CN108022239A - A kind of sparkling wine browning detection device and method based on machine vision - Google Patents
A kind of sparkling wine browning detection device and method based on machine vision Download PDFInfo
- Publication number
- CN108022239A CN108022239A CN201711266254.8A CN201711266254A CN108022239A CN 108022239 A CN108022239 A CN 108022239A CN 201711266254 A CN201711266254 A CN 201711266254A CN 108022239 A CN108022239 A CN 108022239A
- Authority
- CN
- China
- Prior art keywords
- image
- sample
- mrow
- msub
- sum
- 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
- 235000015040 sparkling wine Nutrition 0.000 title claims abstract description 70
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000001514 detection method Methods 0.000 title claims abstract description 26
- 238000002835 absorbance Methods 0.000 claims abstract description 18
- 238000012545 processing Methods 0.000 claims abstract description 14
- 238000005260 corrosion Methods 0.000 claims abstract description 7
- 230000007797 corrosion Effects 0.000 claims abstract description 7
- 239000011159 matrix material Substances 0.000 claims description 46
- 235000014101 wine Nutrition 0.000 claims description 34
- 241000219095 Vitis Species 0.000 claims description 27
- 235000009754 Vitis X bourquina Nutrition 0.000 claims description 27
- 235000012333 Vitis X labruscana Nutrition 0.000 claims description 27
- 235000014787 Vitis vinifera Nutrition 0.000 claims description 27
- 238000012937 correction Methods 0.000 claims description 25
- 238000002372 labelling Methods 0.000 claims description 16
- 230000006870 function Effects 0.000 claims description 14
- 238000005070 sampling Methods 0.000 claims description 10
- NOEGNKMFWQHSLB-UHFFFAOYSA-N 5-hydroxymethylfurfural Chemical compound OCC1=CC=C(C=O)O1 NOEGNKMFWQHSLB-UHFFFAOYSA-N 0.000 claims description 9
- 230000001133 acceleration Effects 0.000 claims description 9
- 238000002474 experimental method Methods 0.000 claims description 6
- 230000003287 optical effect Effects 0.000 claims description 6
- 239000000203 mixture Substances 0.000 claims description 5
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 239000000284 extract Substances 0.000 claims description 4
- 239000006260 foam Substances 0.000 claims description 4
- LFULEKSKNZEWOE-UHFFFAOYSA-N propanil Chemical compound CCC(=O)NC1=CC=C(Cl)C(Cl)=C1 LFULEKSKNZEWOE-UHFFFAOYSA-N 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 4
- 239000004793 Polystyrene Substances 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 230000008676 import Effects 0.000 claims description 3
- 229920002223 polystyrene Polymers 0.000 claims description 3
- 230000001954 sterilising effect Effects 0.000 claims description 3
- 230000003628 erosive effect Effects 0.000 claims description 2
- 230000005251 gamma ray Effects 0.000 claims 1
- 238000004458 analytical method Methods 0.000 abstract description 12
- 230000008569 process Effects 0.000 abstract description 6
- 230000005540 biological transmission Effects 0.000 abstract description 5
- 239000003153 chemical reaction reagent Substances 0.000 abstract description 5
- HYBBIBNJHNGZAN-UHFFFAOYSA-N furfural Chemical compound O=CC1=CC=CO1 HYBBIBNJHNGZAN-UHFFFAOYSA-N 0.000 abstract description 2
- OKKJLVBELUTLKV-UHFFFAOYSA-N Methanol Chemical class OC OKKJLVBELUTLKV-UHFFFAOYSA-N 0.000 abstract 2
- 238000011160 research Methods 0.000 description 13
- 230000008859 change Effects 0.000 description 9
- 235000019994 cava Nutrition 0.000 description 7
- 235000013305 food Nutrition 0.000 description 7
- 238000003860 storage Methods 0.000 description 7
- 230000008602 contraction Effects 0.000 description 6
- 239000003814 drug Substances 0.000 description 6
- 238000012544 monitoring process Methods 0.000 description 6
- 230000032683 aging Effects 0.000 description 5
- 238000004519 manufacturing process Methods 0.000 description 5
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 4
- LFQSCWFLJHTTHZ-UHFFFAOYSA-N Ethanol Chemical compound CCO LFQSCWFLJHTTHZ-UHFFFAOYSA-N 0.000 description 4
- 230000008049 biological aging Effects 0.000 description 4
- 150000001875 compounds Chemical class 0.000 description 4
- 229940079593 drug Drugs 0.000 description 4
- 238000005259 measurement Methods 0.000 description 4
- 240000004808 Saccharomyces cerevisiae Species 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000000855 fermentation Methods 0.000 description 3
- 230000004151 fermentation Effects 0.000 description 3
- 239000003205 fragrance Substances 0.000 description 3
- 238000010438 heat treatment Methods 0.000 description 3
- 238000004128 high performance liquid chromatography Methods 0.000 description 3
- 238000005286 illumination Methods 0.000 description 3
- 239000003550 marker Substances 0.000 description 3
- 230000000877 morphologic effect Effects 0.000 description 3
- 230000001590 oxidative effect Effects 0.000 description 3
- ISWSIDIOOBJBQZ-UHFFFAOYSA-N Phenol Chemical compound OC1=CC=CC=C1 ISWSIDIOOBJBQZ-UHFFFAOYSA-N 0.000 description 2
- RAHZWNYVWXNFOC-UHFFFAOYSA-N Sulphur dioxide Chemical compound O=S=O RAHZWNYVWXNFOC-UHFFFAOYSA-N 0.000 description 2
- 239000008280 blood Substances 0.000 description 2
- 210000004369 blood Anatomy 0.000 description 2
- 229910002092 carbon dioxide Inorganic materials 0.000 description 2
- 239000001569 carbon dioxide Substances 0.000 description 2
- 238000004587 chromatography analysis Methods 0.000 description 2
- 238000013016 damping Methods 0.000 description 2
- 238000005530 etching Methods 0.000 description 2
- 239000000796 flavoring agent Substances 0.000 description 2
- 235000019634 flavors Nutrition 0.000 description 2
- 235000019674 grape juice Nutrition 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 235000013824 polyphenols Nutrition 0.000 description 2
- 230000001568 sexual effect Effects 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Chemical compound O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 235000020097 white wine Nutrition 0.000 description 2
- 235000021446 Apple puree Nutrition 0.000 description 1
- 102000004506 Blood Proteins Human genes 0.000 description 1
- 108010017384 Blood Proteins Proteins 0.000 description 1
- 102000004190 Enzymes Human genes 0.000 description 1
- 108090000790 Enzymes Proteins 0.000 description 1
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 description 1
- 240000008790 Musa x paradisiaca Species 0.000 description 1
- 241000220324 Pyrus Species 0.000 description 1
- 241000235342 Saccharomycetes Species 0.000 description 1
- 101100129590 Schizosaccharomyces pombe (strain 972 / ATCC 24843) mcp5 gene Proteins 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 238000000862 absorption spectrum Methods 0.000 description 1
- 230000009056 active transport Effects 0.000 description 1
- 150000001413 amino acids Chemical class 0.000 description 1
- 230000003078 antioxidant effect Effects 0.000 description 1
- 235000019606 astringent taste Nutrition 0.000 description 1
- 235000021015 bananas Nutrition 0.000 description 1
- 238000005452 bending Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000010364 biochemical engineering Methods 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 239000003518 caustics Substances 0.000 description 1
- 235000019993 champagne Nutrition 0.000 description 1
- 238000004737 colorimetric analysis Methods 0.000 description 1
- 238000004040 coloring Methods 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 239000007799 cork Substances 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000007872 degassing Methods 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 239000012153 distilled water Substances 0.000 description 1
- 238000001035 drying Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000002255 enzymatic effect Effects 0.000 description 1
- 239000000686 essence Substances 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- RJGBSYZFOCAGQY-UHFFFAOYSA-N hydroxymethylfurfural Natural products COC1=CC=C(C=O)O1 RJGBSYZFOCAGQY-UHFFFAOYSA-N 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000001727 in vivo Methods 0.000 description 1
- 150000002632 lipids Chemical class 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 239000007791 liquid phase Substances 0.000 description 1
- 235000020044 madeira Nutrition 0.000 description 1
- 230000002503 metabolic effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- VIKNJXKGJWUCNN-XGXHKTLJSA-N norethisterone Chemical compound O=C1CC[C@@H]2[C@H]3CC[C@](C)([C@](CC4)(O)C#C)[C@@H]4[C@@H]3CCC2=C1 VIKNJXKGJWUCNN-XGXHKTLJSA-N 0.000 description 1
- 238000011017 operating method Methods 0.000 description 1
- 235000021017 pears Nutrition 0.000 description 1
- 238000001556 precipitation Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 239000010453 quartz Substances 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N silicon dioxide Inorganic materials O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 229910001220 stainless steel Inorganic materials 0.000 description 1
- 239000010935 stainless steel Substances 0.000 description 1
- 238000010561 standard procedure Methods 0.000 description 1
- 238000004659 sterilization and disinfection Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 235000019640 taste Nutrition 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000032258 transport Effects 0.000 description 1
- 238000000870 ultraviolet spectroscopy Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30128—Food products
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
- Image Analysis (AREA)
Abstract
The problem of present invention proposes a kind of sparkling wine browning detection device and method based on machine vision, complicated to solve existing detection process, of high cost, and applicability is not strong;The sample being placed on using the camera collection of smart mobile phone on the diffused light source panel of vasculum bottom, obtain the digital picture of sparkling wine, the processing such as gray processing, binaryzation, corrosion carries out digital picture by machine vision and image processing techniques, by analyzing tri- passages of image RGB, quality status stamp using blue channel attenuation percentage as sparkling wine, its content height correlation with 2 furfural of the absorbance at 420nm and 5 methylol.The present invention only needs to obtain the transmission image of the sparkling wine sample in different brown stain stages, can be carried out at the same time the analysis of multiple brand sparkling wine samples, avoids the sample treatment of complexity, it can carry out Multi-example analysis, speed is fast, and cost is low, without expensive special instrument and other chemical reagent.
Description
Technical field
The present invention relates to sparkling wine brown stain and the technical field of image procossing, more particularly to one kind to be based on machine vision
Sparkling wine browning detection device and method.
Background technology
At this stage, nearly all also exist to the good opinion and demand of fizz in production sparkling wine, consumer all over the world
Constantly increase.The principle of fizz is to add sugar and yeast in the wine made to carry out second of alcohol hair in the container of closing
Ferment, fermentation process produce carbon dioxide be limited in bottle become wine in bubble source (Serra-Cayuela A,
Jourdes M,Riu-Aumatell M,et al.Kinetics of browning,phenolics,and 5-
hydroxymethylfurfural in commercial sparkling wines.[J].Journal of
Agricultural&Food Chemistry,2014,62(5):1159-1166).The sugar of addition is converted into by the yeast of addition
Alcohol and carbon dioxide so that bubble is produced and seethed, this is one of most important feature of sparkling wine.
Sparkling wine consumption figure rapid development, according to the statistics of international grape wine tissue, in recent years sparkling wine life
Production increases 40%, and fizz grape wine accounts for the 8% of global vintage.China and the U.S. become the master for promoting this growth
Power army.
Spain's slips and French champagne are foremost sparkling wine brands.Slips be by produced in conventional processes by
Name the high-quality fizz of regulation protection in original producton location.Grape used in slips wine is by strictly selecting, in juice-extracting process
It must be careful as, optimal grape juice could be so obtained in preferable maturity.Grape juice is then poured into stainless steel barrel
Cold fermentation is carried out, goes and tastes to determine its quality again after standing, and selectively allocated.Last allotment is carried out
Afterwards, wine is fitted into bottle, and is at least placed 9 months in wine cabinet, but usually this storage period can be longer.During this period, grape
Wine will produce secondary fermentation in bottle, form complicated organoleptic attribute such as fragrance, color and the foam performance of grape wine, become
Steep in wine, and produce yeast precipitation.After aging period, saccharomycete can carefully be taken out, is filled to bottleneck with a collection of grape wine,
With cork closure, and add strip of paper used for sealing.In addition, according to the duration of ageing in bottle, there are two kinds of different classifications:At least 15
The Reserva, Gran Reserva (Serra-Cayuela A, Aguilera-Curiel the M A, Riu- of at least 30 months of the moon
Aumatell M,et al.Browning during biological aging and commercial storage of
Cava sparkling wine and the use of 5-HMF as a quality marker[J].Food Research
International,2013,53(1):226-231).Except improving grape quality, the ageing time can also influence production cost,
The final price of aging wine is higher than new grape wine.
In terms of physical chemistry and organoleptic attribute, grape wine is a kind of dynamic products, therefore in biological ageing and is stored
Complicated chemical change occurs for Cheng Zhonghui.These chemical changes may cause the change of organoleptic feature, particularly fragrance, flavor and
Color (research [D] the Jinan of Wang Yu peaks Wine Aroma influence factors:Shandong Light Ind College, 2010).
Brown stain is to be related to oxidizing process (Li H, Guo A, the Wang H.Mechanisms of sugar, lipid, amino acid and phenol
of oxidative browning of wine[J].Food chemistry,2008,108(1):1-13.), brown stain reduces
Aesthetic quality's (color, the change of flavor and fragrance and increase of astringent taste) of grape wine, therefore during processing and storage
Brown stain must be controlled.Once bottle is sealed, brown stain cannot regulate and control again.Therefore, winery, which must be found, to have indicated
Steep in wine the mark and parameter of brown stain, and pass through the brown stain that technological means detects sparkling wine.It is necessary in actual production to open
The fast and reliable method of hair monitoring browning degree, and the parameter of definition instruction sparkling wine quality.
Existing research has been based on different quality status stamps and e measurement technology, it is proposed that several to be used to quantify and characterize
The method of pickled grape wine browning.As UV-Visible absorption spectrum (Ultraviolet-visible spectroscopy,
Abbreviation UV-VIS) and high performance liquid chromatography (High Performance Liquid Chromatography, abbreviation HPLC)
(Serra-Cayuela A,Aguilera-Curiel M A,Riu-Aumatell M,et al.Browning during
biological aging and commercial storage of Cava sparkling wine and the use of
5-HMF as a quality marker[J].Food Research International,2013,53(1):226-
231.)。
Absorbance (A at 420nm wavelength420) be used as white wine brown stain monitoring parameter (Kallithraka S,
Salacha M I,Tzourou I.Changes in phenolic composition and antioxidant
activity of white wine during bottle storage:Accelerated browning test versus
bottle storage[J].Food Chemistry,2009,113(2):500-505.), A420The increase of value and browning
Directly related (Ibarz A, Pagan J, Garza S.Kinetic models of non-enzymatic browning in
apple puree.[J].Journal of the Science of Food&Agriculture,2000,80(8):1162-
1168.).However, in previous studies, the A as the quality status stamp of slips sparkling wine420Parameter is proved to have low
Sensitivity and low specificity, use HMF 5 hydroxymethyl 2 furaldehyde (5-HMF) content instead as more effectively mark (Serra- afterwards
Cayuela A,Aguilera-Curiel M A,Riu-Aumatell M,et al.Browning during biological
aging and commercial storage of Cava sparkling wine and the use of 5-HMF as a
quality marker[J].Food Research International,2013,53(1):226-231.), but to the change
Compound, which carries out chromatography in laboratory, to be needed to consume substantial amounts of time and reagent, and cost is expensive.
In existing research method, fluorescent excitation-emmision spectra combination parallel transport is successfully used to monitor
The research of the browning of four kinds of sparkling wines, by as by monitoring A420Or 5-HMF contents realize brown stain detection
Quick alternative.In order to study more preferable alternative solution, researcher proposes a kind of new method based on Colorimetric techniques, uses
Browning in the slips sparkling wine that monitoring accelerates in brown stain experiment.Using fixed by International Commission on Illumination (CIE)
Lab color modes measurement browning degree (the Raquel P F that the Lab color modes or Richard.S.Hunter of justice are founded
G,Joao Barroca M.Evaluation of the Browning Kinetics for Bananas and Pears
Submitted to Convective Drying[J].Current Biochemical Engineering,2014,1:165-
172.), but these methods need to use special colorimeter or spectrophotometer.
In recent years, lifted with the sensor performance of smart mobile phone, the application study based on smart mobile phone is also increasing.Intelligence
Can mobile phone relative experimental room specialty instrument and equipment, with cost it is low, it is easy to carry, can collection in worksite and analysis, result it is easy
In share the advantages that.
The content of the invention
Brown stain detection for existing sparkling wine needs to use special colorimeter or spectrophotometer, and detection process is multiple
Miscellaneous, the not strong technical problem of applicability, the present invention proposes a kind of sparkling wine browning detection dress based on machine vision
Put and method, utilize machine vision and image processing techniques to carry out the detection of sparkling wine browning, it is only necessary to obtain different
The transmission image of the sparkling wine sample in brown stain stage, it is possible to be carried out at the same time point of the sparkling wine sample of multiple brands
Analysis, device is cheap, without expensive special instrument and other chemical reagent.
In order to achieve the above object, the technical proposal of the invention is realized in this way:A kind of blistering based on machine vision
Grape wine browning detection device and method, its step are as follows:
Step 1:Experimental provision is built, prepares sample;
Step 2:Sample image is obtained using the camera of smart mobile phone, by the RGB image IG of acquisitioniCarry out gray processing and
Binary conversion treatment, obtains binary image BINi, i=1,2 ..., 6;
Step 3:By image BINiIn pixel all negate, obtain image IRi;With structural element to IRiCarry out rotten
Erosion, obtains image IERi, the constriction zone in hole of the acquisition equipped with sample;
Step 4:By image IERiConnected component labeling is carried out, each type of sample is identified according to connected component labeling;
Step 5:With image IG1R passages, G passages and the B in region of contraction sample well of primary sample of standard lead to
Correction factor is calculated on the basis of the average in road, and to image IGiEach passage carries out color correction, obtains correction chart as IGCi, i=
2,…6;
Step 6:Calculate the image IGC after correctioniThe R, G in region of shrinkage hole of each brand sample, channel B
Average, chooses parameter of the channel B as brown stain;
Step 7:Channel B Color invariants and blueness decay percentage are calculated, blueness decay percentage has line with the time
Sexual intercourse, blueness decay percentage are consistent with the browning of sparkling wine, the quality status stamp as brown stain detection.
The sample is Brut, Brut Reserva, Brut Gran tetra- kinds of best seller brands of Reserva and Semiseco
Slips sparkling wine, four kinds of grape wine have different sugar contents and productive year;10mL is taken to be respectively charged into the amber of 4 20mL
In amber color bottle, and in N2Deaerate under air-flow;This 4 amber vials equipped with different brands fizz are placed on dark
In environment, and Conservation environment temperature is 8 DEG C;Using the fizz of the different brands in this 4 bottles as the experiment of follow-up 10 days
Standard sample, and by this 4 amber vials equipped with standard sample be named as Bottle1, Bottle2, Bottle3,
Bottle4;10mL is respectively taken to be respectively charged into the amber vial of 4 20mL the sparkling wine of four brands again, and in N2
Deaerate under air-flow;This 4 amber vials equipped with different brands fizz are placed on to the environment of complete darkness, are then placed
In an oven, 65 ± 1 DEG C of constant temperature is set, carries out accelerating brown stain experiment;By this 4 equipped with the amber small of acceleration brown stain sample
Bottle is named as Bottle5, Bottle6, Bottle7, Bottle8;Gathered by the sampling time point of 0,2,4,6,8 and 10 day
Sample message, when the time interval of collection is 48 small;In each sampling time point, (1) extracts Brut, Brut first
Each 1 part of Reserva, the standard sample of Brut Gran Reserva and Semiseco, are placed sequentially in 96 orifice plates from top to bottom
In the hole at middle part, 1 row are lined up vertically, are then named as the 1st row;(2) then extraction Brut, Brut Reserva, Brut Gran
Reserva and Semiseco acceleration brown stain each 4 parts of sample, then each brand put 1 row, be sequentially placed into the 2nd of 96 orifice plates
~5 row.
The experimental provision includes vasculum and light source, and light source is arranged on the lower section of vasculum, and collection lower box part, which is equipped with, to be used
Collection hole is equipped with the top of the orifice plate for holding sample, vasculum, collection hole is located at the surface of orifice plate.
The vasculum is made of black foam core plate, covered with dumb light black velvet paper inside vasculum;Vasculum
For the rectangular parallelepiped structure of height 80cm, the difference between diagonal and center vertical optical path is less than 5%;The light source can for intensity
The diffusing reflection light source of control;Orifice plate is 96 orifice plates, and orifice plate is manufactured by the pure polystyrene of import optical clear, at gamma rays sterilizing
Reason.
Obtain sample image method be:Slips sparkling wine sample is arranged in hole using 0.3ml micropipettes
The middle part of plate, by the panel of orifice plate placement light source, vasculum is covered on orifice plate, collection hole is located at the upper of the middle of orifice plate
Side, smart mobile phone is placed on the collection hole at the top of vasculum, using the built-in camera of smart mobile phone with highest resolution ratio
Image is obtained, and saves as JPEG;The primary sample arrangement of standard in the orifice plate, obtains all samples into same image.
The method that the RGB image of acquisition is carried out to gray processing and binary conversion treatment is:
The RGB image IG that smart mobile phone is obtainediIn each pixel tri- components of R, G, B average value as image
Gray value, i.e. Grayi(x, y)=(RIG,i(x,y)+GIG,i(x,y)+BIG,i(x, y))/3, obtain gray level image Grayi;
To gray level image GrayiOptimal threshold T is calculated using Otsu algorithms, optimal threshold T is so that the ash of σ maximums
Angle value gt:If the gray level of gray level image is L, then tonal range is [0, L-1], and gray level image is calculated most using Ostu methods
Good threshold value is:σ=Max [w0(gt)×(u0(gt)-u)^2+w1(gt)×(u1(gt)-u) ^2)], wherein, w0For prospect ratio,
u0For prospect gray average, w1For background ratio, u1For background gray average, u is the average of view picture gray level image.
T is the optimal threshold of segmentation figure picture, and gray level image is divided into 2 parts according to optimal threshold T, obtains binaryzation
Image BINi(x,y):
Wherein, Grayi(x, y) is the value of the pixel at (x, y) place in gray level image;
By binary image BINiInto negate (pixel value is 1 to be changed into 0, pixel value be 0 be changed into 1) obtain binary picture
As IRi;Using the imerode function pair binary images IR in MatlabiCarry out etching operation:IERi=imerode (IRi,
D1), wherein, IERiRepresent binary image IRiBy a diameter of 8 circular configuration element D1Obtained image, i=after corrosion
1,2,3,6;D1=strel (' disk', 4), disk represents circular shape.
By image IERiCarry out connected component labeling method be:Using connected component labeling function in Matlab
bwlabel:[LJi,LNUMi]=bwlabel (IERi, 8), wherein, LNUMiRepresent image IERiThe quantity of middle connected region, LJi
Represent and image IERiThe identical matrix of size, matrix L JiContain mark image IERiIn each connected region classification mark
Label, the value of these labels is 1,2 ..., LNUMi, 8 represent it is by 8 neighbor searching connected regions;
The method of the position of each type of sample of the identification is:The sample area that connected component labeling shrinks each
Domain is as a connected region, for each connected region, using the regionprops functions in Matlab:STATSi=
regionprops(LJi, ' Centroid'), ' Centroid' attributes are exactly the barycenter of connected region, STATSiContain image
IERiIn each connected region (centre of form coordinate (RXi,nu,LYi,nu), wherein, the value of subscript n u is 1~20.
The acquisition methods of correction chart picture are:With image IG1The R in region of contraction sample well of primary sample of standard lead to
The average R in road, G passages and channel B1, ave、G1, aveAnd B1, aveOn the basis of, in follow-up different time sampled point acquired image
Standard primary sample where the R passages in region of contraction sample well, the average of G passages and channel B be RI, ave、Gi,ave
And BI, aveFor reference, three correction factor C are calculatedi,R、Ci,GAnd Ci,B:
Wherein, Ci,R、Ci,G、Ci,BFor image IGiR passages, the correction factor of G passages and channel B, i=2,3 ..., 6;
By image IGiAll pixels point R passages RIG, iIt is multiplied by Ci,R, G passages GIG, iIt is multiplied by Ci,G, channel B BIG, iIt is multiplied by
Ci,B, i=2,3 ..., 6, obtain tri- passages of new RGB, the image of composition is correction chart as IGCi。
By image IER1In the row coordinate of centre of form coordinate be LY1,1~LY1,4The pixel of connected region put 1, it is other
Pixel is set to 0, and obtains matrix J Z1, by original image IG1In all pixels point R passages value as matrix RIG,1, by original
Beginning image IG1In all pixels point G passages value as matrix GIG,1, by original image IG1In all pixels point B
The value of passage is as matrix BIG,1;By matrix J Z1Dot product matrix RIG,1Obtain matrix D CR1, by matrix J Z1Dot product matrix GIG,1
To matrix D CG1, by matrix J Z1Dot product matrix BIG,1Obtain matrix D CB1;
R1, ave=sum (sum (DCR1))/sum(MJ1),
G1, ave=sum (sum (DCG1))/sum(MJ1),
B1, ave=sum (sum (DCB1))/sum(MJ1),
Wherein, sum (sum (DCR1)) it is matrix D CR1The summation of middle all elements value, sum (sum (DCG1)) it is matrix
DCG1The summation of middle all elements value, sum (sum (DCB1)) it is matrix D CB1The summation of middle all elements value, sum (MJ1) it is figure
As IER1In the row coordinate of centre of form coordinate be LY1,1~LY1,4Connected region sum of all pixels, MJ1=regionprops
(JZ1, ' Area'), ' Area' attributes are exactly the number of pixels in image regional.
By sample image IGCiChannel B be converted to respective color invariant:bi=BIGC,i,ave/(RIGC,i,ave+GIGC,i,ave
+BIGC,i,ave), calculate blueness decay percentage %Bt:
Wherein, bt0Channel B invariant when being t=0, btIt is the Color invariants of the channel B of time t;Blueness decay hundred
Divide and compare %BtRepresent that browning is:Y=Y0+ kt,
Wherein, Y is the absorbance or 5-HMF contents or %B at 420nmt, Y0It is that absorbance at 420nm or 5-HMF contain
Amount or %BtInitial value, t is the time, and k is velocity constant.
Beneficial effects of the present invention:Sparkling wine sample is placed in the hole in the centre position of 96 orifice plates, orifice plate is put
Put on the diffused light source panel of vasculum bottom, using smart mobile phone as image capture device, from adopting at the top of vasculum
Collect the digital picture that sparkling wine is obtained in hole, carry out Digital Image Processing and analysis, blistering is detected in RGB color
The brown stain of grape wine, determines the browning degree of sparkling wine;Absorbance and acquisition with two kinds of widely used research 420nm
The method of 5-HMF contents is compared, and blue channel attenuation percentage and common browning index height correlation, are that fizz is brown
The new mark become.The present invention only needs to obtain the transmission image of the sparkling wine sample in different brown stain stages, it is possible to while into
The analysis of the sparkling wine brown stain sample of the multiple brands of row, avoids the sample treatment of complexity, and can synchronously carry out more
Sample analysis, speed is fast, and device price is low, without expensive special instrument and other chemical reagent.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is attached drawing needed in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments of invention, for those of ordinary skill in the art, without creative efforts, can be with
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is the flow chart of the present invention.
Fig. 2 is the structural scheme of mechanism of image capture apparatus of the present invention.
Fig. 3 is putting in order for inventive samples.
Fig. 4 is the gray level image of sample of the present invention.
Fig. 5 is the binary image of sample of the present invention.
Fig. 6 is the image after present invention corrosion.
Fig. 7 causes R, G of various brands sample areas, the Change in Mean schematic diagram of channel B for acceleration brown stain.
Fig. 8 is various brands sample B channel attenuation and the chart of percentage comparison picture of heating time.
Fig. 9 is the absorbance at various brands sample 420nm and the graph of a relation picture of blue channel attenuation percentage.
Figure 10 is various brands sample 5-HMF contents and the graph of a relation of blue channel decay percentage.
Figure 11 is the qualitative comparison of several algorithms.
Embodiment
Below in conjunction with the attached drawing in the embodiment of the present invention, the technical solution in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other under the premise of not making the creative labor
Embodiment, belongs to the scope of protection of the invention.
As shown in Figure 1, a kind of sparkling wine browning detection device and method based on machine vision, its step is such as
Under:
Step 1:Experimental provision is built, prepares sample.
Using the slips sparkling wine of four kinds of best seller brands as research object, including Brut, Brut Reserva,
Brut Gran Reserva and Semiseco, four kinds of grape wine have different sugar contents and productive year.During initial samples,
The quality parameter of each type of slips grape wine is as shown in table 1.Surveyed using the standard method of international grape and grape wine tissue
Measure content sugar, alcohol content, pH, the free and total sulfur dioxide of every kind of sample.Wherein, alcohol content meets on slips wine bottle
The value of sign.
The quality parameter of 14 kinds of brand Cava sparkling wines of table
10mL is respectively taken to be respectively charged into the amber vial of 4 20mL the sparkling wine of four brands, and in N2Gas
Flow down degassing.This 4 amber vials equipped with different brands fizz are placed in the environment of dark afterwards, and preserve ring
Border temperature is 8 DEG C, places the time of 10 days in the present context, its brown stain can be ignored.By the different brands in this 4 bottles
Standard sample of the fizz as subsequent experimental, and this 4 amber vials equipped with standard sample are named as
Bottle1、Bottle2、Bottle3、Bottle4。
And then, then by the sparkling wine of four brands respectively take 10mL to be respectively charged into the amber vial of 4 20mL,
And in N2Deaerate under air-flow.This 4 amber vials equipped with different brands fizz are placed on to the environment of complete darkness, so
After place in an oven, 65 ± 1 DEG C of constant temperature of setting, carries out accelerating brown stain experiment.And by this 4 equipped with acceleration brown stain sample
Amber vial is named as Bottle5, Bottle6, Bottle7, Bottle8.
Sample message is gathered at the 0th, 2,4,6,8 and 10 day (6 sampling time points), the time interval of collection is small for 48
When.In each sampling time point, (1) extracts Brut, Brut Reserva, Brut Gran Reserva and Semiseco first
Each 1 part of standard sample, in the hole at the middle part for being placed sequentially in 96 orifice plates from top to bottom, line up 1 row vertically, be then named as
1 row;(2) then extraction Brut, Brut Reserva, Brut Gran Reserva and Semiseco acceleration brown stain sample
Each 4 parts, then each brand put 1 row, be sequentially placed into 96 orifice plates the 2nd~5 row, as shown in Figure 3.Therefore, 120 are analyzed altogether
Sample (6 sampling time point × 4 Cava brand × 5 part (4 parts accelerate brown stain sample and 1 part of standard sample)).
For experimental provision as shown in Fig. 2, including vasculum 2 and light source 4, light source 4 is arranged on the lower section of vasculum 2, vasculum 2
Lower part, which is equipped with, is used to holding 96 orifice plates 3 of sample, and the top of vasculum 2 is equipped with collection hole 1, collection hole 1 be located at orifice plate 3 just on
Side.Vasculum 1 is made of black foam core plate, internal covered with dumb light black velvet paper, to eliminate internal light reflection.Adopt
Header 1 is the cuboid of high 80cm so that difference between diagonal and center vertical optical path is less than 5%, so as to reduce due to
Light passes through difference caused by different paths.Light source 4 is the diffusing reflection light source of intensity controlled.Orifice plate 3 is 96 orifice plates, by import
The pure polystyrene manufacture of optical clear, through gamma rays sterilization treatment.
Step 2:Sample image is obtained using the camera of smart mobile phone, the RGB image of acquisition is subjected to gray processing and two-value
Change is handled.
In the hole at the middle part that slips sparkling wine sample is arranged in orifice plate 3 using micropipette (0.3ml), per hole
Add 0.15ml samples, orifice plate 3 is placed on the panel of light source 4, vasculum 2 is covered on the aperture plate 3, collection hole 1 is located at orifice plate 3
Surface.Smart mobile phone is placed on the collection hole 1 at the top of vasculum, using the built-in camera of smart mobile phone with highest
Resolution ratio obtains image (3888 × 5152=20030976 pixels), and saves as jpeg format.Every time during collection, orifice plate is obtained
In all samples into same image, caused by being rocked to avoid the fluctuation due to light source or smart mobile phone influence.For
Raising efficiency, and can be compared between different sample images, all can be in the 1st row of orifice plate when gathering image every time
The primary sample of placement standard, is derived from each Bottle1, Bottle2, Bottle3, Bottle4 using time point.
Sample is placed in the center of 96 orifice plates, is from left to right arranged in order, the 1st is classified as the primary sample of standard (often
Kind of 1 part of sparkling wine, respectively from Bottle1, Bottle2, Bottle3 and Bottle4, totally 4 parts), the 2nd is classified as Brut samples
4 parts of product (come from Bottle5), and the 3rd is classified as 4 parts of Brut Reserva samples (coming from Bottle6), and the 4th is classified as Brut Gran
4 parts of Reserva samples (come from Bottle7), and the 5th is classified as 4 parts of Semiseco samples (coming from Bottle8), as shown in Figure 3.
The image comprising the sparkling wine sample for accelerating brown stain and the sample of standard sparkling wine is obtained first, rear
Continuous image of the sampling time point collection comprising brown stain sample and standard sample.By obtaining 6 figures in 6 sampling time points
Picture, at the 1st time, obtains the standard sample in orifice plate and accelerated the image IG of the sample of brown stain by 0 day1;At the 2nd time, obtain
The image IG of standard sample in orifice plate and the sample by acceleration brown stain in 2 days2;At the 3rd time, the standard sample in orifice plate is obtained
With the image IG of the sample by acceleration brown stain in 4 days3;In the 4th, obtain the standard sample in orifice plate and accelerated by 6 days brown
The image IG of the sample of change4;In the 5th, obtain the standard sample in orifice plate and accelerated the image of the sample of brown stain by 8 days
IG5;At the 6th time, obtain the standard sample in orifice plate and accelerated the image IG of the sample of brown stain by 10 days6.Each sampling time
Before point obtains image, the relevant position that sample is put into orifice plate will be taken out from Bottle1~Bottle8, as shown in Figure 3.Adopt
When collecting the image of sample, ensure that each column equipped with sample is in vertical state as far as possible in the preview form of camera so that obtain
Image close in Fig. 3 sample place ideal position, in order to follow-up image procossing.
The uniformity of illumination and measurement is studied using blank, blank is placed on diffusing reflection light source, and using intelligent hand
The built-in camera of machine obtains image IB (3888 × 5152=20030976 pixels) with highest resolution ratio, by image IB average marks
For 322 × 243=78246 blocks, (each piece is expressed as BKNum1, num2, the value that the value of num1 is 1~322, num2 is 1~243),
Every piece of size is 16 × 16=256 pixels.Count BKNum1, num2RGB triple channels average BKNum1, num2, ave。
Wherein, RIB(x,y)、GIB(x, y) and BIB(x, y) is represented in image IB in the R passages at pixel (x, y) place respectively
Value, the value of G passages and the value of channel B.The value range that the value range of x is 0~3887, y is 0~5151.
Count BKNum1, num2, aveIn maximum MAXbkAnd minimum value MINbk, by calculating, maximum difference (MAXbk-
MINbk)/MAXbkLess than 1%, show that the influence of exterior lighting is smaller, meet requirement of the present invention to the uniformity of illumination.This hair
The bright view data using aperture plate center position, reduces error as far as possible with this.
The RGB image IG that smart mobile phone is obtainedi(value of i is 1~6) is converted to gray level image Grayi, using average value
Method, that is, extract the average value of RGB triple channels as gray value:
Grayi(x, y)=(RIG,i(x,y)+GIG,i(x,y)+BIG,i(x,y))/3
Wherein, RIG,i(x,y)、GIG,i(x,y)、BIG,i(x, y) is respectively image IGiIn pixel (x, y) place R, G,
The value of channel B;Grayi(x, y) represents gray value of the gray level image at pixel (x, y) place, as shown in Figure 4.
It needs to be determined that threshold value could be to gray level image GrayiCarry out binary conversion treatment, maximum variance between clusters (Ostu) meter
Calculate it is simple, stablize it is effective, be passed through in practical application frequently with definite threshold value method.Ostu algorithms are considered as image segmentation
Preferred algorithm that middle threshold value is chosen, the algorithm from brightness of image and contrast influence, therefore in Digital Image Processing
To being widely applied.Ostu algorithms are the gamma characteristics by image, divide the image into background and prospect two parts.Background is with before
Inter-class variance between scape is bigger, illustrates that the two-part difference for forming image is bigger, prospect mistake is divided into background or portion when part
Point background mistake, which is divided into prospect, all can cause two parts difference to diminish, therefore, make inter-class variance maximum segmentation mean wrong point it is general
Rate is minimum.
If the gray level of gray level image is L, then tonal range is [0, L-1], and gray level image is calculated using Ostu algorithms
Optimal threshold is:
σ=Max [w0(gt)×(u0(gt)-u)^2+w1(gt)×(u1(gt)-u)^2)]
Wherein, w0For prospect ratio, u0For prospect gray average, w1For background ratio, u1For background gray average, u is whole
The average of width gray level image so that the gray value gt of σ maximums is exactly the optimal threshold T of segmentation figure picture.
To gray level image Grayi(x, y) finds optimal threshold T using criterion above, divides the image into as 2 parts,
Obtain binary image BINi(x,y)。
Wherein, Grayi(x, y) is gray level image GrayiIn (x, y) place pixel value, the value of i is 1~6 (collection altogether
6 images, obtain 1 image every time), the image BIN after binaryzationi(x, y) is as shown in Figure 5.
Step 3:By image BINiIn pixel all negate, obtain binary image IRi;With structural element to image
IRiCorroded to obtain image IERi, the constriction zone in the hole equipped with sample is obtained, prevents the edge in the hole equipped with sample to sample
The image procossing of product interferes.
D1Represent a diameter of 4 circular configuration element, its establishment function is:
D1=strel (' disk', 4)
Its implication is that one radius of establishment is 4 (the circular configuration element of i.e. a diameter of 8), disk expression circular shapes.
The function of strel functions is structural texture element (Structuring element), and so-called structural element, can regard one as
Small image is opened, it is commonly used in the morphology operations (such as expansion, burn into opening operation and closed operation) of image.
By image BINiIn pixel all negate, obtain image IRi.That is image BINiIn the value of original pixel be
0, then it is changed into 1;The value of pixel originally is 1, then is changed into 0.
Using the imerode functions in Matlab by binary image IRiCarry out the etching operation in Morphological scale-space, tool
The operating method of body is:
IERi=imerode (IRi,D1)
Wherein, IERiRepresent binary image IRiBy a diameter of 8 circular configuration element D1Obtained image after corrosion,
I=1,2,6.As shown in Figure 6.Utilize circular configuration element D1To binary image IRiCarry out at morphologic corrosion
Reason, eliminates the border for the emptying aperture for not holding sample in image, while also removes the border in the hole for holding sample, obtains in sample
The constriction zone in the hole of the image in heart district domain, i.e. sample, prevents the edge in the hole equipped with sample from being caused to the image procossing of sample
Interference.In image equipped with sample hole region in, by using morphologic caustic solution by the sample areas in image into
Go corrosion (achieveed the purpose that shrink sample areas), only with the pixel inside hole, eliminate the influence on the border in hole.
Step 4:By image IERiConnected component labeling is carried out, each type of sample is identified according to connected component labeling
Position.
Image IERiConnected component labeling is carried out, for handling the image in the hole equipped with different samples.If E with
F is connected, and F is connected with G, then E is connected with G.Visually apparently, the point to communicate with each other forms a region, and disconnected
Point forms different regions.Such a all collection formed that communicates with each other are collectively referred to as a connected region.Binaryzation
The most important method of graphical analysis is exactly connected component labeling, it passes through the mark to white pixel in binary image (target)
Allow each individually connected region to form an identified block, can further obtain the area of these blocks, profile, external
Rectangle, barycenter and the not geometric parameter such as bending moment.
The present invention uses connected component labeling function bwlabel in Matlab:[LJi,LNUMi]=bwlabel (IERi,
8), wherein LNUMiRepresent image IERiThe quantity of middle connected region, due to there is 20 samples in each image, so LNUMi
=20.LJiExpression and IERiFor the identical matrix of size, LJiMatrix contains mark image IERiIn each connected region class
Distinguishing label, the value of these labels is 1,2 ..., LNUMi, 8 in formula represent it is by 8 neighbor searching connected regions.
The algorithm of function bwlabel is a traversing graph picture, and writes down what is continuously rolled into a ball and mark in every a line (or row)
It is of equal value right, then by equivalence to being re-flagged to original image.This algorithm is higher one of current efficiency, is calculated
It is of equal value right for eliminating with Dulmage-Mendelsohn decomposition algorithms that sparse matrix has been used in method.
By image IERiConnected component labeling is carried out, realizes sample areas that each shrinks as a connected region,
For each connected region, using regionprops functions in Matlab:STATSi=regionprops (LJi,'
Centroid'), ' Centroid' attributes be exactly the region barycenter (i.e. the centre of form of connected region).
The STATS being calculated according to regionprops functionsiContain image IERiIn each connected region it is (i.e. every
A sample areas) centre of form coordinate (RXi,nu,LYi,nu), wherein the value of nu is 1~20.The centre of form of these connected regions is according to row
Coordinate is arranged as LY from small to largei,1~LYi,20, the wherein LY of row coordinate minimumi,1~LYi,4This 4 sample areas, are sample
(row share four samples, are followed successively by Brut standard samples, Brut from top to bottom in standard sample region in this image
Reserva standard samples, Brut Gran Reserva standard samples, Semiseco standard samples);Row coordinate is LYi,5~
LYi,8This 4 sample areas, are the Brut sample areas in sample image;Row coordinate is LYi,9~LYi,12This 4 sample areas
Domain, is the Brut Reserva sample areas in sample image;Row coordinate is LYi,13~LYi,16This 4 sample areas, are
Brut Gran Reserva sample areas in sample image;Row coordinate is LYi,17~LYi,20This 4 sample areas, are sample
Semiseco sample areas in this image.
Step 5:With image IG1R passages, G passages and the B in region of contraction sample well of primary sample of standard lead to
Correction factor is calculated on the basis of the average in road, and to image IGiEach passage carries out color correction, obtains correction chart as IGCi, i=
2,…6。
By image IER1In the row coordinate of centre of form coordinate be LY1,1~LY1,4The pixel of connected region put 1, it is other
Pixel is set to 0, and obtains matrix J Z1, then by original image IG1In all pixels point R passages value as matrix RIG,1,
By original image IG1In all pixels point G passages value as matrix GIG,1, by original image IG1In all pixels point
Channel B value as matrix BIG,1.By JZ1Dot product RIG,1Obtain DCR1, by JZ1Dot product GIG,1Obtain DCG1, by JZ1Dot product
BIG,1Obtain DCB1。
Using the regionprops functions in matlab:MJ1=regionprops (JZ1, ' Area'), calculate image IG1
Standard sample shrinkage hole region R passages average.The function include ' Area' attributes are exactly image regional
Middle pixel total number.
R1, ave=sum (sum (DCR1))/sum(MJ1),
G1, ave=sum (sum (DCG1))/sum(MJ1),
B1, ave=sum (sum (DCB1))/sum(MJ1),
Wherein, sum (sum (DCR1)) it is matrix D CR1The summation of middle all elements value, sum (sum (DCG1)) it is matrix
DCG1The summation of middle all elements value, sum (sum (DCB1)) it is matrix D CB1The summation of middle all elements value, sum (MJ1) it is figure
As IER1In the row coordinate of centre of form coordinate be LY1,1~LY1,4Connected region sum of all pixels.Similarly, G is calculated1, ave
And B1, ave。
Similarly, image IG is calculatediStandard sample contraction sample well region R passages, G passages and channel B
Average RI, ave、GI, aveAnd BI, ave, wherein i=2,3 ..., 6.
Then color correction is carried out.With image IG1Standard primary sample contraction sample well region R passages, G
The average R of passage and channel B1, ave、G1, aveAnd B1, aveOn the basis of, follow-up different time sampled point acquired image IGi(i=
2,3 ..., 6) in standard primary sample where the R passages in region in hole, the average of G passages and channel B be Ri,ave、
GI, aveAnd BI, aveTo refer to (wherein i=2,3 ..., 6), three correction factor C are calculatedi,R、Ci,GAnd Ci,B(wherein i=2,
3,…,6)。
Wherein, Ci,R、Ci,G、Ci,BFor image IGiR passages, the correction factor of G passages and channel B.
By image IGiAll pixels point R passages RIG, iIt is multiplied by Ci,R, G passages GIG, i is multiplied by Ci,G, channel B BIG, iIt is multiplied by
Ci,B, tri- passages of new RGB are obtained, the image of composition is named as correction chart as IGCi, wherein i=2,3 ..., 6.
Step 6:Calculate the image IGC after correctioniThe R, G in region of shrinkage hole of each brand sample, channel B
Average, chooses parameter of the channel B as brown stain.
Using method before, correction chart is obtained as IGCiThe various brands of (i=1,2 ..., 6) accelerate the receipts of brown stain sample
The average R of R, G, B triple channel in contracting regionIGC,i,ave、GIGC,i,ave、BIGC,i,ave, and draw, as shown in Figure 7.Fig. 7 shows companion
With heating time, R, G of sample area and the differentiation of channel B value.Channel B is mainly reflected in as can be seen that accelerating to change, and R
It is held essentially constant with G values.With the increase of heating time, this shows as deeper yellow, i.e., changes to more shallow brown.
The change of the channel B of Brut and GRes samples has obvious linear dependence, and in Res and SS, 48 it is small when after brown stain
Start, hereafter blue channel value and linear dependence again between the time.Original sparkling wine sample shows it
The Light Difference of color, as Brut and Brut Reserva (Res) sample is more like each other.And Semiseco (SS) and Brut
Gran Reserva (GRes) show larger difference, and the maximum difference between sample is blue channel.
Step 7:Calculate channel B Color invariants and blueness decay percentage;Blueness decay percentage has line with the time
Sexual intercourse, is consistent with the browning of sparkling wine.
Except having used RIGC,i,ave、GIGC,i,ave、BIGC,i,aveThe change of value is analyzed, and additionally uses R, G, B color not
Variable studies brown stain.By sample image IGCiChannel B be converted to respective color invariant (bi=BIGC,i,ave/
(RIGC,i,ave+GIGC,i,ave+BIGC,i,ave)), to quantify and compare brown stain.Channel B Color invariants are similar to absorbance, can be with
Calculate blueness decay percentage (%Bt).B at this timeiCorresponding to bt。
Wherein, bt0Channel B invariant when being t=0, btIt is the Color invariants of the channel B of time t.Fig. 8 shows institute
The %B of the sample of researchtChange with time, it can be seen that brown stain is reflected in %BtIncrease linearly over time.
For Brut,R2For 0.99.
For Brut Reserva,R2For 0.93.
For Semiseco,R2For 0.93.
For Brut Gran Reserva,R2For 0.96.
Wherein, xdayFor chronomere (my god).
Shimadzu ultraviolet specrophotometer (UV-3600, Duisburg, Germany) in use 10mm path lengths
The quartz cuvette and distilled water of degree as reference, measure absorbance of the Cava sparkling wines of 4 kinds of brands at 420nm.
In 420nm (A420) absorbance be multiplied by 1000 times, be expressed as milli absorbance unit (milli-absorbance units, letter
Claim mAU).5-HMF is measured in each sample according to the method for international grape and grape wine tissue, using with level Four L-7100
Hitachi's liquid chromatograph (Hitachi) of pump, L-7455 diode array detector and LaChrom chromatographic columns carries out 4 kinds of brands
Cava sparkling wines liquid-phase chromatographic analysis.Image capture device in the present invention is 7 smart mobile phone of Huawei's honor, is had
5.2 inches of touch display screens, resolution ratio are the pixel of 1920 pixels × 1080,20,000,000 pixel camera head (5152 × 3888 pixels
Image), support PDAF phase-detection rapid focus, F2.0 apertures.
In order to assess detection result of the present invention, by %BtWith other conventional brown stain monitoring parameters (such as A420With 5-HMF contents)
It is compared, Fig. 9 is shown in 420nm (A420) place measurement absorbance and %BtBetween correlation.
Linear regression analysis is carried out to determine the relation between these brown stain indexs:
For Brut,R2For 0.98.
For Brut Reserva,R2For 0.98.
For Semiseco,R2For 0.99.
For Brut Gran Reserva,R2For 0.96.
Wherein, x is sample in 420nm (A420) place absorbance, y %Bt。
The result shows that %BtThe trend similar to the browning index presentation represented by A420.It should be noted that two kinds of distinct methods
Between have height correlation.Correspond to the color in the border between blueness and purple in view of 420nm, absorption band will be given
Sample band Yellow-to-orange/brown, correspondingly, these changes will reflect in the channel B of image.Equally it is significantly, it is different
There are uniformity between sample, their slope is closely similar, particularly Brut, Res and SS sample.
On the other hand, Figure 10 shows %BtCorrelation between 5-HMF contents.Linear regression analysis is carried out to determine
Relation between these brown stain indexs:
For Brut,R2For 0.93.
For Brut Reserva,R2For 0.97.
For Semiseco,R2For 0.99.
For Brut Gran Reserva,R2For 0.91.
Wherein, x be sample 5-HMF contents, y %Bt。
As can be seen that there is very high correlation between them, and particularly Brut, Res and SS samples.Should be considerable
It is that 5-HMF is not the compound of unique instruction grape wine brown stain, therefore the difference of slope value is attributable to depositing for other compounds
Also the color of grape wine is being influenced.
On %Bt, 5-HMF contents and A420The characteristics of the results show of acquisition browning, it can be by following
Equation describes:
Y=Y0+kt (8)
Wherein, Y is absorbance (mAU) or 5-HMF contents or the %B at 420nmt, t is time (in units of day), and k is
Velocity constant (is expressed as A420MAU/day, mg/L/ days of 5-HMF, for BtThe daily percentage specific damping of blue channel).
Zero order kinetics (zero-order elimination kineics) refers to medicine in vivo with constant speed
Eliminate, no matter that is, plasma drug level height, the medication amount eliminated in the unit interval are constant.Produce the master of zero kinetics
Will be the reason is that the saturation history that drug metabolic enzyme, drug transporters and medicine are combined with plasma protein, zero kinetics
There is the characteristics of active transport.Zero order kinetics is set up from natural force system, such as electromagnetic force, biomethanics, gravitation.Zero
The dynamic (dynamical) rich connotation of level, natural day is into enlightenment is far-reaching, is widely used.The present invention uses the correlation theory of zero order kinetics,
Carry out the research of sparkling wine brown stain.The formula of zero order kinetics is identical with above formula, and (at this time, k is zero-order constant, Y0For
Initial blood concentration, blood concentration when Y is t).The velocity constant calculated for the zero order kinetics of each sample and monitoring side
Method is as shown in table 2.As can be seen that use %BtZero order kinetics (table 2 and Fig. 8) is also observed as collative variables.In addition, to the greatest extent
Pipe assessment brown stain speed constant value it is different according to different application methods, but the relation between result follow really it is identical
Trend.
2 zero order kinetics parameter (A of table420, 5-HMF, %Bt)
As the time increases, sparkling wine ferments (from Brut Gran Reserva to Brut Reserva) continuous,
Brown stain velocity constant increases (from Brut to Semiseco) with the increase of sugared content.Show that the constant speed of sugared content increases
Add, as was expected, and 5-HMF is formed and initial sugared content height correlation, because 5-HMF synthesis speeds are sugared dependences
(Camara J S,Alves M A,Marques J C.Changes in volatile composition of Madeira
wines during their oxidative ageing[J].Analytica Chimica Acta,2006,563(1):
188-197.).Therefore, grape wine (the Brut Gran of the sparkling wine at least influenced by browning seemingly first water
Reserva, Brut Reserva).Figure 11 summarize to graphically acquisition results, it can be seen that the present invention makes brown stain
Journey is characterized, and completely the same with known method.
The present invention proposes a kind of colorimetric method for the browning for being used to monitor sparkling wine, and this method speed is fast, consumption
Take low.Using the camera of smart mobile phone as harvester, and for analyzing the transmission image of slips grape wine, avoid
Complicated sample treatment, and single step Multi-example analysis can be carried out.The present invention proposes a kind of new control parameter, with the time
Passage, blue channel percentage specific damping, this causes brown stain dynamics to be able to study and characterize.That is R, G passage is in browning
In be kept approximately constant, browning has time dependence, with the time increase and mainly influence channel B.The present invention proposes will
Quality status stamps of the blue channel attenuation percentage %Bt as sparkling wine, the value and the absorbance and 5- hydroxyl first at 420nm
The content height correlation of base -2- furfurals, the content of absorbance and HMF 5 hydroxymethyl 2 furaldehyde at 420nm is grape wine brown stain
Common quality status stamp, the result of acquisition and existing research method are completely the same.Therefore, new parameter and furfural compounds are deposited
In height correlation, be reflected in the correlation of A420 and in lesser degree it is related with 5-HMF concentration because this is only brown stain
One of coloring compound involved in process.The present invention the result shows that %Bt is a good brown stain descriptor, it is only necessary to obtain
Take the transmission image of the sparkling wine sample in different brown stain stages, it is possible to be carried out at the same time the sparkling wine sample of multiple brands
This analysis, device is cheap, without expensive special instrument and other chemical reagent.
Method proposed by the invention can be as the alternative solution of conventional method, and may be used as sparkling wine
Key Quality Con trolling index.Method proposed by the invention has been proven that its serviceability, therefore based on the further of this method
Research will be significantly.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
With within principle, any modification, equivalent replacement, improvement and so on, should all be included in the protection scope of the present invention god.
Claims (10)
1. a kind of sparkling wine browning detection method based on machine vision, it is characterised in that its step is as follows:
Step 1:Experimental provision is built, prepares sample;
Step 2:Sample image is obtained using the camera of smart mobile phone, by the RGB image IG of acquisitioniCarry out gray processing and binaryzation
Processing, obtains binary image BINi, i=1,2 ..., 6;
Step 3:With structural element to binary image BINiCorroded, obtain image IEi, obtain the hole equipped with sample
Constriction zone;
Step 4:By image IEiIn pixel all negate, obtain image IERi;By image IERiCarry out connected component labeling,
Each type of sample is identified according to connected component labeling;
Step 5:With image IG1The R passages in region of shrinkage hole of primary sample of standard, the average of G passages and channel B be
Benchmark correction factor, and to image IGiEach passage carries out color correction, obtains correction chart as IGCi, i=2 ... 6;
Step 6:Calculate the image IGC after correctioniThe R, G in region of shrinkage hole of each brand sample, the average of channel B,
Choose parameter of the channel B as brown stain;
Step 7:Channel B Color invariants and blueness decay percentage are calculated, blueness decay percentage has linear close with the time
System, blueness decay percentage are consistent with the browning of sparkling wine, the quality status stamp as brown stain detection.
2. the sparkling wine browning detection method according to claim 1 based on machine vision, it is characterised in that
The sample is Brut, the blistering of the slips of Brut Reserva, Brut Gran tetra- kinds of best seller brands of Reserva and Semiseco
Grape wine, four kinds of grape wine have different sugar contents and productive year;10mL is taken to be respectively charged into the amber vial of 4 20mL
In, and in N2Deaerate under air-flow;This 4 amber vials equipped with different brands fizz are placed in the environment of dark,
And Conservation environment temperature is 8 DEG C;Standard sample using the fizz of the different brands in this 4 bottles as the experiment of follow-up 10 days
Product, and this 4 amber vials equipped with standard sample are named as Bottle1, Bottle2, Bottle3, Bottle4;Again
10mL is respectively taken to be respectively charged into the amber vial of 4 20mL the sparkling wine of four brands, and in N2Deaerate under air-flow;
This 4 amber vials equipped with different brands fizz are placed on to the environment of complete darkness, are then placed within baking oven, if
Constant temperature 65 ± 1 DEG C fixed, carries out accelerating brown stain experiment;This 4 amber vials equipped with acceleration brown stain sample are named as
Bottle5、Bottle6、Bottle7、Bottle8;In the sampling time point collection sample letter by 0,2,4,6,8 and 10 day
Breath, when the time interval of collection is 48 small;In each sampling time point, (1) extracts Brut, Brut Reserva, Brut first
Each 1 part of the standard sample of Gran Reserva and Semiseco, in the hole at the middle part for being placed sequentially in 96 orifice plates from top to bottom, is erected
It is in line to be arranged into 1, then it is named as the 1st row;(2) then extraction Brut, Brut Reserva, Brut Gran Reserva and
Semiseco acceleration brown stain each 4 parts of sample, then each brand put 1 row, be sequentially placed into 96 orifice plates the 2nd~5 row.
3. the sparkling wine browning detection method according to claim 1 based on machine vision, it is characterised in that
The experimental provision includes vasculum (2) and light source (4), and light source (4) is arranged on the lower section of vasculum (2), vasculum (2) lower part
Equipped with the orifice plate (3) for holding sample, collection hole (1) is equipped with the top of vasculum (2), collection hole (1) is being located at orifice plate (3) just
Top.
4. the sparkling wine browning detection method according to claim 3 based on machine vision, it is characterised in that
The vasculum (1) is made of black foam core plate, and vasculum (1) is internal covered with dumb light black velvet paper;Vasculum (1)
For the rectangular parallelepiped structure of height 80cm, the difference between diagonal and center vertical optical path is less than 5%;The light source (4) is strong
Spend controllable diffusing reflection light source;Orifice plate (3) is 96 orifice plates, and orifice plate (3) is manufactured by the pure polystyrene of import optical clear, through gamma
Ray sterilizing processing.
5. the sparkling wine browning detection method based on machine vision according to Claims 2 or 3, its feature exist
In the method for obtaining sample image is:Slips sparkling wine sample is arranged in orifice plate (3) using 0.3ml micropipettes
Middle part, by orifice plate (3) place light source (4) panel on, vasculum (2) is covered on orifice plate (3), collection hole (1) be located at hole
The surface of plate (3), smart mobile phone is placed on the collection hole (1) at the top of vasculum, uses the built-in camera of smart mobile phone
Image is obtained with highest resolution ratio, and saves as JPEG;The primary sample arrangement of standard in the orifice plate, obtains all samples and arrives
In same image.
6. the sparkling wine browning detection method according to claim 1 based on machine vision, it is characterised in that
The method that the RGB image of acquisition is carried out to gray processing and binary conversion treatment is:
The RGB image IG that smart mobile phone is obtainediIn each pixel tri- components of R, G, B gray scale of the average value as image
Value, i.e. Grayi(x, y)=(RIG,i(x,y)+G IG,i(x,y)+B IG,i(x, y))/3, obtain gray level image Grayi;
To gray level image GrayiOptimal threshold T is calculated using Otsu algorithms, optimal threshold T is so that the gray value of σ maximums
gt:If the gray level of gray level image is L, then tonal range is [0, L-1], and the optimal threshold of gray level image is calculated using Ostu methods
It is worth and is:σ=Max [w0(gt)×(u0(gt)-u)^2+w1(gt)×(u1(gt)-u) ^2)], wherein, w0For prospect ratio, u0For
Prospect gray average, w1For background ratio, u1For background gray average, u is the average of view picture gray level image.
T is the optimal threshold of segmentation figure picture, and gray level image is divided into 2 parts according to optimal threshold T, obtains binary image
BINi(x,y):
<mrow>
<msub>
<mi>BIN</mi>
<mi>i</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mn>0</mn>
</mtd>
<mtd>
<mrow>
<msub>
<mi>Gray</mi>
<mi>i</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo><</mo>
<mi>T</mi>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mn>1</mn>
</mtd>
<mtd>
<mrow>
<msub>
<mi>Gray</mi>
<mi>i</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>&GreaterEqual;</mo>
<mi>T</mi>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
</mrow>
Wherein, Grayi(x, y) is the value of the pixel at (x, y) place in gray level image;
By binary image BINiNegate to obtain binary image IRi, using the imerode function pairs IR in MatlabiCarry out rotten
Erosion operation:IERi=imerode (IRi,D1), wherein, IERiRepresent binary image IRiBy a diameter of 8 circular configuration element
D1Obtained image after corrosion, i=1,2,3 ..., 6;D1=strel (' disk', 4), disk represents circular shape.
7. the sparkling wine browning detection method according to claim 6 based on machine vision, it is characterised in that
By image IERiCarry out connected component labeling method be:Using connected component labeling function bwlabel in Matlab:[LJi,
LNUMi]=bwlabel (IERi, 8), wherein, LNUMiRepresent image IERiThe quantity of middle connected region, LJiExpression and image
IERiThe identical matrix of size, matrix L JiContain mark image IERiIn each connected region class label, these labels
Value be 1,2 ..., LNUMi, 8 represent it is by 8 neighbor searching connected regions;
The method of the position of each type of sample of the identification is:The sample areas that connected component labeling shrinks each is made
For a connected region, for each connected region, using the regionprops functions in Matlab:STATSi=
regionprops(LJi, ' Centroid'), ' Centroid' attributes are exactly the barycenter of connected region, STATSiContain image
IERiIn each connected region centre of form coordinate (RXi,nu,LYi,nu), wherein, the value of subscript n u is 1~20.
8. the sparkling wine browning detection method according to claim 1 based on machine vision, it is characterised in that
The acquisition methods of correction chart picture are:With image IG1Standard primary sample shrinkage hole region R passages, G passages and B
The average R of passage1, ave、G1, aveAnd B1, aveOn the basis of, standard in follow-up different time sampled point acquired image it is original
The average of the R passages in the region of the shrinkage hole where sample, G passages and channel B is RI, ave、GI, aveAnd BI, aveFor reference, calculate
Three correction factor Ci,R、Ci,GAnd Ci,B:
<mrow>
<msub>
<mi>C</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>R</mi>
</mrow>
</msub>
<mo>=</mo>
<mfrac>
<msub>
<mi>R</mi>
<mrow>
<mn>1</mn>
<mo>,</mo>
<mi>a</mi>
<mi>v</mi>
<mi>e</mi>
</mrow>
</msub>
<msub>
<mi>R</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>a</mi>
<mi>v</mi>
<mi>e</mi>
</mrow>
</msub>
</mfrac>
<mo>,</mo>
<msub>
<mi>C</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>C</mi>
</mrow>
</msub>
<mo>=</mo>
<mfrac>
<msub>
<mi>G</mi>
<mrow>
<mn>1</mn>
<mo>,</mo>
<mi>a</mi>
<mi>v</mi>
<mi>e</mi>
</mrow>
</msub>
<msub>
<mi>G</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>a</mi>
<mi>v</mi>
<mi>e</mi>
</mrow>
</msub>
</mfrac>
<mo>,</mo>
<msub>
<mi>L</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>B</mi>
</mrow>
</msub>
<mo>=</mo>
<mfrac>
<msub>
<mi>B</mi>
<mrow>
<mn>1</mn>
<mo>,</mo>
<mi>a</mi>
<mi>v</mi>
<mi>e</mi>
</mrow>
</msub>
<msub>
<mi>B</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>a</mi>
<mi>v</mi>
<mi>e</mi>
</mrow>
</msub>
</mfrac>
</mrow>
Wherein, Ci,R、Ci,G、Ci,BFor image IGiR passages, the correction factor of G passages and channel B, i=2,3 ..., 6;
By image IGiAll pixels point R passages RIG, iIt is multiplied by Ci,R, G passages GIG, iIt is multiplied by Ci,G, channel B BIG, iIt is multiplied by Ci,B, i
=2,3 ..., 6, obtain tri- passages of new RGB, the image of composition is correction chart as IGCi。
9. the sparkling wine browning detection method based on machine vision according to claim 7 or 8, its feature exist
In by image IER1In the row coordinate of centre of form coordinate be LY1,1~LY1,4The pixel of connected region put 1, other pixels
Set to 0, obtain matrix J Z1, by original image IG1In all pixels point R passages value as matrix RIG,1, by original image
IG1In all pixels point G passages value as matrix GIG,1, by original image IG1In all pixels points channel B
Value is used as matrix BIG,1;By matrix J Z1Dot product matrix RIG,1Obtain matrix D CR1, by matrix J Z1Dot product matrix GIG,1Obtain matrix
DCG1, by matrix J Z1Dot product matrix BIG,1Obtain matrix D CB1;
R1, ave=sum (sum (DCR1))/sum(MJ1),
G1, ave=sum (sum (DCG1))/sum(MJ1),
B1, ave=sum (sum (DCB1))/sum(MJ1),
Wherein, sum (sum (DCR1)) it is matrix D CR1The summation of middle all elements value, sum (sum (DCG1)) it is matrix D CG1In
The summation of all elements value, sum (sum (DCB1)) it is matrix D CB1The summation of middle all elements value, sum (MJ1) it is image IER1
In the row coordinate of centre of form coordinate be LY1,1~LY1,4Connected region sum of all pixels, MJ1=regionprops (JZ1,'
Area'), ' Area' attributes are exactly number of pixels in image regional.
10. the sparkling wine browning detection method according to claim 1 based on machine vision, its feature exist
In by sample image IGCiChannel B be converted to respective color invariant:bi=BIGC,i,ave/(RIGC,i,ave+GIGC,i,ave+
BIGC,i,ave), calculate blueness decay percentage %Bt:
Wherein, bt0Channel B invariant when being t=0, btIt is the Color invariants of the channel B of time t;Blueness decay percentage
Compare %BtRepresent that browning is:Y=Y0+ kt,
Wherein, Y is the absorbance or 5-HMF contents or %B at 420nmt, Y0It is the absorbance or 5-HMF contents at 420nm
Or %BtInitial value, t is the time, and k is velocity constant.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711266254.8A CN108022239B (en) | 2017-12-05 | 2017-12-05 | Bubbling wine browning process detection device and method based on machine vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711266254.8A CN108022239B (en) | 2017-12-05 | 2017-12-05 | Bubbling wine browning process detection device and method based on machine vision |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108022239A true CN108022239A (en) | 2018-05-11 |
CN108022239B CN108022239B (en) | 2020-03-10 |
Family
ID=62078258
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711266254.8A Active CN108022239B (en) | 2017-12-05 | 2017-12-05 | Bubbling wine browning process detection device and method based on machine vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108022239B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109211815A (en) * | 2018-10-30 | 2019-01-15 | 福建省农业科学院食用菌研究所(福建省蘑菇菌种研究推广站) | A kind of method for quantitatively evaluating of white mushroom cap damage brown stain |
CN112734688A (en) * | 2020-12-11 | 2021-04-30 | 江苏大学 | Potato browning detection method based on machine vision |
CN116183535A (en) * | 2023-04-23 | 2023-05-30 | 中国科学技术大学 | Method, system, equipment and medium for analyzing spectrum of macromolecular solution aging process |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101282489A (en) * | 2008-04-24 | 2008-10-08 | 北京中星微电子有限公司 | Light source detection apparatus and method as well as image processing method |
CN101714257A (en) * | 2009-12-23 | 2010-05-26 | 公安部第三研究所 | Method for main color feature extraction and structuring description of images |
US20130245148A1 (en) * | 2010-12-07 | 2013-09-19 | Allergan, Inc. | Process for texturing materials |
CN105303189A (en) * | 2014-07-29 | 2016-02-03 | 阿里巴巴集团控股有限公司 | Method and device for detecting specific identification image in predetermined area |
CN107424198A (en) * | 2017-07-27 | 2017-12-01 | 广东欧珀移动通信有限公司 | Image processing method, device, mobile terminal and computer-readable recording medium |
-
2017
- 2017-12-05 CN CN201711266254.8A patent/CN108022239B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101282489A (en) * | 2008-04-24 | 2008-10-08 | 北京中星微电子有限公司 | Light source detection apparatus and method as well as image processing method |
CN101714257A (en) * | 2009-12-23 | 2010-05-26 | 公安部第三研究所 | Method for main color feature extraction and structuring description of images |
US20130245148A1 (en) * | 2010-12-07 | 2013-09-19 | Allergan, Inc. | Process for texturing materials |
CN105303189A (en) * | 2014-07-29 | 2016-02-03 | 阿里巴巴集团控股有限公司 | Method and device for detecting specific identification image in predetermined area |
CN107424198A (en) * | 2017-07-27 | 2017-12-01 | 广东欧珀移动通信有限公司 | Image processing method, device, mobile terminal and computer-readable recording medium |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109211815A (en) * | 2018-10-30 | 2019-01-15 | 福建省农业科学院食用菌研究所(福建省蘑菇菌种研究推广站) | A kind of method for quantitatively evaluating of white mushroom cap damage brown stain |
CN112734688A (en) * | 2020-12-11 | 2021-04-30 | 江苏大学 | Potato browning detection method based on machine vision |
CN116183535A (en) * | 2023-04-23 | 2023-05-30 | 中国科学技术大学 | Method, system, equipment and medium for analyzing spectrum of macromolecular solution aging process |
CN116183535B (en) * | 2023-04-23 | 2023-08-29 | 中国科学技术大学 | Method, system, equipment and medium for analyzing spectrum of macromolecular solution aging process |
Also Published As
Publication number | Publication date |
---|---|
CN108022239B (en) | 2020-03-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Biasoto et al. | Acceptability and preference drivers of red wines produced from Vitis labrusca and hybrid grapes | |
Medina et al. | Increased flavour diversity of Chardonnay wines by spontaneous fermentation and co-fermentation with Hanseniaspora vineae | |
Junior et al. | Potential use of Starmerella bacillaris as fermentation starter for the production of low-alcohol beverages obtained from unripe grapes | |
Izquierdo Cañas et al. | Influence of sequential inoculation of Wickerhamomyces anomalus and Saccharomyces cerevisiae in the quality of red wines | |
CN108022239A (en) | A kind of sparkling wine browning detection device and method based on machine vision | |
Song et al. | Pinot Noir wine composition from different vine vigour zones classified by remote imaging technology | |
Mina et al. | Contribution of yeast in wine aroma and flavour | |
Kelkar et al. | Modeling the effects of initial nitrogen content and temperature on fermentation kinetics of hard cider | |
Domizio et al. | Spontaneous and inoculated yeast populations dynamics and their effect on organoleptic characters of Vinsanto wine under different process conditions | |
Moreno-García et al. | Biological aging status characterization of Sherry type wines using statistical and oenological criteria | |
Holt et al. | Influence of yeast strain on Shiraz wine quality indicators | |
Scansani et al. | The film-forming Pichia spp. in a winemaker's toolbox: A simple isolation procedure and their performance in a mixed-culture fermentation of Vitis vinifera L. cv. Gewürztraminer must | |
Versari et al. | Effect of co‐inoculation with yeast and bacteria on chemical and sensory characteristics of commercial Cabernet Franc red wine from Switzerland | |
Minnaar et al. | Anthocyanins and polyphenols in Cabernet Franc wines produced with Saccharomyces cerevisiae and Torulaspora delbrueckii yeast strains: Spectrophotometric analysis and effect on selected sensory attributes | |
Schwinn et al. | Impact of fermentation temperature on required heat dissipation, growth and viability of yeast, on sensory characteristics and on the formation of volatiles in Riesling | |
Rayne et al. | Chromatic characteristics and optically derived compositional descriptors of micro-oxygenated wines from Vitis vinifera cv. Merlot and Cabernet Sauvignon | |
Sáenz-Navajas et al. | Understanding microoxygenation: Effect of viable yeasts and sulfur dioxide levels on the sensory properties of a Merlot red wine | |
Varela et al. | Volatile aroma composition and sensory profile of Shiraz and Cabernet Sauvignon wines produced with novel Metschnikowia pulcherrima yeast starter cultures | |
Charnock et al. | The Maillard reaction in traditional method sparkling wine | |
Roullier-Gall et al. | Yeast interaction on Chardonnay wine composition: Impact of strain and inoculation time | |
Francesca et al. | Technological screening and application of Saccharomyces cerevisiae strains isolated from fermented honey by-products for the sensory improvement of Spiritu re fascitrari, a typical Sicilian distilled beverage | |
Xia et al. | effects of mixed fermentation on the aroma compounds of ‘Italian Riesling’dry white wine in eastern foothill of Helan mountain | |
Xia et al. | Effect of cold stabilization duration on organic acids and aroma compounds during Vitis vinifera L. cv. Riesling wine bottle storage | |
Grebneva et al. | Understanding yeast impact on 1, 1, 6-Trimethyl-1, 2-dihydronaphthalene formation in riesling wine through a formation-Pathway-Informed hydrolytic assay | |
Makarov et al. | Yeast race effect on the quality of base and young sparkling wines. |
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 |