, the correlativity of other sensors is low or and uncorrelated.Show, 2,9,7,6,4, No. 8 sensors and quality deterioration process have stable significant correlation.The contribution rate that this result and load on analyze identification sensor has good consistance, therefore, these six sensors is used for following model as selected sensor array and builds and pattern-recognition.
Table 14 DEG C the storage index of quality of pork and the correlation analysis of 10 sensor responses
| Sensor sequence number | Storage time | Organoleptic acceptability | TVBN | Total plate count |
| 1 | -.808** | .817** | -.710** | -.717** |
| 2 | .727** | -.747** | .652** | .709** |
| 3 | -.804** | .812** | -.714** | -.712** |
| 4 | .537* | -.559** | .493* | .574** |
| 5 | -.748** | .755** | -.676** | -.665** |
| 6 | .593** | -.610** | .558** | .597** |
| 7 | .668** | -.681** | .471* | .635** |
| 8 | .533* | -.551** | .484* | .570** |
| 9 | .688** | -.698** | .522* | .622** |
| 10 | 0.133 | -0.177 | 0.124 | 0.296 |
*, significant difference (P < 0.05); *, difference extremely significantly (P < 0.01)
Step C5: modelling verification
By gathering the smell of the chilled pork of known grade of freshness, the principal component model established is adopted to carry out back sentencing cluster analysis to known sample, judge that it sorts out accuracy by known data point and the Clustering Effect building model data set, return and sentence accuracy=(gross sample number-mistake sample number) × 100%/gross sample number.The reliability that accuracy carrys out evaluation model is sentenced by returning.Particularly, comprising:
Step C5-1: return and sentence model of cognition
Equally with 40 ~ 45 seconds of sensor response curve stationary phase for feature carries out data extraction, sensor array after adopting six to optimize carries out principal component analysis (PCA), obtain the differentiation design sketch as Fig. 8, the Clustering Effect of different storage time is good, along with the prolongation of storage time, from coordinate axis left-to-right limit, there is good linear relationship, namely 1,3,5 day good data acquisition of freshness closely, 9, the data acquisition of 11,13 days corrupt grades linearly increases along with PC1, and secondary fresh 7 days are in linear crossing with fresh and corrupt grade.Generally, each data acquisition linearly increases from left to right along with PC1, with PC2 has good discreteness from bottom to up, and total contribution rate of PC1+PC2 is 95.21%.Show the not only quantity minimizing of the sensor array after optimizing, and its clustering recognition effect has better linear relationship.
Step C5-2: model returns sentences accuracy
In order to verify the reliability of the sensor array after optimization further, the present invention adopts known sample to carry out back sentencing its validity identified of analysis verification.The sensor array of 6 stable significant correlations after optimizing is adopted to carry out back sentencing checking to the different grade of freshness of pork, result is as table 2, although the sample of fresh, secondary fresh and corrupt grade returns sentence accuracy all lower than 10 sensor arraies, but after optimizing, returning of sensor array is sentenced accuracy and is reached 82.5%, 77.2%, 82.2% respectively, and overall average accuracy reaches 80.6%.On the basis of having simplified 4 sensors, accuracy is sentenced in its time still can reach 80%, shows that these 6 sensors are advantage identification sensor arrays of meat quality deterioration, can meet the requirement passed judgment on fast.
Accuracy is sentenced in returning of the different pork grading of table 2 sensor array identification
Compared with prior art, the invention has the advantages that: adopt the contribution rate of load on analyte sensors array under existing pattern-recognition first, and by setting up the correlativity of sensor array response and the index of quality, screening and optimization advantage identification sensor array, and return and sentence accuracy and can reach more than 80%, not only reach and simplify identification sensor quantity, also reach the effect that pork freshness judges fast simultaneously.
Above embodiment is only for illustration of the present invention; and be not limitation of the present invention; the those of ordinary skill of relevant technical field; without departing from the spirit and scope of the present invention; can also make a variety of changes and modification; therefore all equivalent technical schemes also belong to category of the present invention, and scope of patent protection of the present invention should be defined by the claims.
Claims (5)
1., for the sensor array optimization method that meat detects, it is characterized in that, comprising:
S1: the index of quality data obtaining tested meat sample;
S2: gathered by the smell of sensor array to described tested meat sample, obtain the response data of sensor array;
S3: carry out feature extraction to described response data, obtains the characteristic of described sensor array;
S4: the storage number of days principal component model that principal component analysis (PCA) obtains described tested meat sample is carried out to described characteristic, according to described storage number of days principal component model, the contribution rate that load on analysis obtains each sensor in described sensor array is carried out to described characteristic;
S5: carry out correlation analysis between the characteristic of described sensor array and described index of quality data;
S6: the result according to described contribution rate and described correlation analysis is optimized described sensor array.
2. the sensor array optimization method detected for meat according to claim 1, is characterized in that, also comprise after step S6:
Carry out back appraising to the sensor array after described optimization and analyse.
3. the sensor array optimization method detected for meat according to claim 1, it is characterized in that, the index of quality data of described tested meat sample comprise at least one in sensory evaluation data, physico-chemical analysis data and microorganism detection data.
4. the sensor array optimization method detected for meat according to claim 1, it is characterized in that, described physico-chemical analysis data are TVBN data.
5. the sensor array optimization method detected for meat according to claim 1, it is characterized in that, described tested meat sample is pork.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410523077.7A CN104483460A (en) | 2014-09-30 | 2014-09-30 | Sensor array optimization method for meat detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410523077.7A CN104483460A (en) | 2014-09-30 | 2014-09-30 | Sensor array optimization method for meat detection |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104483460A true CN104483460A (en) | 2015-04-01 |
Family
ID=52758028
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410523077.7A Pending CN104483460A (en) | 2014-09-30 | 2014-09-30 | Sensor array optimization method for meat detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104483460A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109002859A (en) * | 2018-07-25 | 2018-12-14 | 郑州轻工业学院 | Sensor array feature selecting and array optimization method based on principal component analysis |
| CN109799269A (en) * | 2019-01-24 | 2019-05-24 | 山东工商学院 | Electronic nose gas sensor array optimization method based on behavioral characteristics different degree |
| CN113125657A (en) * | 2021-04-02 | 2021-07-16 | 中国肉类食品综合研究中心 | Refrigerated meat product spoilage monitoring and early warning system and application thereof |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO1999042820A1 (en) * | 1998-02-24 | 1999-08-26 | Wma Airsense Analysentechnik Gmbh | Method and device for identifying gaseous compounds |
| KR20030079351A (en) * | 2002-04-03 | 2003-10-10 | 노봉수 | Method for detecting the volatile compounds of irradiated meat by using electronic nose |
| CN1453584A (en) * | 2003-06-02 | 2003-11-05 | 江苏大学 | Fast non-destructive detection method and device of food smell based on gas sensor array technology |
| CN101419212A (en) * | 2008-12-08 | 2009-04-29 | 浙江大学 | Bird egg freshness detection device and method by utilizing volatile matter |
| CN102297930A (en) * | 2011-07-20 | 2011-12-28 | 浙江大学 | Method for identifying and predicting freshness of meat |
| CN103424526A (en) * | 2013-08-01 | 2013-12-04 | 浙江工商大学 | Device and method for detecting freshness of beef |
-
2014
- 2014-09-30 CN CN201410523077.7A patent/CN104483460A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO1999042820A1 (en) * | 1998-02-24 | 1999-08-26 | Wma Airsense Analysentechnik Gmbh | Method and device for identifying gaseous compounds |
| KR20030079351A (en) * | 2002-04-03 | 2003-10-10 | 노봉수 | Method for detecting the volatile compounds of irradiated meat by using electronic nose |
| CN1453584A (en) * | 2003-06-02 | 2003-11-05 | 江苏大学 | Fast non-destructive detection method and device of food smell based on gas sensor array technology |
| CN101419212A (en) * | 2008-12-08 | 2009-04-29 | 浙江大学 | Bird egg freshness detection device and method by utilizing volatile matter |
| CN102297930A (en) * | 2011-07-20 | 2011-12-28 | 浙江大学 | Method for identifying and predicting freshness of meat |
| CN103424526A (en) * | 2013-08-01 | 2013-12-04 | 浙江工商大学 | Device and method for detecting freshness of beef |
Non-Patent Citations (2)
| Title |
|---|
| 刘寿春等: "基于电子鼻技术判定冷鲜罗非鱼片品质劣变进程", 《食品科学》 * |
| 钟赛意等: "电子鼻系统识别罗非鱼品质劣变的传感器筛选研究", 《广东农业科学》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109002859A (en) * | 2018-07-25 | 2018-12-14 | 郑州轻工业学院 | Sensor array feature selecting and array optimization method based on principal component analysis |
| CN109002859B (en) * | 2018-07-25 | 2022-07-05 | 郑州轻工业学院 | Sensor array feature selection and array optimization method based on principal component analysis |
| CN109799269A (en) * | 2019-01-24 | 2019-05-24 | 山东工商学院 | Electronic nose gas sensor array optimization method based on behavioral characteristics different degree |
| CN109799269B (en) * | 2019-01-24 | 2023-09-22 | 山东工商学院 | Electronic nose gas sensor array optimization method based on dynamic feature importance |
| CN113125657A (en) * | 2021-04-02 | 2021-07-16 | 中国肉类食品综合研究中心 | Refrigerated meat product spoilage monitoring and early warning system and application thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hristov et al. | Symposium review: Uncertainties in enteric methane inventories, measurement techniques, and prediction models | |
| Rajamäki et al. | Application of an electronic nose for quality assessment of modified atmosphere packaged poultry meat | |
| Zhang et al. | Evaluation of peach quality indices using an electronic nose by MLR, QPST and BP network | |
| Panigrahi et al. | Design and development of a metal oxide based electronic nose for spoilage classification of beef | |
| CN105044298B (en) | A kind of Eriocheir sinensis class grade of freshness detection method based on machine olfaction | |
| Sohn et al. | Implementation of an electronic nose for continuous odour monitoring in a poultry shed | |
| CN102621192B (en) | Method for detecting freshness of mangos by aid of electronic nose | |
| Zhang et al. | Electronic nose with an air sensor matrix for detecting beef freshness | |
| Yongwei et al. | Monitoring storage time and quality attribute of egg based on electronic nose | |
| Wijaya et al. | Gas concentration analysis of resistive gas sensor array | |
| CN102323267A (en) | System and method used for rapidly evaluating freshness of raw meat products | |
| Balasubramanian et al. | Spoilage identification of beef using an electronic nose system | |
| CN104330382A (en) | Safety classification method of fresh beef | |
| CN107102035B (en) | Non-destructive testing method for pork frozen storage period based on electronic nose | |
| CN104483458A (en) | Method for predicting shelf life of cold chain pork and system thereof | |
| CN108254513A (en) | A kind of judgment method and device of fish shelf life | |
| CN104483460A (en) | Sensor array optimization method for meat detection | |
| Damuchali et al. | Evaluation of a field olfactometer in odour concentration measurement | |
| CN103983676A (en) | Quick pork freshness nondestructive testing method based on gas sensor technology | |
| CN106556680A (en) | It is a kind of based on Electronic Nose Technology fast prediction meat and its method for product fat oxidation degree | |
| CN104330441A (en) | Method and system for determining fish meat quality change | |
| Castrica et al. | Evaluation of smart portable device for food diagnostics: A preliminary study on Cape Hake fillets (M. Capensis and M. Paradoxus) | |
| Grassi et al. | E-sensing systems for shelf life evaluation: A review on applications to fresh food of animal origin | |
| CN114729910A (en) | Sensor tuning-sensor specific selection for internet of things e-nose application using gradient boosting decision trees | |
| Tan et al. | E-nose screening of pesticide residue on chilli and double-checked analysis through different data-recognition algorithms |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150401 |
|
| RJ01 | Rejection of invention patent application after publication |