CN110517160A - A kind of quality grading method and quality grading system of agricultural product - Google Patents
A kind of quality grading method and quality grading system of agricultural product Download PDFInfo
- Publication number
- CN110517160A CN110517160A CN201910712867.2A CN201910712867A CN110517160A CN 110517160 A CN110517160 A CN 110517160A CN 201910712867 A CN201910712867 A CN 201910712867A CN 110517160 A CN110517160 A CN 110517160A
- Authority
- CN
- China
- Prior art keywords
- quality
- agricultural product
- sample
- matrix
- training
- 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
- 238000000034 method Methods 0.000 title claims abstract description 38
- 239000011159 matrix material Substances 0.000 claims abstract description 92
- 238000013441 quality evaluation Methods 0.000 claims abstract description 71
- 238000012549 training Methods 0.000 claims abstract description 71
- 238000000513 principal component analysis Methods 0.000 claims abstract description 17
- 238000001303 quality assessment method Methods 0.000 claims abstract description 10
- 239000000126 substance Substances 0.000 claims abstract description 8
- 244000000010 microbial pathogen Species 0.000 claims abstract description 6
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 6
- 238000001514 detection method Methods 0.000 claims description 19
- 238000007781 pre-processing Methods 0.000 claims description 15
- 238000012545 processing Methods 0.000 claims description 13
- 230000003287 optical effect Effects 0.000 claims description 8
- 238000012216 screening Methods 0.000 claims description 8
- 238000011156 evaluation Methods 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 5
- 238000004891 communication Methods 0.000 claims description 3
- 239000002932 luster Substances 0.000 claims description 3
- 239000002699 waste material Substances 0.000 abstract 1
- 230000008859 change Effects 0.000 description 7
- 238000009659 non-destructive testing Methods 0.000 description 4
- 238000010521 absorption reaction Methods 0.000 description 3
- 235000013399 edible fruits Nutrition 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000004806 packaging method and process Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 239000003637 basic solution Substances 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 230000001186 cumulative effect Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000000149 penetrating effect Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 239000000243 solution Substances 0.000 description 2
- 230000004308 accommodation Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 235000013305 food Nutrition 0.000 description 1
- 235000001497 healthy food Nutrition 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 235000015097 nutrients Nutrition 0.000 description 1
- 239000000447 pesticide residue Substances 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
- 238000002310 reflectometry Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 241000894007 species Species 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000017105 transposition Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K17/00—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations
- G06K17/0022—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations arrangements or provisions for transferring data to distant stations, e.g. from a sensing device
- G06K17/0025—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations arrangements or provisions for transferring data to distant stations, e.g. from a sensing device the arrangement consisting of a wireless interrogation device in combination with a device for optically marking the record carrier
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- General Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Primary Health Care (AREA)
- Computer Networks & Wireless Communication (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Economics (AREA)
- Mining & Mineral Resources (AREA)
- Marine Sciences & Fisheries (AREA)
- Animal Husbandry (AREA)
- Agronomy & Crop Science (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The present invention relates to computer fields, more particularly to a kind of quality grading method and quality grading system of agricultural product, the method includes determining agricultural product quality evaluation index, principal component-Adaboost conjunctive model is constructed, the principal component matrix of quality index is obtained according to principal component analysis in the model；It is trained according to principal component matrix of the Adaboost algorithm to quality evaluation index, completes the quality assessment result that will need the agricultural product assessed input principal component-Adaboost conjunctive model that the agricultural product can be obtained after training；Wherein agricultural product quality evaluation index includes at least density, freshness, conductivity, color, water content, microbial pathogens amount and the chemical residues amount of agricultural product；The present invention can effectively assess the quality of agricultural product by the validity feature of agricultural product, and waste of human resource is reduced.
Description
Technical Field
The invention relates to the field of computers, in particular to a quality grading method and a quality grading system for agricultural products.
Background
From the agricultural aspect, agriculture is not only a basic industry but also a strategic industry, and must be concerned with solving the most basic clothes and food accommodation problem of the masses.
In recent years, with the development of society, the living standard of people is gradually improved, and people are pursuing green, environment-friendly and healthy food, including agricultural products. The quality of agricultural products is closely related to the life of consumers, influences the enthusiasm of agricultural producers during planting, and has great influence on the development of the folk lives of the future China. However, the phenomena of improper cultivation, seasonal climate influence, excessive pesticide residue, additive use and the like exist in the production process of agricultural products, so that the produced agricultural products have poor quality, and the marketing and the consumer purchasing of the agricultural products are influenced.
Most of the current agricultural product quality assessment is manually completed, the process is time-consuming and labor-consuming, the assessment effect is poor, and no method can effectively assess and grade the quality of the agricultural product through the effective characteristics of the agricultural product.
Disclosure of Invention
In order to effectively evaluate and grade the quality of agricultural products through the effective characteristics of the agricultural products, the invention provides a quality grading method and a quality grading system of the agricultural products, wherein the quality grading method comprises the steps of determining quality evaluation indexes of the agricultural products, constructing a principal component-Adaboost combined model, and obtaining a principal component matrix of the quality evaluation indexes in the model according to principal component analysis; training a principal component matrix of a quality evaluation index according to an Adaboost algorithm, and inputting the agricultural product to be evaluated into a principal component-Adaboost combined model after training to obtain a quality evaluation result of the agricultural product; wherein the quality evaluation indexes of the agricultural products at least comprise density, freshness, conductivity, color and luster, water content, microbial pathogen amount and chemical residual amount of the agricultural products.
Further, as shown in fig. 1, the principal component matrix for obtaining the quality assessment index according to the principal component analysis includes:
s101, preprocessing data of the quality evaluation indexes of the agricultural products;
s102, establishing an autocorrelation matrix of the quality index of the agricultural product, and calculating and acquiring a characteristic value and a characteristic vector of the quality index of the agricultural product according to the autocorrelation matrix;
s103, calculating the variance contribution rate and the accumulated variance contribution rate, selecting an index with the accumulated variance contribution rate exceeding 85% in the quality indexes of the agricultural products, and counting the number of the indexes with the accumulated variance contribution rate exceeding 85%;
and S104, calculating a principal component matrix of the quality index of the agricultural product according to the selected characteristic vector of the quality index of the agricultural product.
Further, the data preprocessing of the agricultural product quality evaluation index comprises:
S101A, the agricultural product quality evaluation indexes comprise n samples, each sample comprises m quality evaluation indexes, each sample is taken as a row, each quality evaluation index is taken as a column, and an n x m-order agricultural product quality evaluation index matrix is formed;
S101B, carrying out standardization processing on each index in the matrix to obtain a standardized matrix; the normalization process is represented as:
wherein, x'ijAn element representing the ith row and the jth column in the normalized matrix; x is the number ofijThe element which represents the ith row and the jth column in the quality evaluation index matrix of the agricultural product;representing the average value of j-th row indexes in the quality evaluation index matrix of the agricultural products; sjAnd the variance of the jth column index in the agricultural product quality evaluation index matrix is represented.
Further, establishing an autocorrelation matrix of the quality index of the agricultural product, and calculating and obtaining the eigenvalue and the eigenvector of the quality index of the agricultural product according to the autocorrelation matrix comprises: constructing a characteristic polynomial of the autocorrelation matrix, making the characteristic polynomial of the autocorrelation matrix equal to 0, solving characteristic values, and bringing the characteristic values into one by one, wherein the solved basic solution system is the characteristic vector, and the autocorrelation matrix of the agricultural product quality evaluation index matrix is expressed as follows:
S＝R'TR'/(N-1)；
wherein S is the autocorrelation of the agricultural product quality evaluation index matrixA matrix; r' is a standardized matrix generated after the sample is subjected to standardized processing; n is the number of samples; superscript T represents the transpose of the matrix; the eigenvalue vector solved from the autocorrelation matrix is denoted λ ═ λ1,λ2,…,λm]And the feature vector is expressed as U ═ U1,u2,…,um](ii) a Wherein λmExpressed as the m-th value, u, in the eigenvalue vectormRepresenting the m-th value in the feature vector, i.e. according to λmAnd solving the obtained feature vector.
Further, as shown in fig. 1, inputting the principal component matrix of the quality assessment index into the Adaboost algorithm model for training includes:
s201, inputting training samples into an Adaboost algorithm model, and setting the initial weight of each sample to be 1/n; wherein the training samples comprise n samples, wherein the ith sample is represented as (x)i,yi)，xiIs the feature vector of the i-th training sample, yiThe evaluation result of the ith sample;
s202, setting the maximum iteration times of an Adaboost algorithm model, and performing k rounds of training on the quality index of the agricultural product to obtain k basic classifiers;
s203, in the training process, optimizing the sample relearning weight of the basic classifier by using a fuzzy rule corresponding to fitness, and taking the relearning weight of the current basic trainer as the sample weight of the next basic trainer;
s204, combining all the basic classifiers into a strong classifier, and inputting the quality index data of the agricultural products to be tested to obtain an evaluation result; wherein combining all basic classifiers into a strong classifier is represented as:
H(x)＝sign(sum(αih(xi))),i∈{1,2,...,k}；
wherein H (x) is a strong classifier; h (x)i) Is the output of the ith basic classifier; alpha is alphaiIs the weight of the ith basic classifier.
Further, optimizing the obtained weight of the basic classifier by using the fuzzy rule corresponding to the fitness includes:
S203A, calculating the fitness of the sample, and screening out the sample with the maximum fitness and the fuzzy rule corresponding to the sample;
S203B, calculating a classification error rate and a classification weight according to the fuzzy rule when the fitness is maximum;
S203C, updating the optimal weight of the classifier according to the classification error rate and the classification weight.
Further, the fitnessExpressed as:
wherein,classify the weak classifiers with the correct rate expressed asωkFor the weight of each sample after training,for the k sample in the fuzzy rule betatDegree of membership, x, of the fuzzy set to which the pair belongskA k training sample in the characteristic sample set is obtained;classify the weak classifiers with an error rate expressed askmaxClassifying an error rate threshold for the weak classifier; k | yk＝ytRepresents the training result y of the k-th sample among the n training sampleskAnd quality evaluation result y in the feature sampletAre equal.
Further, calculating the classification error rate and the classification weight includes:
wherein epsilon (. beta.)t) For blurring the rule betatThe classification error rate of (2); thetatIs the weight of the weak classifier; omegaiIs the weight of the sample i and,for the k sample in the fuzzy rule betatDegree of membership, x, of the fuzzy set to which the pair belongsiIs the ith training sample; i | yi＝ytAnd (4) showing.
Further, the step of obtaining the optimal weight value of the classifier according to the classification error rate and the classification weight value updating includes:
wherein, ω isi(t +1) is the optimal weight of the sample used for training the classifier according to the error rate of classification; alpha is alphatIs a normalization factor; thetatIs the weight of the weak classifier;for the k sample in the fuzzy rule betatThe degree of membership of the fuzzy set to which the pair belongs.
A quality grading system of agricultural products comprises an agricultural product quality detection device, a quality grading server and a user side, wherein the agricultural product quality detection device is installed in an agricultural product processing workshop and comprises an acoustic characteristic analyzer, an electrical characteristic analyzer, an optical characteristic analyzer, an X-ray and laser analyzer, a two-dimensional code generator and a communication module; the quality grading server comprises an agricultural product database and an agricultural product quality evaluation unit; the agricultural product quality evaluation unit is used for screening the grading indexes from the agricultural product quality detection device, calculating according to the grading indexes to obtain the grading information, and storing the grading information and the grading indexes in the agricultural product database, and comprises a main component analysis data preprocessing module and an Adaboost classifier training module; the principal component analysis data preprocessing module comprises a parameter standardization unit, an autocorrelation matrix acquisition unit and a principal component number calculation unit; the Adaboost classifier training module comprises a weight initialization unit, a training weak classifier unit and a weak classifier combination strong classifier unit; wherein:
the agricultural product quality detection device is used for acquiring an agricultural product quality evaluation index from an agricultural product nondestructive analyzer;
the parameter standardization unit is used for standardizing the quality evaluation indexes of the obtained agricultural products and eliminating the influence of the quality indexes due to inconsistent dimensions and units;
the autocorrelation matrix obtaining unit is used for establishing an autocorrelation matrix of the quality index of the agricultural product and obtaining the eigenvalue and the eigenvector of the quality index of the agricultural product according to the autocorrelation matrix;
the main component number calculating unit is used for calculating the variance contribution rate and the accumulated variance contribution rate of each agricultural product quality index and determining the number of main components of the agricultural product quality index according to the accumulated variance contribution rate;
the weight initialization unit is used for initializing the weight of each sample;
the training weak classifier unit is used for updating the weights of all training samples;
the weak classifier combination strong classifier unit is used for increasing the weight of a weak classifier with small classification error rate, reducing the weight of a weak classifier with large classification error rate and combining the weak classifiers into a strong classifier.
The method is characterized in that dimension reduction processing is carried out on the whole original data based on principal component analysis and a classification model of Adaboost, a plurality of weak classifiers are trained by utilizing an Adaboost algorithm, and the weak classifiers are combined into a strong classifier to classify the quality of agricultural products; the model is built by combining principal component analysis and Adaboost; firstly, carrying out dimensionality reduction processing on the whole original data by utilizing principal component analysis to reduce information redundancy; secondly, a strong classifier is trained by using an Adaboost algorithm and used for classifying agricultural products, so that the quality evaluation effect of the agricultural products is achieved.
Drawings
FIG. 1 is a schematic flow diagram of a method of quality grading agricultural products according to the present invention;
FIG. 2 is a schematic diagram of the structure of a quality grading system for agricultural products according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention provides a quality grading method and a quality grading system of agricultural products, which comprises the steps of determining quality evaluation indexes of the agricultural products, constructing a principal component-Adaboost combined model, and obtaining a principal component matrix of the quality evaluation indexes in the model according to principal component analysis; training a principal component matrix of a quality evaluation index according to an Adaboost algorithm, and inputting the agricultural product to be evaluated into a principal component-Adaboost combined model after training to obtain a quality evaluation result of the agricultural product; wherein the quality evaluation indexes of the agricultural products at least comprise density, freshness, conductivity, color and luster, water content, microbial pathogen amount and chemical residual amount of the agricultural products.
Further, as shown in fig. 1, the principal component matrix for obtaining the quality assessment index according to the principal component analysis includes:
s101, preprocessing data of the quality evaluation indexes of the agricultural products;
s102, establishing an autocorrelation matrix of the quality index of the agricultural product, and calculating and acquiring a characteristic value and a characteristic vector of the quality index of the agricultural product according to the autocorrelation matrix;
s103, calculating the variance contribution rate and the accumulated variance contribution rate, selecting an index with the accumulated variance contribution rate exceeding 85% in the quality indexes of the agricultural products, and counting the number of the indexes with the accumulated variance contribution rate exceeding 85%; where the variance contribution σ (h) is calculated as:
the calculation of the cumulative variance contribution δ (h) is expressed as:
wherein λ isiIs the ith eigenvalue of the autocorrelation matrix S; h is the total quality index number of the agricultural products; and m is the number of the quality indexes of the agricultural products of which the cumulative variance contribution rate exceeds 85 percent.
And S104, calculating a principal component matrix of the quality index of the agricultural product according to the selected characteristic vector of the quality index of the agricultural product.
Further, the data preprocessing of the agricultural product quality evaluation index comprises:
S101A, the agricultural product quality evaluation indexes comprise n samples, each sample comprises m quality evaluation indexes, each sample is taken as a row, each quality evaluation index is taken as a column, and an n x m-order agricultural product quality evaluation index matrix is formed;
S101B, carrying out standardization processing on each index in the matrix to obtain a standardized matrix; the normalization process is represented as:
wherein, x'ijAn element representing the ith row and the jth column in the normalized matrix; x is the number ofijThe element which represents the ith row and the jth column in the quality evaluation index matrix of the agricultural product;representing the average value of j-th row indexes in the quality evaluation index matrix of the agricultural products; sjAnd the variance of the jth column index in the agricultural product quality evaluation index matrix is represented.
Further, establishing an autocorrelation matrix of the agricultural product quality index, calculating and obtaining the characteristic value and the characteristic vector of the agricultural product quality index according to the autocorrelation matrix, including constructing a characteristic polynomial of the autocorrelation matrix, making the characteristic polynomial of the autocorrelation matrix equal to 0, solving the characteristic value, bringing the characteristic values one by one, solving a basic solution system which is the characteristic vector, and expressing the autocorrelation matrix of the agricultural product quality evaluation index matrix as follows:
S＝R'TR'/(N-1)；
s is an autocorrelation matrix of an agricultural product quality evaluation index matrix; r' is a standardized matrix generated after the sample is subjected to standardized processing; n is the number of samples; transposition of superscript T representation matrix the eigenvalue vector obtained from the autocorrelation matrix solution is represented as λ ═ λ1,λ2,…,λm]And the eigenvector matrix is expressed as U ═ U1,u2,…,um](ii) a Wherein λmExpressed as the m-th value, u, in the eigenvalue vectormRepresenting the m-th value in the feature vector, i.e. according to λmThe principal component matrix of the quality index is represented as X by the solved eigenvectorn×m＝R'n×hUh×m(ii) a Wherein, Xn×mRepresents a main component matrix, R 'representing a quality index'n×hRepresenting the normalized sample to produce a normalized matrix, Uh×mRepresents a feature vector of oneThe feature value corresponds to a feature vector, and the dimension of each feature vector is h.
Adaboost is an iterative algorithm, and the core idea is to train different training sets against the same training setClassifier(weak classifiers), then the weak classifiers are integrated to form a stronger final classifier (strong classifier), as shown in fig. 1, the invention inputs the principal component matrix of the quality evaluation index into an Adaboost algorithm model for training, including:
s201, inputting training samples into an Adaboost algorithm model, and setting the initial weight of each sample to be 1/n; wherein the training samples comprise n samples, wherein the ith sample is represented as (x)i,yi)，xiIs the feature vector of the i-th training sample, yiThe evaluation result of the ith sample;
s202, setting the maximum iteration times of an Adaboost algorithm model, and performing k rounds of training on the quality index of the agricultural product to obtain k basic classifiers;
s203, in the training process, optimizing the sample relearning weight of the basic classifier by using a fuzzy rule corresponding to fitness, and taking the relearning weight of the current basic trainer as the sample weight of the next basic trainer;
s204, combining all the basic classifiers into a strong classifier, and inputting the quality index data of the agricultural products to be tested to obtain an evaluation result; wherein combining all basic classifiers into a strong classifier is represented as:
H(x)＝sign(sum(αih(xi))),i∈{1,2,...,k}；
wherein H (x) is a strong classifier; h (x)i) Is the output of the ith basic classifier; alpha is alphaiIs the weight of the ith basic classifier.
Further, optimizing the obtained weight of the basic classifier by using the fuzzy rule corresponding to the fitness includes:
S203A, calculating the fitness of the sample, and screening out the sample with the maximum fitness and the fuzzy rule corresponding to the sample;
S203B, calculating a classification error rate and a classification weight according to the fuzzy rule when the fitness is maximum;
S203C, updating the optimal weight of the classifier according to the classification error rate and the classification weight.
Further, the fitnessExpressed as:
wherein,classify the correct rate for weak classifiers, i.e. in k rounds of training, sample xkTraining classification result y of (i ═ 1,2, …, n)kQuality assessment result y in exact sum sample sett(t is equal to 1,2, …, n) and is represented byωkFor the weight of each sample after training,for the k sample in the fuzzy rule betatDegree of membership, x, of the fuzzy set to which the pair belongskA k training sample in the characteristic sample set is obtained;classifying error rates for weak classifiers, i.e. in k rounds of training, sample xkTraining classification result y of (i ═ 1,2, …, n)kQuality assessment result y in exact sum sample sett(t ═ 1,2, …, n) inequality, expressed askmaxClassifying an error rate threshold for the weak classifier; k | yk＝ytRepresents the training result y of the k-th sample among the n training sampleskAnd featuresQuality evaluation result y in sampletAre equal.
Further, calculating the classification error rate and the classification weight includes:
wherein epsilon (. beta.)t) For blurring the rule betatThe classification error rate of (2); thetatIs the weight of the weak classifier; omegaiIs the weight of the sample i and,for the k sample in the fuzzy rule betatDegree of membership, x, of the fuzzy set to which the pair belongsiIs the ith training sample; i | yi＝ytIt is indicated that the training result of the ith sample is equal to the quality evaluation result of the feature sample in the n training samples.
Fuzzy rule beta adopted by the inventiontIs defined as: if the sample set is I ═ x1,x2,…,xn}，xjIs S ═ S1，S2，...，SnjJ is 1,2 …, n, and y is the categoryj∈{y1,y2,…,ym}; then the Adaboost classifier input variable is x and the resulting fuzzy rule can be expressed as:
βj:if x1 is S1j and x2 is S2j and … and xn is Snj then Y＝yj。
where Y represents the output category result under the assumed conditions, i.e., if x1, x2, etc. are in the corresponding vague set, the output category is consistent with the sample species.
Further, the step of obtaining the optimal weight value of the classifier according to the classification error rate and the classification weight value updating includes:
wherein, ω isi(t +1) is the optimal weight of the sample used to train the classifier, ω, based on the error rate of the classificationi(t) is the weight of the classifier at the last iteration; alpha is alphatTo normalize the factors, the goal is to uniformly map the data to [0,1]In the interval, eliminating the influence of the data dimension and the numerical value; thetatIs the weight of the weak classifier;for the k sample in the fuzzy rule betatThe degree of membership of the fuzzy set to which the pair belongs.
Utilizing the optimized classifier to perform quality data characteristic set sampling on n unknown agricultural productsClassifying to obtain quality grading result y of agricultural productsmax(xk) It can be expressed as:
a quality grading system of agricultural products is shown in figure 2 and comprises an agricultural product quality detection device, a quality grading server and a user side, wherein the agricultural product quality detection device is installed in an agricultural product processing workshop and comprises an acoustic characteristic analyzer, an electrical characteristic analyzer, an optical characteristic analyzer, an X-ray and laser analyzer, a two-dimensional code generator and a communication module; the quality grading server comprises an agricultural product database and an agricultural product quality evaluation unit; the agricultural product quality evaluation unit is used for screening the grading indexes from the agricultural product quality detection device, calculating according to the grading indexes to obtain grading information, and storing the grading information and the grading indexes of the agricultural products in the agricultural product database, and comprises a main component analysis data preprocessing module and an Adaboost classifier training module; the principal component analysis data preprocessing module comprises a parameter standardization unit, an autocorrelation matrix acquisition unit and a principal component number calculation unit; the Adaboost classifier training module comprises a weight initialization unit, a training weak classifier unit and a weak classifier combination strong classifier unit; wherein:
the agricultural product quality detection device is used for acquiring an agricultural product quality evaluation index from an agricultural product nondestructive analyzer;
the parameter standardization unit is used for standardizing the quality evaluation indexes of the obtained agricultural products and eliminating the influence of the quality indexes due to inconsistent dimensions and units;
the autocorrelation matrix obtaining unit is used for establishing an autocorrelation matrix of the quality index of the agricultural product and obtaining the eigenvalue and the eigenvector of the quality index of the agricultural product according to the autocorrelation matrix;
the main component number calculating unit is used for calculating the variance contribution rate and the accumulated variance contribution rate of each agricultural product quality index and determining the number of main components of the agricultural product quality index according to the accumulated variance contribution rate;
the weight initialization unit is used for initializing the weight of each sample;
the training weak classifier unit is used for updating the weights of all training samples;
the weak classifier combination strong classifier unit is used for increasing the weight of a weak classifier with small classification error rate, reducing the weight of a weak classifier with large classification error rate and combining the weak classifiers into a strong classifier.
The two-dimensional code of the agricultural product and the classification information of the agricultural product are stored in an agricultural product database, the two-dimensional code of the agricultural product is associated with the classification information and the storage position of the classification information of the agricultural product in the agricultural product database, and when a user terminal scans the two-dimensional code for query, the quality classification server calls the classification information and the classification index of the agricultural product according to the two-dimensional code and sends the classification information and the classification index to the user terminal.
In the invention, the principal component analysis data preprocessing module screens out different quality indexes of agricultural products according to different agricultural products, so that computing resources for all quality indexes can be saved, and the agricultural products can be analyzed in a targeted manner; the invention further classifies and trains the screened indexes by using an Adaboost classifier training module, so as to obtain the grade of the agricultural product; in an actual application scene, screening out quality indexes by using a principal component analysis data preprocessing module, inputting the quality indexes into a trained Adaboost classifier for the agricultural products with the screened quality indexes, and obtaining quality grades, wherein the Adaboost classifier training module trains the Adaboost classifier according to historical classification data, the historical data comprises the quality indexes and grading information of the agricultural products, and similarly, the principal component analysis data preprocessing module screens out the quality indexes according to the quality indexes of the agricultural products in the historical data.
After agricultural products circulate in the market, when a user purchases the agricultural products, the label on the agricultural product packaging bag can be scanned to read out the specific quality index information of the factory-grade one-grade of the agricultural products from the quality grading server, so that the information transparence is realized, and the user is helped to know the quality information of the purchased agricultural products.
Nondestructive testing is an important acquisition way of detection indexes in existing agricultural product assessment, the nondestructive testing is an operation performed before packaging after primary cleaning and screening procedures are performed on agricultural products in an agricultural product processing factory, wherein an acoustic characteristic analyzer refers to the change characteristics of reflection, scattering and absorption, attenuation, propagation speed, acoustic impedance, inherent frequency and the like of the agricultural products under the action of sound waves, and the change characteristics are changed along with the change of internal tissues of the agricultural products so as to reflect the rule of interaction between the sound waves and the agricultural products, therefore, the biological characteristics of the agricultural products can judge the quality of the agricultural products and grade the agricultural products, the sound waves used for detecting the agricultural products are low-energy ultrasonic waves, the change of physical or chemical characteristics of the agricultural products cannot be caused, and the nondestructive testing is realized; the acoustic characteristic analyzer is mainly used for detecting the maturity and the hardness of agricultural products;
the electrical characteristic analyzer is based on the principle that physiological change of agricultural products is accompanied with change of dielectric characteristic parameters, reflects the quality of the agricultural products by utilizing the change of electrical and magnetic characteristics of the agricultural products in an electrical and magnetic field, and mainly detects the density, hardness, freshness, conductivity, water content, internal and external damages of the agricultural products and the like;
the optical characteristic analyzer can have different absorption or reflection characteristics under the irradiation of light rays with different wavelengths, namely the spectral reflectivity or the absorption rate of fruits is larger than that of other parts in a certain specific wavelength, the nondestructive detection of the quality of agricultural products can be realized by combining the characteristic with an optical detection device, and at present, 3 methods are mainly used for the optical detection of the internal quality of fruits: the regular reflection light method, the diffuse reflection light method and the transmission light method mainly comprise 3 methods for optical detection of the internal quality of the fruit at present: regular reflection light method, diffuse reflection light method, transmission light method, etc., mainly detecting nutrient components of agricultural products, such as sugar content, moisture content, etc.;
the X-ray and laser analyzer has better penetrating ability according to X-ray, the laser has better monochromaticity, the intensity after penetrating is detected, the image is converted according to a certain method, and the image is analyzed to achieve the purpose of nondestructive detection, the density of the agricultural product is much smaller than that of substances such as metal, the required X-ray intensity is very weak, generally called soft X-ray, and the soft X-ray can be used for nondestructive detection on the basic physical properties such as surface defect, density, internal lesion and the like and the internal and external quality of the agricultural product, wherein the internal and external quality comprises the chemical residue, the surface defect, the microbial pathogen quantity and the like.
In actual production, one analyzer or a combination of two or more of an acoustic characteristic analyzer, an electrical characteristic analyzer, an optical characteristic analyzer and a laser analyzer can be adopted according to production needs, so that the analyzer can at least detect the density, freshness, conductivity, color, water content, microbial pathogen amount and chemical residual amount of the agricultural products.
After the nondestructive testing is completed, the quality grading server classifies agricultural products according to quality indexes of the detected agricultural products in a grading manner, an agricultural product processing workshop packages the agricultural products after the classification, the quality grading server generates a label for each agricultural product through a label generator, the label is adhered to a packaging bag of the agricultural products, and the label can be a two-dimensional code.
The user scans the label through a user side, such as a mobile phone, and the like, and the outgoing quality grade and the specific grade quality index of the agricultural product can be obtained from the quality grade server.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (10)
1. A quality grading system of agricultural products comprises an agricultural product quality detection device, a quality grading server and a user side, wherein the agricultural product quality detection device is installed in an agricultural product processing workshop and comprises an acoustic characteristic analyzer, an electrical characteristic analyzer, an optical characteristic analyzer, an X-ray and laser analyzer, a two-dimensional code generator and a communication module; the quality grading server comprises an agricultural product database and an agricultural product quality evaluation unit; the agricultural product quality evaluation method is characterized in that the agricultural product quality evaluation unit is used for screening out grading indexes from an agricultural product quality detection device, calculating according to the grading indexes to obtain the grading information, and storing the grading information and the grading indexes in the agricultural product database, and the agricultural product quality evaluation unit comprises a principal component analysis data preprocessing module and an Adaboost classifier training module; the principal component analysis data preprocessing module comprises a parameter standardization unit, an autocorrelation matrix acquisition unit and a principal component number calculation unit; the Adaboost classifier training module comprises a weight initialization unit, a training weak classifier unit and a weak classifier combination strong classifier unit; wherein:
the agricultural product quality detection device is used for acquiring an agricultural product quality evaluation index from an agricultural product nondestructive analyzer;
the parameter standardization unit is used for standardizing the quality evaluation indexes of the obtained agricultural products and eliminating the influence of the quality indexes due to inconsistent dimensions and units;
the autocorrelation matrix obtaining unit is used for establishing an autocorrelation matrix of the quality index of the agricultural product and obtaining the eigenvalue and the eigenvector of the quality index of the agricultural product according to the autocorrelation matrix;
the main component number calculating unit is used for calculating the variance contribution rate and the accumulated variance contribution rate of each agricultural product quality index and determining the number of main components of the agricultural product quality index according to the accumulated variance contribution rate;
the weight initialization unit is used for initializing the weight of each sample;
the training weak classifier unit is used for updating the weights of all training samples;
the weak classifier combination strong classifier unit is used for increasing the weight of a weak classifier with small classification error rate, reducing the weight of a weak classifier with large classification error rate and combining the weak classifiers into a strong classifier.
2. A quality grading method of agricultural products is characterized by comprising the steps of determining quality evaluation indexes of the agricultural products, constructing a principal component-Adaboost combined model, and obtaining principal component matrixes of the quality evaluation indexes in the model according to principal component analysis; training a principal component matrix of a quality evaluation index according to an Adaboost algorithm, and inputting the agricultural product to be evaluated into a principal component-Adaboost combined model after training to obtain a quality evaluation result of the agricultural product; wherein the quality evaluation indexes of the agricultural products at least comprise density, freshness, conductivity, color and luster, water content, microbial pathogen amount and chemical residual amount of the agricultural products.
3. The method of claim 1, wherein the obtaining of the principal component matrix of the quality assessment index from the principal component analysis comprises:
s101, preprocessing data of the quality evaluation indexes of the agricultural products;
s102, establishing an autocorrelation matrix of the quality index of the agricultural product, and calculating and acquiring a characteristic value and a characteristic vector of the quality index of the agricultural product according to the autocorrelation matrix;
s103, calculating the variance contribution rate and the accumulated variance contribution rate, selecting an index with the accumulated variance contribution rate exceeding 85% in the quality indexes of the agricultural products, and counting the number of the indexes with the accumulated variance contribution rate exceeding 85%;
and S104, calculating a principal component matrix of the quality index of the agricultural product according to the selected characteristic vector of the quality index of the agricultural product.
4. The quality grading method for agricultural products of claim 2, wherein the data preprocessing of the quality assessment indexes of agricultural products comprises:
S101A, the agricultural product quality evaluation indexes comprise n samples, each sample comprises m quality evaluation indexes, each sample is taken as a row, each quality evaluation index is taken as a column, and an n x m-order agricultural product quality evaluation index matrix is formed;
S101B, carrying out standardization processing on each index in the matrix to obtain a standardized matrix; the normalization process is represented as:
wherein, x'ijAn element representing the ith row and the jth column in the normalized matrix; x is the number ofijThe element which represents the ith row and the jth column in the quality evaluation index matrix of the agricultural product;representing the average value of j-th row indexes in the quality evaluation index matrix of the agricultural products; sjAnd the variance of the jth column index in the agricultural product quality evaluation index matrix is represented.
5. The quality grading method for agricultural products according to claim 2, wherein the step of establishing an autocorrelation matrix of the quality index of the agricultural product, the step of obtaining the eigenvalue and the eigenvector of the quality index of the agricultural product by calculation according to the autocorrelation matrix comprises the steps of constructing an eigen polynomial of the autocorrelation matrix, making the eigen polynomial of the autocorrelation matrix equal to 0, solving the eigenvalue, and bringing the eigenvalues one by one, wherein the solution system is the eigenvector, and the autocorrelation matrix of the quality evaluation index matrix of the agricultural product is expressed as:
S＝R'TR'/(N-1)；
s is an autocorrelation matrix of an agricultural product quality evaluation index matrix; r' is a standardized matrix generated after the sample is subjected to standardized processing; n is the number of samples; the superscript T represents the transpose of the matrix.
6. The method for grading the quality of agricultural products according to claim 1, wherein the training of the principal component matrix of the quality assessment index by inputting the principal component matrix into an Adaboost algorithm model comprises:
s201, inputting training samples into an Adaboost algorithm model, and setting the initial weight of each sample to be 1/n; wherein the training samples comprise n samples, wherein the ith sample is represented as (x)i,yi)，xiIs the feature vector of the i-th training sample, yiThe evaluation result of the ith sample;
s202, setting the maximum iteration times of an Adaboost algorithm model, and performing k rounds of training on the quality indexes of the agricultural products to obtain k basic classifiers;
s203, in the training process, the sample relearning weight of the basic classifier is optimized by using the fuzzy rule corresponding to the fitness, and the relearning weight of the current basic trainer is used as the weight of the sample input into the next basic trainer;
s204, combining all the basic classifiers into a strong classifier, and inputting the quality index data of the agricultural products to be tested to obtain an evaluation result; wherein combining all basic classifiers into a strong classifier is represented as:
H(x)＝sign(sum(αih(xi))),i∈{1,2,...,k}；
wherein H (x) is a strong classifier; h (x)i) Is the output of the ith basic classifier; alpha is alphaiIs the weight of the ith basic classifier.
7. The quality grading method for agricultural products according to claim 5, wherein optimizing the obtained weight of the basic classifier by using the fuzzy rule corresponding to the fitness comprises:
S203A, calculating the fitness of the sample, and screening out the sample with the maximum fitness and the fuzzy rule corresponding to the sample;
S203B, calculating a classification error rate and a classification weight according to the fuzzy rule when the fitness is maximum;
S203C, updating the optimal weight of the classifier according to the classification error rate and the classification weight.
8. A quality grading method for agricultural products according to claim 6, characterized in that said fitness isExpressed as:
wherein,classify the weak classifiers with the correct rate expressed asωkFor the weight of each sample after training,for the k sample in the fuzzy rule betatDegree of membership, x, of the fuzzy set to which the pair belongskA k training sample in the characteristic sample set is obtained;classify the weak classifiers with an error rate expressed askmaxClassifying an error rate threshold for the weak classifier; k | yk＝ytRepresents the training result y of the k-th sample among the n training sampleskAnd quality evaluation result y in the feature sampletEqual; k | yk≠ytRepresents the training result y of the k-th sample among the n training sampleskAnd quality evaluation result y in the feature sampletNot equal.
9. The quality grading method for agricultural products of claim 6, wherein calculating the classification error rate and the classification weight comprises:
wherein epsilon (. beta.)t) For blurring the rule betatThe classification error rate of (2); thetatIs the weight of the weak classifier; omegaiIs the weight of the sample i and,for the k sample in the fuzzy rule betatDegree of membership, x, of the fuzzy set to which the pair belongsiIs the ith training sample; i | yi≠ytRepresents the training result y of the k-th sample among the n training samplesiAnd quality evaluation result y in the feature sampletNot equal.
10. The quality grading method for agricultural products according to claim 6, wherein the updating the optimal weight of the classifier according to the classification error rate and the classification weight comprises:
wherein, ω isi(t +1) is a weight value used for training the classifier sample obtained according to the error rate of classification during the t +1 th iteration; alpha is alphatIs a normalization factor; thetatIs the weight of the weak classifier;for the k sample in the fuzzy rule betatThe degree of membership of the fuzzy set to which the pair belongs.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910712867.2A CN110517160A (en) | 2019-08-02 | 2019-08-02 | A kind of quality grading method and quality grading system of agricultural product |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910712867.2A CN110517160A (en) | 2019-08-02 | 2019-08-02 | A kind of quality grading method and quality grading system of agricultural product |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110517160A true CN110517160A (en) | 2019-11-29 |
Family
ID=68624862
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910712867.2A Pending CN110517160A (en) | 2019-08-02 | 2019-08-02 | A kind of quality grading method and quality grading system of agricultural product |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110517160A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112201355A (en) * | 2020-11-04 | 2021-01-08 | 成都东方天呈智能科技有限公司 | Construction method of health assessment iterative classifier model |
| CN112966541A (en) * | 2020-09-23 | 2021-06-15 | 北京豆牛网络科技有限公司 | Automatic fruit and vegetable goods inspection method and system, electronic equipment and computer readable medium |
| CN113435967A (en) * | 2021-06-22 | 2021-09-24 | 布瑞克农业大数据科技集团有限公司 | Method and system for automatically determining marketing content of agricultural products |
| CN113705947A (en) * | 2020-10-28 | 2021-11-26 | 四川沁链云科科技有限公司 | Agricultural product rating system based on block chain technology |
| CN116664169A (en) * | 2023-08-01 | 2023-08-29 | 贵州中昂科技有限公司 | Online agricultural product selling method and system |
| CN116976917A (en) * | 2023-07-31 | 2023-10-31 | 金景(海南)科技发展有限公司 | Agricultural brand certificate system construction method based on blockchain technology |
| CN117726359A (en) * | 2024-02-08 | 2024-03-19 | 成都纳宝科技有限公司 | Interactive marketing method, system and equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102654494A (en) * | 2012-04-16 | 2012-09-05 | 昆明理工大学 | Method for establishing quality identification and detection standard for agricultural products |
| CN103646251A (en) * | 2013-09-14 | 2014-03-19 | 江南大学 | Apple postharvest field classification detection method and system based on embedded technology |
| CN103955626A (en) * | 2014-05-21 | 2014-07-30 | 江苏省农业科学院 | Dried edamame quality evaluation model and construction method |
| CN104021396A (en) * | 2014-06-23 | 2014-09-03 | 哈尔滨工业大学 | Hyperspectral remote sensing data classification method based on ensemble learning |
| CN108985336A (en) * | 2018-06-19 | 2018-12-11 | 三峡大学 | A kind of fruit quality classification method based on multi-layer SVM |
-
2019
- 2019-08-02 CN CN201910712867.2A patent/CN110517160A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102654494A (en) * | 2012-04-16 | 2012-09-05 | 昆明理工大学 | Method for establishing quality identification and detection standard for agricultural products |
| CN103646251A (en) * | 2013-09-14 | 2014-03-19 | 江南大学 | Apple postharvest field classification detection method and system based on embedded technology |
| CN103955626A (en) * | 2014-05-21 | 2014-07-30 | 江苏省农业科学院 | Dried edamame quality evaluation model and construction method |
| CN104021396A (en) * | 2014-06-23 | 2014-09-03 | 哈尔滨工业大学 | Hyperspectral remote sensing data classification method based on ensemble learning |
| CN108985336A (en) * | 2018-06-19 | 2018-12-11 | 三峡大学 | A kind of fruit quality classification method based on multi-layer SVM |
Non-Patent Citations (6)
| Title |
|---|
| 侯升飞: "基于图像处理与光谱技术的大豆等级识别方法的研究", 《中国优秀博硕士学位论文全文数据库(硕士)农业科技辑》 * |
| 孙熙，李夏苗: "基于boosting 算法的交通事件检测", 《交通运输系统工程与信息》 * |
| 孟永宏等: "基于主成分分析法的美八苹果品质综合评价体系构建", 《食品工业科技》 * |
| 曾乐根: "基于机器学习的4G网络工程质量评估", 《电脑知识与技术》 * |
| 樊丁宇等: "新疆杏品种果实鲜食品质主要评价指标的选择", 《中国农学通报》 * |
| 王南南: "基于高光谱成像技术的多指标综合决策香蕉品质等级研究", 《中国优秀硕士学位论文全文数据库工程科技Ⅰ辑(月刊)》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112966541A (en) * | 2020-09-23 | 2021-06-15 | 北京豆牛网络科技有限公司 | Automatic fruit and vegetable goods inspection method and system, electronic equipment and computer readable medium |
| CN112966541B (en) * | 2020-09-23 | 2023-12-05 | 北京豆牛网络科技有限公司 | Fruit and vegetable automatic checking method, system, electronic equipment and computer readable medium |
| CN113705947A (en) * | 2020-10-28 | 2021-11-26 | 四川沁链云科科技有限公司 | Agricultural product rating system based on block chain technology |
| CN112201355A (en) * | 2020-11-04 | 2021-01-08 | 成都东方天呈智能科技有限公司 | Construction method of health assessment iterative classifier model |
| CN112201355B (en) * | 2020-11-04 | 2024-01-16 | 成都东方天呈智能科技有限公司 | Construction method of health evaluation iterative classifier model |
| CN113435967A (en) * | 2021-06-22 | 2021-09-24 | 布瑞克农业大数据科技集团有限公司 | Method and system for automatically determining marketing content of agricultural products |
| CN113435967B (en) * | 2021-06-22 | 2022-04-12 | 布瑞克农业大数据科技集团有限公司 | Method and system for automatically determining marketing content of agricultural products |
| CN116976917A (en) * | 2023-07-31 | 2023-10-31 | 金景(海南)科技发展有限公司 | Agricultural brand certificate system construction method based on blockchain technology |
| CN116976917B (en) * | 2023-07-31 | 2024-05-24 | 金景(海南)科技发展有限公司 | Agricultural brand certificate system construction method based on blockchain technology |
| CN116664169A (en) * | 2023-08-01 | 2023-08-29 | 贵州中昂科技有限公司 | Online agricultural product selling method and system |
| CN117726359A (en) * | 2024-02-08 | 2024-03-19 | 成都纳宝科技有限公司 | Interactive marketing method, system and equipment |
| CN117726359B (en) * | 2024-02-08 | 2024-04-26 | 成都纳宝科技有限公司 | Interactive marketing method, system and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110517160A (en) | A kind of quality grading method and quality grading system of agricultural product | |
| Raikar et al. | Classification and Grading of Okra-ladies finger using Deep Learning | |
| Bhatt et al. | Automatic apple grading model development based on back propagation neural network and machine vision, and its performance evaluation | |
| Tian et al. | Early detection of freezing damage in oranges by online Vis/NIR transmission coupled with diameter correction method and deep 1D-CNN | |
| Sabzi et al. | Non-destructive estimation of physicochemical properties and detection of ripeness level of apples using machine vision | |
| Iraji | Comparison between soft computing methods for tomato quality grading using machine vision | |
| Benmouna et al. | Convolutional neural networks for estimating the ripening state of fuji apples using visible and near-infrared spectroscopy | |
| US20230048665A1 (en) | Two-stage frequency selection method and device for microwave frequency sweep data | |
| CN115908257A (en) | Defect recognition model training method and fruit and vegetable defect recognition method | |
| Saleh et al. | Palm oil classification using deep learning | |
| Devi et al. | IoT-deep learning based prediction of amount of pesticides and diseases in fruits | |
| Mohamadzadeh Moghadam et al. | Nondestructive classification of saffron using color and textural analysis | |
| CN117292494A (en) | Signal identification method, system, computer equipment and medium for sound and vibration fusion | |
| Tian et al. | Detection of soluble solid content in apples based on hyperspectral technology combined with deep learning algorithm | |
| Tian et al. | Research on apple origin classification based on variable iterative space shrinkage approach with stepwise regression–support vector machine algorithm and visible‐near infrared hyperspectral imaging | |
| Goyal et al. | Tomato ripeness and shelf-life prediction system using machine learning | |
| CN110263969B (en) | Dynamic prediction system and prediction method for quality of apples with shelf life | |
| Zhang et al. | Non-destructive evaluation of the edible rate for pomelo using X-ray imaging method | |
| Masuda et al. | Application of deep learning diagnosis for multiple traits sorting in peach fruit | |
| Jian et al. | Predicting bruise susceptibility in apples using Vis/SWNIR technique combined with ensemble learning | |
| Okere et al. | Vis-NIR and SWIR hyperspectral imaging method to detect bruises in pomegranate fruit | |
| Wu et al. | Moisture measurement of tea leaves during withering using multifrequency microwave signals optimized by ant colony optimization | |
| Gao et al. | Mass detection of walnut based on X‐ray imaging technology | |
| Fisher et al. | An image processing and machine learning solution to automate Egyptian cotton lint grading | |
| Yu et al. | Nondestructive testing of external defects on Nanguo pear |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191129 |