CN116563258A - Method for rapidly estimating color quality of hydrangea - Google Patents
Method for rapidly estimating color quality of hydrangea Download PDFInfo
- Publication number
- CN116563258A CN116563258A CN202310566659.2A CN202310566659A CN116563258A CN 116563258 A CN116563258 A CN 116563258A CN 202310566659 A CN202310566659 A CN 202310566659A CN 116563258 A CN116563258 A CN 116563258A
- Authority
- CN
- China
- Prior art keywords
- color
- color quality
- quality
- hydrangea
- calculating
- 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
- 235000014486 Hydrangea macrophylla Nutrition 0.000 title claims abstract description 54
- 241001092080 Hydrangea Species 0.000 title claims abstract description 52
- 238000000034 method Methods 0.000 title claims abstract description 50
- 241001164374 Calyx Species 0.000 claims abstract description 19
- 241000272503 Sparassis radicata Species 0.000 claims abstract description 10
- 238000011156 evaluation Methods 0.000 claims abstract description 8
- 229930002877 anthocyanin Natural products 0.000 claims description 18
- 235000010208 anthocyanin Nutrition 0.000 claims description 18
- 239000004410 anthocyanin Substances 0.000 claims description 18
- 150000004636 anthocyanins Chemical class 0.000 claims description 18
- 238000007637 random forest analysis Methods 0.000 claims description 15
- 230000017260 vegetative to reproductive phase transition of meristem Effects 0.000 claims description 15
- ISWSIDIOOBJBQZ-UHFFFAOYSA-N Phenol Chemical compound OC1=CC=CC=C1 ISWSIDIOOBJBQZ-UHFFFAOYSA-N 0.000 claims description 13
- GAMYVSCDDLXAQW-AOIWZFSPSA-N Thermopsosid Natural products O(C)c1c(O)ccc(C=2Oc3c(c(O)cc(O[C@H]4[C@H](O)[C@@H](O)[C@H](O)[C@H](CO)O4)c3)C(=O)C=2)c1 GAMYVSCDDLXAQW-AOIWZFSPSA-N 0.000 claims description 13
- 229930003944 flavone Natural products 0.000 claims description 13
- 150000002212 flavone derivatives Chemical class 0.000 claims description 13
- 235000011949 flavones Nutrition 0.000 claims description 13
- VHBFFQKBGNRLFZ-UHFFFAOYSA-N vitamin p Natural products O1C2=CC=CC=C2C(=O)C=C1C1=CC=CC=C1 VHBFFQKBGNRLFZ-UHFFFAOYSA-N 0.000 claims description 13
- 238000012360 testing method Methods 0.000 claims description 11
- 230000008859 change Effects 0.000 claims description 10
- 238000012549 training Methods 0.000 claims description 8
- 238000013528 artificial neural network Methods 0.000 claims description 7
- 238000005516 engineering process Methods 0.000 claims description 4
- 238000004737 colorimetric analysis Methods 0.000 claims description 3
- 238000000589 high-performance liquid chromatography-mass spectrometry Methods 0.000 claims description 3
- 239000006185 dispersion Substances 0.000 claims description 2
- 238000005286 illumination Methods 0.000 description 10
- 238000011282 treatment Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 6
- 238000005562 fading Methods 0.000 description 6
- 238000012706 support-vector machine Methods 0.000 description 6
- 238000012423 maintenance Methods 0.000 description 5
- 241000123241 Sparassis Species 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 4
- 230000003247 decreasing effect Effects 0.000 description 4
- 241000196324 Embryophyta Species 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 239000000047 product Substances 0.000 description 3
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 2
- 244000267823 Hydrangea macrophylla Species 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 2
- 230000001934 delay Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000003306 harvesting Methods 0.000 description 2
- 238000003064 k means clustering Methods 0.000 description 2
- 238000004895 liquid chromatography mass spectrometry Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000004060 metabolic process Effects 0.000 description 2
- BDAGIHXWWSANSR-UHFFFAOYSA-N methanoic acid Natural products OC=O BDAGIHXWWSANSR-UHFFFAOYSA-N 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000013441 quality evaluation Methods 0.000 description 2
- 230000011514 reflex Effects 0.000 description 2
- OSWFIVFLDKOXQC-UHFFFAOYSA-N 4-(3-methoxyphenyl)aniline Chemical compound COC1=CC=CC(C=2C=CC(N)=CC=2)=C1 OSWFIVFLDKOXQC-UHFFFAOYSA-N 0.000 description 1
- 241001598984 Bromius obscurus Species 0.000 description 1
- 101100136092 Drosophila melanogaster peng gene Proteins 0.000 description 1
- 241000220151 Saxifragaceae Species 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 229930014669 anthocyanidin Natural products 0.000 description 1
- 150000001453 anthocyanidins Chemical class 0.000 description 1
- 235000008758 anthocyanidins Nutrition 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000004720 fertilization Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 229930003935 flavonoid Natural products 0.000 description 1
- 150000002215 flavonoids Chemical class 0.000 description 1
- 235000017173 flavonoids Nutrition 0.000 description 1
- 235000019253 formic acid Nutrition 0.000 description 1
- 238000000227 grinding Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000002386 leaching Methods 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 239000002932 luster Substances 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000012528 membrane Substances 0.000 description 1
- 239000002207 metabolite Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 229910052757 nitrogen Inorganic materials 0.000 description 1
- 235000016709 nutrition Nutrition 0.000 description 1
- 230000035764 nutrition Effects 0.000 description 1
- 150000002989 phenols Chemical class 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 238000002945 steepest descent method Methods 0.000 description 1
- 239000006228 supernatant Substances 0.000 description 1
- 238000002834 transmittance Methods 0.000 description 1
- 238000011179 visual inspection Methods 0.000 description 1
- 238000005303 weighing 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- 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
- G06T2207/30181—Earth observation
- G06T2207/30188—Vegetation; Agriculture
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The invention discloses a method for rapidly estimating the color quality of an embroidered ball, which comprises the following steps: (1) sample collection; (2) Obtaining an sparassis crispa image, and extracting multi-color space component values: l, a, b, H, S, V, R, G, B; (3) Determining the color quality index in the calyx, and calculating the color quality comprehensive score; (4) constructing a flower color quality estimation model: using the color component value brightness L, a, b, H, S, V, R, G, B as an independent variable and the corresponding color quality comprehensive score as an independent variable to construct a color quality estimation model; (5) classifying the color quality of the embroidered ball: classifying the quality of the hydrangea according to the color quality comprehensive score; (6) evaluation of the color of the hydrangea ball: and (3) collecting the embroidered ball inflorescence image to be evaluated, obtaining color component values, calculating the comprehensive score of the color quality, and grading the color quality. The method can be used for rapidly, nondestructively and quantitatively estimating the color quality of the hydrangea.
Description
Technical Field
The invention relates to a method for evaluating the color quality of an embroidered ball, in particular to a method for rapidly estimating the color quality of an embroidered ball based on a multi-color space model.
Background
The Hydrangea (Hydrangea macrophylla) is also called Hydrangea, saxifragaceae Hydrangea, and is cultivated in the Yangtze river basin of the original country, and is loved in a half-shade environment, and the natural flowering period is concentrated in 5-7 months. The hydrangea has various varieties, full flower shape and changeable and controllable flower color, is popular worldwide, can be applied to landscaping in landscaping, can be used as potted flowers and cut flowers for sale in the flower market, and has high economic benefit and good market prospect.
The flower color is an important quality index of ornamental plants, and the formation of the flower color is regulated by environmental factors such as temperature, illumination and the like and the nutrition level. The color of the hydrangea is obviously changed in the flowering process, obvious fading phenomenon can occur in the later development period of the calyx, and the quality of the color is rapidly reduced and the ornamental time is shortened due to unreasonable maintenance management or harvesting measures. For example, the excessive illumination of hydrangea causes burnt edges of the calyx, the fading rate of the flower is accelerated, the flowering period is shortened, the excessive shading causes the color to lighten, the content of the total anthocyanin is reduced, and the moderate shading is beneficial to the accumulation of the total anthocyanin and delays the color fading. At present, the related hydrangea has been put out of corresponding standards (LY/T1732-2008; GB/T28680-2012) in terms of quality product grades of potted flowers and cut flowers, and the classification method of the quality grades of the hydrangea is a visual inspection method, wherein qualitative descriptions such as 'uniform color, pure color, natural color change, luster' and the like are still adopted in the aspect of color quality evaluation, and a rapid, lossless and quantitative estimation method is not yet available. The anthocyanidin and its metabolite are key factors affecting plant flower color, and the response to environmental changes is more sensitive and rapid than the external flower color phenotype. In recent years, the study shows that the quality of the hydrangea flower is closely related to the content of the total anthocyanin, the total flavone and the total phenol which are internal physiological metabolism indexes. Studies by Schreiber et al (2011) showed that anthocyanin gradually accumulated as the flowering of the hydrangea progresses, reaching a maximum in the full bloom stage and decreasing in content after fading. The total flavone and total phenol have auxiliary effect on anthocyanin, so that the color is plump. The study by Peng et al (2021) showed that with advancing flowering, the total flavonoids and total phenols content of the hydrangea show a decreasing trend, highest in the bud period, gradually decreasing thereafter, and lowest in the final flowering period. In the existing research or based on a single color space model (Lab), the external color phenotype evaluation is carried out by naked eye observation or using a color measuring instrument, and anthocyanin, total flavone and total phenol content are used as color quality evaluation basis, so that the method is more scientific, comprehensive and accurate, the color quality change of the hydrangea in the flowering process can be prejudged, and the basis is provided for a producer to timely adjust the follow-up management, cultivation, harvesting and transportation strategy. However, the existing method for measuring the content of the total anthocyanin, the total flavone and the total phenol in the hydrangea calyx has the problems of long time consumption, strong specialization, high cost and the like although the accuracy is higher and the universality is better. Therefore, how to rapidly, nondestructively and quantitatively estimate and grade the color quality of the hydrangea becomes an important problem to be solved at present.
Disclosure of Invention
The invention aims to solve the defects in the prior art and provides a rapid, lossless and quantitative method for estimating the color quality of the hydrangea.
In order to achieve the above purpose, the technical scheme provided by the invention is as follows:
a method for rapidly estimating the color quality of an hydrangea ball comprises the following steps:
(1) Sample collection
Selecting different embroidered ball inflorescence samples, continuously observing the color change, and collecting fresh calyx for standby from the selected embroidered ball inflorescence samples in the full flowering period, the full flowering end period and the final flowering period;
(2) Acquiring an image of the sparassis crispa, extracting multi-color space component values
And (3) acquiring corresponding inflorescence images of the sparassis crispa samples selected in the step (1) before fresh calyx acquisition, and extracting color component values of 3 color spaces Lab, HSV and RGB from the sparassis crispa samples: brightness L, redness a, yellowness B, hue H, saturation S, brightness V, red R, green G, blue B;
(3) Determining the color quality index in the calyx, and calculating the color quality comprehensive score
Taking fresh calyx collected in the step (1), and measuring the color quality index: total anthocyanin, total flavone and total phenol content; after normalizing the obtained color quality indexes, calculating the weight of each index through an entropy weight method, and calculating to obtain a color quality comprehensive score;
(4) Building a color quality estimation model
Taking the color component values L, the redness a, the yellowness B, the hue H, the saturation S, the brightness V, the red R, the green G and the blue B obtained in the step (2) as independent variables, and the corresponding comprehensive scores of the color quality as independent variables to construct a color quality estimation model;
(5) Quality classification of silk ball flower color
Classifying the quality of the hydrangea according to the comprehensive score of the color quality, marking the acquired image, dividing the acquired image into a training set and a testing set, and classifying the color quality by utilizing random forest classification;
(6) Hydrangea suit evaluation
Collecting the embroidered ball inflorescence image to be evaluated, obtaining a color component value according to the method of the step (2), calculating a color quality comprehensive score according to the color quality estimation model of the step (4), and grading the color quality according to the step (5).
In more detail, the method for rapidly estimating the color quality of the hydrangea comprises the following specific steps:
(1) And (3) collecting field samples: selecting different embroidered ball inflorescence samples to be hung, continuously observing the color change, and collecting fresh calyx for standby from the hung inflorescences in the full-bloom period, the full-bloom end period and the final-bloom period to obtain different color quality samples.
Specifically, the full-bloom stage means that the hydrangea calyx is fully unfolded, and the whole inflorescence is colored; the end stage of full bloom refers to the fading of the flower calyx and inflorescence flower, and the end stage refers to the substantial fading of the flower calyx and inflorescence.
(2) Obtaining an sparassis crispa image, and extracting multi-color space component values: obtaining an image of the sparassis crispa: a single-lens reflex camera (Canon, EOS 90D) was used to obtain a high-definition (resolution: 6960X 4640) hydrangea image in RAW format. The pictures are unified into JPG pictures in RGB mode using Python PL library, and the pictures are unified into 100px pictures. The image K-means clustering method is used for extracting and segmenting the sparassis image, and a image color summarizer open source tool is used for extracting color component values L, a, b, H, S, V, R, G, B of 3 color spaces Lab, HSV and RGB.
Specifically, L is brightness, a is redness, and b is yellowness; h is hue, S is saturation, and V is brightness; r is red, G is green, and B is blue.
(3) Determining the contents of total anthocyanin, total flavone and total phenol, and establishing a comprehensive score of the color quality: and obtaining the color quality index by a conventional measuring method. The anthocyanin is measured by adopting a high performance liquid chromatography mass spectrometry (LC-MS), the total flavone is measured by adopting a colorimetric method, and the total phenol is measured by adopting a Folin-Ciocalteau method. After normalizing the measured index value, calculating the weight of each measured index by an entropy weight method to obtain the comprehensive index value of the color quality
(4) Building a flower color quality estimation model: color component value L, a, b, H, S, V, R, G, B is used as an independent variable, the color quality comprehensive score is used as an independent variable, and a color quality estimation model is constructed through random forest, support vector machine and bp neural network regression.
(5) Grading the color quality of the hydrangea: the color quality of the embroidered ball is classified according to the color quality comprehensive score, the acquired image is marked and divided into a training set and a testing set, and the color quality is classified by utilizing random forest classification.
(6) Hydrangea suit evaluation
Collecting an inflorescence image of an hydrangea needing to be evaluated, acquiring a color component value according to the method of the step (2), calculating a color quality comprehensive score according to the color quality estimation model of the step (4), and grading the color quality according to the step (5).
Compared with the prior art, the invention has the following advantages: the invention extracts color component values based on a multicolor space model, combines the measurement of physiological metabolism indexes affecting the color, utilizes a machine learning model to construct a color quality estimation model, quantitatively estimates the color quality of the hydrangea, judges the quality grade of the classified colors, can rapidly, nondestructively and quantitatively estimate and grade the color quality of the hydrangea by directly acquiring color information when estimating the color quality on site, and provides basis for optimizing the hydrangea maintenance management technology and the product quality grade grading method in production practice.
The method can quantitatively describe the color quality of the hydrangea, predicts the color quality change, and compared with the classification by conventional subjective observation, the method has the advantage that the data obtained by testing and statistics are more scientific, comprehensive and accurate.
The estimation method can be further applied to monitoring of color quality under the influence of cultivation and maintenance technologies such as fertilization and temperature, and evaluating of the color quality of the hydrangea with different nursery and varieties, and has important significance for standardized and standardized production of hydrangea products and promotion of high-quality development of industries.
Drawings
FIG. 1 is a flowchart of a method for rapidly estimating the color quality of an hydrangea according to the present invention;
FIG. 2 is a schematic diagram of the acquisition of the component values of the multiple color spaces of the selected hydrangea suit sample;
FIG. 3 is a graph showing the comparison of the changes of the color of the hydrangea with different cultivation treatments in the embodiment of the effect of the invention; t1: full illumination; t2:50% shading.
Detailed Description
The present invention will be described in detail with reference to the accompanying drawings. The following examples will assist those skilled in the art in further understanding the present invention, but are not intended to limit the invention in any way. It should be noted that variations and modifications could be made by those skilled in the art without departing from the inventive concept. These are all within the scope of the present invention.
Example 1
The invention discloses a method for rapidly estimating the color quality of an embroidered ball, which comprises the following steps:
(1) And (3) collecting field samples: ball of red series hydrangea variety 'Huashouju' (Hydrangea macrophylla)
'Hanaterari') is a test material, and plants with consistent growth vigor are selected and placed in nursery greenhouses respectively. 2 different cultivation treatments, namely full illumination (T1) and 50% shading (T2), are set, and the color change is continuously observed, so that samples with different color qualities are obtained. 50% shading the illumination intensity is controlled by a shading net (4 needles, light transmittance 50%). Before the experiment starts, an illuminometer (DT-8809A, shenzhen Hua Chengchang) is used for measuring illumination intensity in sunny weather, so that the shading degree is accurate; 50% of the shading strength is 736 mu mol.m -2 ·s -1 The total light is 1471 mu mol m -2 ·s -1 . Continuously observing the color change, and collecting fresh calyx from the card-hanging inflorescences in different flowering periods for standby. Fig. 3 shows that the color change of the hydrangea with different illumination intensities can be found to have obvious differences in the color phenotype characteristics of the hydrangea with different illumination intensities when the T1 and the T2 are processed, and reliable sample data is provided for the subsequent model establishment.
(2) Obtaining an sparassis crispa image, and extracting multi-color space component values: a single-lens reflex camera (Canon, EOS 90D) was used to obtain a high-definition (resolution: 6960X 4640) hydrangea image in RAW format. The pictures are unified into JPG pictures in RGB mode using Python PL library, and the pictures are unified into 100px pictures. Because uneven illumination and interference of complex background can influence the image segmentation effect, a color space capable of effectively extracting a complete target needs to be searched. The acquired image has rich color information, and the sparassis image has obvious color difference in the color space. The method can effectively divide the sparassis image based on the image K-means clustering method, and better divide the sparassis image from the background such as soil, leaves and the like. The image color summarizer open source tool was used to extract 3 color space Lab, HSV and RGB color component values, with a K-means cluster number of 5, and the average pixel value was used as the base color feature parameter value for each component to construct a dataset (as shown in fig. 2).
(3) And measuring the content of total anthocyanin, total flavone and total phenol. Fresh calyx of flowers at different flowering periods is collected from the put-on inflorescences. Accurately weighing 0.5g of fresh calyx, fully grinding with liquid nitrogen, leaching with 500 mu L of 2% methanolic formic acid, standing for 24h at 4 ℃ in the dark, centrifuging at 13500rpm for 9min, and filtering the supernatant with a 0.22 mu m microporous filter membrane for later use. And (3) determining the content of the total anthocyanin by using a high performance liquid chromatography mass spectrometry (LC-MS), determining the content of the total flavone by using a colorimetric method, and determining the content of the total phenol by using a Folin-Ciocalteau method. (4) establishing a comprehensive score of the flower color quality: the entropy weighting method is a method for calculating index weight according to the variation of each index, belongs to an objective weighting method, and can more accurately judge the influence of each variable on the whole. The greater the degree of variation of the index, the more information quantity it reflects, and the higher the corresponding weight. Calculating a weight value according to a formula (1), calculating an entropy value according to a formula (2), calculating an entropy weight value according to a formula (3), and calculating a color quality comprehensive index value of each sample according to a formula (4). Calculating the weights of all indexes of the total anthocyanin, the total flavone and the total phenol by an entropy weight method to obtain a comprehensive index value of the color quality, wherein the formula is as follows:
calculating weights:
calculating an entropy value:
calculating entropy weight:
calculating the comprehensive score of the flower color quality:
in the formula (1): x is x ij The j-th index value, y, for the i-th sample ij Is x ij Values normalized by dispersion; p (P) ij The j index weight of the i sample, E ij Entropy value of jth index for ith sample, w j For the j index value entropy weight, FQCI is the color quality comprehensive score. The weight of each index is shown in table 1, and the weight calculation result of the entropy weight method shows that the total anthocyanin plays a leading role in the comprehensive index of the color quality, and the total anthocyanin accounts for 59.284 percent.
Table 1 results of calculation of weights of the respective indices based on the entropy weight method
(5) Building a flower color quality estimation model: color component value L, a, b, H, S, V, R, G, B is taken as independent variable, and color quality comprehensive score is taken as dependent variable and is processed by random forest, support vector machine and Bp
And constructing a color quality estimation model by neural network regression. The regression algorithm was run in Windows 10,RAM 16GB,CPU AMD Ryzen 75800H, the programming language python3.9, and the tool PyCharm. The dataset was divided into training and test sets (7:3) and 3-fold cross-validation was performed.
The random forest, the support vector machine and the Bp neural network regression are all classical machine learning algorithms. The random forest regression is integration of decision trees, sample observation and characteristic variables of a modeling data set are respectively and randomly sampled, each sampling result is a tree, each tree can generate rules and judgment values conforming to the attribute of the tree, and finally, the rules and judgment values of all decision trees are integrated, so that regression of a random forest algorithm is realized. The support vector machine regression uses a nonlinear mapping to map data into a high-dimensional data feature space, thereby converting nonlinearities into linearity. The bp neural network is a multi-layer feedforward network trained according to an error back propagation algorithm, and is one of the most widely applied neural network models at present. The learning rule of the bp neural network is to continuously adjust the weight and the threshold value of the network by using a steepest descent method through back propagation, so that the square sum of errors of the network is minimized. Model evaluation uses the determination coefficient (R 2 ) And Root Mean Square Error (RMSE) quantization index. When R is 2 The closer the result is to 1 the higher the model accuracy, while the smaller the RMSE value is, the higher the model accuracy is.
The results show that R of 3 models 2 Value Bp neural netThe complex is larger than the support vector machine and random forest, while RMSE is smaller than the support vector machine and random forest, indicating that the fitting ability of Bp neural network is optimal (table 2).
TABLE 2 evaluation of color quality of different models model results
(6) Grading the color quality of the hydrangea: the color quality of the embroidered ball is classified according to the color quality integrated score, and the color quality integrated score is classified into one stage, 0.3 to 0.7 is classified into two stages, and 0 to 0.3 is classified into three stages according to the color quality integrated score of 0.7 to 1 (table 3). Labeling the acquired images according to the first level, the second level and the third level, and carrying out random oversampling treatment on labels of different categories so as to enable the duty ratios of the different categories to be consistent and increase the number of samples to a certain extent. In order to make the training effect of the model better, the variable data which reach the classification balance after the oversampling treatment is standardized, and then the data are divided into a training set and a testing set according to the proportion of 7:3, and the random forest classification is utilized for classifying the flower color quality. The environment of the classification algorithm is Windows 10,RAM 16GB,CPU AMD Ryzen 75800H, the programming language is python3.9, and the tool is PyCharm.
(7) Classification effectiveness is measured by accuracy, recall, precision, and comprehensive estimation (F1-score). The accuracy predicts the percentage of the total sample that is the correct result. The accuracy is the degree of prediction accuracy in the model alignment sample results, the Recall (Recall) is a measure of the model coverage, and the accuracy is a set of relative indicators. The comprehensive estimation rate is a weighted harmonic mean of the precision rate and the recall rate, and can be considered simultaneously when evaluating the performance of the model, so that the comprehensive estimation rate is more reasonable compared with a single index.
(8) The confusion matrix is also called an error matrix, is a specific table display mode, and is output according to the fitting result of the test set, so that whether the model confuses 3 categories or not can be seen, and the diagonal is a predicted correct value. The results are shown in Table 4, and most of the conditions of the test set labels and the prediction results are consistent, most of the samples are correctly predicted, wherein the probability of the first-stage color quality prediction is 0.73, the second-stage color quality is 0.5, the probability of the third-stage prediction is higher, and the probability of the third-stage prediction is 0.96, so that samples with different color qualities are effectively distinguished. According to the test data (true value) and the predicted data (predicted value), the fitting result of the random forest classification is obtained through calculation, and according to the table 5, the comprehensive estimation rates of the random forest model on different flower color quality grades in the test set and the training set are respectively 0.667 and 0.907, the ideal level (F1-score is more than or equal to 0.9) is achieved in the training set, and the overall estimation effect is quite considerable.
TABLE 3 color quality ranking
Note that: in the table, the dividing range of the color quality comprehensive score includes an upper limit value and a lower limit value
TABLE 4 color quality confusion matrix based on random forest classification
TABLE 5 fit of flower quality based on random forest classification
The previous description of the embodiments is provided to facilitate a person of ordinary skill in the art in order to make and use the present invention. It will be apparent to those skilled in the art that various modifications can be readily made to these embodiments and the generic principles described herein may be applied to other embodiments without the use of the inventive faculty. Therefore, the present invention is not limited to the above-described embodiments, and those skilled in the art, based on the present disclosure, should make improvements and modifications without departing from the scope of the present invention.
Effect example 2
Optimizing the hydrangea maintenance management technology by adopting the estimation method
The overall values of the different treatment (total light, 50% shading) flower color qualities using the example 1 estimation method are shown in table 2-1, and the results show that the different treatment flower color qualities are significantly different. With the change of time, the color quality comprehensive value gradually decreases, and the color quality comprehensive index value is lower than the shading value under the whole full illumination treatment, which indicates that the shading effectively improves the color quality and delays the color quality from decreasing.
TABLE 2-1 evaluation of different maintenance management of the color quality of the hydrangea
Claims (7)
1. A method for rapidly estimating the color quality of an hydrangea is characterized by comprising the following steps:
(1) Sample collection
Selecting different embroidered ball inflorescence samples, continuously observing the color change, and collecting fresh calyx for standby from the selected embroidered ball inflorescence samples in the full flowering period, the full flowering end period and the final flowering period;
(2) Acquiring an image of the sparassis crispa, extracting multi-color space component values
And (3) acquiring corresponding inflorescence images of the sparassis crispa samples selected in the step (1) before fresh calyx acquisition, and extracting color component values of 3 color spaces Lab, HSV and RGB from the sparassis crispa samples: brightness L, redness a, yellowness B, hue H, saturation S, brightness V, red R, green G, blue B;
(3) Determining the color quality index in the calyx, and calculating the color quality comprehensive score
Taking fresh calyx collected in the step (1), and measuring the color quality index: total anthocyanin, total flavone and total phenol content; after normalizing the obtained color quality indexes, calculating the weight of each index through an entropy weight method, and calculating to obtain a color quality comprehensive score;
(4) Building a color quality estimation model
Taking the color component values L, the redness a, the yellowness B, the hue H, the saturation S, the brightness V, the red R, the green G and the blue B obtained in the step (2) as independent variables, and the corresponding comprehensive scores of the color quality as independent variables to construct a color quality estimation model;
(5) Quality classification of silk ball flower color
Classifying the quality of the hydrangea according to the comprehensive score of the color quality, marking the acquired image, dividing the acquired image into a training set and a testing set, and classifying the color quality by utilizing random forest classification;
(6) Hydrangea suit evaluation
Collecting the embroidered ball inflorescence image to be evaluated, obtaining a color component value according to the method of the step (2), calculating a color quality comprehensive score according to the color quality estimation model of the step (4), and grading the color quality according to the step (5).
2. The method for rapidly estimating the color quality of the hydrangea according to claim 1, wherein in the step (2), before extracting the color component values, collected images are unified into JPG images in RGB mode, and simultaneously unified into images of 100px, and then the images are segmented by adopting a clustering method based on K means of the images.
3. The method for rapidly estimating the quality of the hydrangea flower according to claim 1, wherein the determination of the total anthocyanin content in the step (3) adopts a high performance liquid chromatography mass spectrometry technology, the determination of the total flavone content adopts a colorimetric method, and the determination of the total phenol content adopts a Folin-Ciocalteau method.
4. The method for rapidly estimating the color quality of the hydrangea according to claim 1, wherein the color quality integrated score in the step (3) is calculated as follows:
calculating weights:
calculating an entropy value:
calculating entropy weight:
calculating the comprehensive score of the flower color quality:
wherein: x is x ij The j-th index value, y, for the i-th sample ij Is x ij Values normalized by dispersion; p (P) ij The j index weight of the i sample, E ij The j index entropy value, w, of the i sample j And (3) the entropy weight of the j-th index value, and FQCI is the comprehensive score of the color quality.
5. The method for rapidly estimating the color quality of an embroidered ball according to claim 4, wherein the weights of the respective indexes are: 59.2884% of total anthocyanin, 16.544% of total flavone and 24.172% of total phenol.
6. The method for rapidly estimating the color quality of the hydrangea according to claim 1, wherein in the step (4), a Bp neural network regression is used to construct a color quality estimation model.
7. The method for rapid estimation of color quality of embroidered ball according to claim 1, wherein the color quality classification in the step (6) is: the color quality comprehensive score value is between 0.7 and 1, the color quality comprehensive score value is between 0.3 and 0.7, and the color quality comprehensive score value is between 0 and 0.3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310566659.2A CN116563258A (en) | 2023-05-19 | 2023-05-19 | Method for rapidly estimating color quality of hydrangea |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310566659.2A CN116563258A (en) | 2023-05-19 | 2023-05-19 | Method for rapidly estimating color quality of hydrangea |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116563258A true CN116563258A (en) | 2023-08-08 |
Family
ID=87497964
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310566659.2A Pending CN116563258A (en) | 2023-05-19 | 2023-05-19 | Method for rapidly estimating color quality of hydrangea |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116563258A (en) |
-
2023
- 2023-05-19 CN CN202310566659.2A patent/CN116563258A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110763698B (en) | Hyperspectral citrus leaf disease identification method based on characteristic wavelength | |
| CN106845497B (en) | Corn early-stage image drought identification method based on multi-feature fusion | |
| Barry et al. | Crown‐scale evaluation of spectral indices for defoliated and discoloured eucalypts | |
| CN112287886B (en) | Wheat plant nitrogen content estimation method based on hyperspectral image fusion map features | |
| Chen et al. | Diagnosing of rice nitrogen stress based on static scanning technology and image information extraction | |
| CN116543316B (en) | Method for identifying turf in paddy field by utilizing multi-time-phase high-resolution satellite image | |
| CN108710766A (en) | A kind of hothouse plants liquid manure machine tune fertilizer calculation method of parameters based on growth model | |
| CN110189793A (en) | The building of wheat nitrogenous fertilizer physiological use efficiency estimation models and wheat varieties with different N efficiency classification based on EO-1 hyperion | |
| CN114677033B (en) | Laboratory cultivation data full-period flow intelligent monitoring analysis management system | |
| CN107886497A (en) | A kind of wheat flag leaf automatic nondestructive detection and measuring method based on image | |
| Parsons et al. | Image analysis and statistical modelling for measurement and quality assessment of ornamental horticulture crops in glasshouses | |
| CN115879836B (en) | Soybean crop remote sensing large-area rapid yield estimation method based on coupling mechanism model | |
| CN112541659A (en) | Agricultural planting agricultural product quality information acquisition sharing system based on big data | |
| CN116563258A (en) | Method for rapidly estimating color quality of hydrangea | |
| CN107853054A (en) | Coconut Evaluation of Cold Tolerance method based on mathematical model | |
| CN112528726A (en) | Aphis gossypii insect pest monitoring method and system based on spectral imaging and deep learning | |
| CN106568722A (en) | Spectrum technology-based facility cucumber disease early warning method, and device | |
| CN116739403A (en) | Fruit comprehensive quality evaluation method based on quality condition distribution | |
| CN115728249A (en) | Prediction method for chlorophyll content of tomato seedlings and processing terminal | |
| CN114612794A (en) | Remote sensing identification method for land covering and planting structure in finely-divided agricultural area | |
| AU2021101996A4 (en) | Nutrient deficiency stress detection and prediction in corn fields from aerial images using machine learning | |
| CN109118488B (en) | Method for identifying cold resistance of new anthurium andraeanum germplasm | |
| CN114494689A (en) | Identification method of tomato drought stress | |
| Ding et al. | Rice lodging area extraction based on YCbCr spatial and texture features | |
| KR20200080432A (en) | Method for diagnosing growth state employing analyzing of image of strawberry |
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 |