CN115615987B - Fruit harvesting and classifying method integrating appearance quality and internal quality detection - Google Patents
Fruit harvesting and classifying method integrating appearance quality and internal quality detection Download PDFInfo
- Publication number
- CN115615987B CN115615987B CN202211172555.5A CN202211172555A CN115615987B CN 115615987 B CN115615987 B CN 115615987B CN 202211172555 A CN202211172555 A CN 202211172555A CN 115615987 B CN115615987 B CN 115615987B
- Authority
- CN
- China
- Prior art keywords
- decision
- fruit
- harvesting
- index
- evaluation
- 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.)
- Active
Links
- 238000003306 harvesting Methods 0.000 title claims abstract description 82
- 235000013399 edible fruits Nutrition 0.000 title claims abstract description 65
- 238000001514 detection method Methods 0.000 title claims abstract description 24
- 238000000034 method Methods 0.000 title claims abstract description 17
- 238000011156 evaluation Methods 0.000 claims abstract description 38
- 230000014509 gene expression Effects 0.000 claims description 28
- 239000013598 vector Substances 0.000 claims description 8
- 238000012804 iterative process Methods 0.000 claims description 7
- 239000011159 matrix material Substances 0.000 claims description 6
- 241000238631 Hexapoda Species 0.000 claims description 5
- 239000007787 solid Substances 0.000 claims description 5
- 241000607479 Yersinia pestis Species 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 3
- 230000006378 damage Effects 0.000 claims description 2
- 208000027418 Wounds and injury Diseases 0.000 claims 1
- 230000008014 freezing Effects 0.000 claims 1
- 238000007710 freezing Methods 0.000 claims 1
- 208000014674 injury Diseases 0.000 claims 1
- 239000002420 orchard Substances 0.000 abstract description 2
- 235000005976 Citrus sinensis Nutrition 0.000 description 15
- 240000002319 Citrus sinensis Species 0.000 description 15
- 235000012055 fruits and vegetables Nutrition 0.000 description 6
- 238000003860 storage Methods 0.000 description 6
- 230000009746 freeze damage Effects 0.000 description 4
- 238000004806 packaging method and process Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 238000005481 NMR spectroscopy Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000000701 chemical imaging Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000005389 magnetism Effects 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 229940088594 vitamin Drugs 0.000 description 1
- 229930003231 vitamin Natural products 0.000 description 1
- 235000013343 vitamin Nutrition 0.000 description 1
- 239000011782 vitamin Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N2021/8466—Investigation of vegetal material, e.g. leaves, plants, fruits
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8854—Grading and classifying of flaws
- G01N2021/8861—Determining coordinates of flaws
Landscapes
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Sorting Of Articles (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a fruit harvesting and classifying method integrating appearance quality and internal quality detection, which considers that the conditions of different time sections, harvesting equipment, orchard management and the like of fruits are different in actual conditions when harvesting and classifying, and can cause variation conditions of indexes in a decision scheme of harvesting and classifying, so that different decision schemes of fruit harvesting and classifying are optimally subjected to fuzzy decision recognition by aiming at the related indexes of the appearance quality and the internal quality, and the evaluation conditions corresponding to the decision schemes of fruit harvesting and classifying and the evaluation results for determining the fruit harvesting and classifying are given to guide fruit harvesting and classifying work.
Description
Technical Field
The invention belongs to the technical field of fruit postharvest treatment, and particularly relates to a method for harvesting and classifying fruits integrating appearance quality and internal quality detection based on multi-objective multi-level system multi-dimensional fuzzy decision.
Background
The harvesting, grading, packaging and transportation of fruits and vegetables are the work to be done before entering into formal storage or marketing, and the commercial treatment after harvesting and harvesting mainly has two effects on fruits. On one hand, the quality and the storability of fruits and vegetables are affected, such as proper-period harvesting, grading and selecting after harvesting, reasonable packaging, pest and disease control after harvesting, timely precooling after harvesting, reasonable stacking and the like, can have a certain influence on the storage and transportation of fruits and vegetables, and on the other hand, the commodity property of fruits and vegetables is affected. In the commodity circulation field, the uniformity of size and specification is required. The appearance is beautiful, and the package is exquisite and proper. Ensure shelf time. This requires grading and packaging of the fruit according to certain criteria after harvesting. Some fruits have beautiful appearance and prolonged shelf life, and are also required to be cleaned and waxed to improve commodity properties of the fruits. Harvesting is the last link of fruit and vegetable production and also the beginning of post-harvest treatment. The harvesting maturity of fruits and vegetables has close relation with the yield, quality and storage characteristics. The judgment of the maturity in production is generally comprehensively considered according to factors such as different varieties, biological characteristics, growth conditions, climate conditions, cultivation management and the like. At the same time, the suitable harvest time is determined from the aspects of regulating market supply, storage, transportation and processing requirements, labor arrangement and the like.
Fruit quality mainly includes external quality and internal quality. External qualities include fruit size, shape, color, odor, etc. The internal quality includes hardness, soluble solids, sugar degree, acidity, vitamins and other components, and the internal browning, freeze injury, insect damage, hollowness and other indexes of the fruit are invisible. The nondestructive detection of fruit quality refers to a detection process for analyzing and acquiring fruit quality by utilizing physical characteristics of fruits such as sound, electricity, light, magnetism and the like and comprehensive detection technologies such as spectral imaging, dielectric characteristics, nuclear magnetic resonance and the like on the premise of not damaging a detection object. In recent years, nondestructive detection technology of fruit quality is mature and perfect, a detection instrument is developed towards portability and intellectualization, detection indexes are changed from single indexes to multiple indexes for simultaneous detection, and accuracy, reliability and timeliness of detection results are gradually improved, so that the nondestructive detection instrument plays a greater promotion role in fruit industry industrialization.
Disclosure of Invention
The invention aims at carrying out optimal fuzzy decision recognition on different decision schemes of fruit harvest classification aiming at the related indexes of the appearance quality and the internal quality of the set, thereby providing the evaluation grade corresponding to the decision scheme of fruit harvest classification to guide fruit harvest classification work.
In order to achieve the above object, the present invention includes the following:
Let the fruit harvest classification decision set be Γ= { η j |1.ltoreq.j.ltoreq.j }, where η j is the J-th fruit harvest classification decision scheme integrating appearance quality and internal quality detection, J is the number of decision schemes, and is defined by C 1 (size), C 2 (shape), C 3 (colour), C 4 (odour), C 4 (hardness), C 5 (soluble solids), C 6 (sugar degree), C 7 (acidity), C 8 (internal browning), C 9 (freeze injury), C 10 (insect pest), C 11 (hollow) and the like to construct an index set c= { C 1,...,C11 } of the decision set Γ, then dividing C 1,...,C4 into appearance quality B 1, Dividing C 5,...,C11 into internal qualities B 2, thereby constructing a gatepost set b= { B 1,B2 }, from the appearance quality B 1 and the internal quality B 2; Because the actual conditions are different when harvesting and classifying are carried out according to the conditions of different time sections, harvesting equipment, orchard management and the like of fruits, the variation condition exists in the index in the decision scheme of the harvesting and classifying, so that different decision schemes of the fruit harvesting and classifying can be divided into H=4 different evaluation grades as decision evaluation sets D in the use process according to the actual demands of sales, storage, processing and the like, and the decision sets D can be expressed as:
d= { D 1 (preferred use), D 2 (recommended use), D 3 (barely used), D 4 (not recommended) } (1)
The data characteristic value matrix of the fruit harvest classifying index corresponding to the category k decision j index I can be expressed as (kxi,j)I×J, wherein 1 is less than or equal to 2,1 is less than or equal to 11, kxi,j is the characteristic value of the category k decision j index I, kri,j is the normalized value of the characteristic value kxi,j of the category k decision j index I, ksi,h is the target value of the standard decision mode of the category k evaluation grade h, wherein 1 is less than or equal to 4, kuh,j is the relative membership degree of the characteristic value of the evaluation grade h to which the decision j belongs and kwi,j Decision of target weight of j index i for class k and/>The method can establish a system comprising Lagrangian constant/>, according to a multi-level system multi-dimensional fuzzy decision theory, for obtaining optimal fuzzy decision recognition under different scenes corresponding to fruit harvesting classification decision schemes integrating appearance quality and internal quality detectionAnd a minimization objective function of xi/>
For a pair ofThe first and second derivatives are obtained with respect to kuh,j,kwi,j:
When (when) When in use, it is obvious thatIt can be seen that kuh,j,kwi,j is present to makeIf there is a minimum, the objective function/>, is minimizedThere is a solution.
kuh,j Available from formula (3) as 0Is represented by the expression:
From (7) and Availability/>Is represented by the expression:
Then the theoretical value expression of kuh,j can be obtained from formulas (7) and (8):
The expression kwi,j for ζ is obtainable from equation (4) being 0:
From (9) and Availability/>Is represented by the expression:
Then the theoretical value expression of kwi,j can be obtained from formulas (9) and (10):
Theoretical value expressions of kuh,j and kwi,j can be substituted into the iterative process shown in fig. 1 to give actual values ku'h,j and kw'i,j corresponding to kuh,j and kwi,j, and to give fruit harvest classification decision schemes for preferential, recommended and marginal use according to the maximum membership principle to give an evaluation result of fruit harvest classification:
Step 1: kxi,j、kri,j and ksi,h can be given out through training set data corresponding to different indexes in the fruit harvesting and classifying process and historical data of a harvesting and classifying decision scheme in the initializing process, if the historical data of the harvesting and classifying decision scheme is not available, all indexes including appearance quality 2C1 and internal quality 2C2 of the harvesting and classifying decision scheme are set, the target weight kwi,j (t) of the index I of the category k decision j is 1/I, and the iteration times t is set to be 0;
step 2: the iteration times t are replaced by t+1, kxi,j、kri,j、ksi,h and kwi,j (t-1) are substituted into a theoretical value expression of kuh,j so as to output kuh,j (t);
Step 3: substituting kxi,j、kri,j、ksi,h and kuh,j (t) into the theoretical value expression of kwi,j to output kwi,j (t);
Step 4: if both conditions | [ kuh,j(t)-kuh,j(t-1)]/kuh,j(t)|≤δu and | [ kwi,j(t)-kwi,j(t-1)]/kwi,j(t)|≤δw ] are satisfied, wherein δ u and δ w are iterative process change thresholds of kuh,j and kwi,j, respectively, kuh,j (t) and kwi,j (t) are output as actual values ku'h,j and kw'i,j, thereby constructing (ku'h,j)H×J and kw'i,j)I×J, otherwise step 2 is entered;
Step 5: according to the decision evaluation set D= { D 1,D2,D3,D4 }, a multi-dimensional fuzzy decision matrix of the category k (ku'h,j)H×J can obtain a fruit harvesting classification decision set E h={eh=j|ku'h,j≥εh corresponding to the evaluation level h, wherein E h is an element in the fruit harvesting classification decision set E h, epsilon h is a decision threshold of the evaluation level h, and then the corresponding fruit harvesting classification decision can be selected to be used according to the order of priority to barely using of the evaluation level h;
Step 6: if the fruit harvest classification decision given in step 5 is j ', a weight vector of appearance quality is constructed from the target weights kwi,j′ of the corresponding category k decision j ' index i (1wi,j′)1×4 and the weight vector of internal quality (2wi,j′)1×7, and the normalized values of the eigenvalues kxi,j′ of the category k decision j ' index i are counted to establish (1ri,j′)1×4 and (2ri,j′)1×7, the evaluation result for determining fruit harvest classification can be expressed as:
Wherein, Using the M (+, ·) calculation mode, ρ is the quality weight.
Drawings
Fig. 1: fruit harvesting and classifying iterative process diagram integrating appearance quality and internal quality detection
Fig. 2: normalized value chart for appearance quality test of Newhol navel orange fruit
Fig. 3: normalized value chart for internal quality test of Newhol navel orange fruit
Fig. 4: neohol navel orange harvesting classification evaluation result graph
Detailed Description
The following will describe in detail the technical solution provided by the present invention in connection with a specific embodiment of fruit harvesting classification integrating appearance quality and internal quality detection by navel orange of new-holl variety, specifically including the following:
Let the Neohol navel orange harvesting classification decision set be Γ= { eta j |1.ltoreq.j.ltoreq.J }, wherein eta j is the J-th Neohol navel orange harvesting classification decision scheme integrating appearance quality and internal quality detection, the number of the past decision schemes is provided with J=10, and the number is provided with C 1 (size), C 2 (shape), C 3 (colour), C 4 (odour), C 4 (hardness), C 5 (soluble solids), C 6 (sugar degree), C 7 (acidity), C 8 (internal browning), C 9 (freeze injury), C 10 (insect pest), C 11 (hollow) and the like to construct an index set c= { C 1,...,C11 } of the decision set Γ, then dividing C 1,...,C4 into appearance quality B 1, Dividing C 5,...,C11 into internal qualities B 2, thereby constructing a gatepost set b= { B 1,B2 }, from the appearance quality B 1 and the internal quality B 2; Different decision schemes for harvesting and classifying the neoil navel orange according to actual demands such as sales, storage and processing can be divided into H=4 different evaluation grades in the use process to be used as a decision evaluation set D, and the decision evaluation set D can be expressed as:
d= { D 1 (preferred use), D 2 (recommended use), D 3 (barely used), D 4 (not recommended) }
The data characteristic value matrix of the Newhall navel orange harvesting classification index corresponding to the category k decision j index I can be expressed as (kxi,j)I×J), wherein 1 is less than or equal to 2,1 is less than or equal to 11, kxi,j is the characteristic value of the category k decision j index I, kri,j is the normalized value of the characteristic value kxi,j of the category k decision j index I, ksi,h is the target value of the standard decision mode of the category k evaluation grade h, wherein 1 is less than or equal to 4, kuh,j is the relative membership degree of the characteristic value of the evaluation grade h to which the decision j belongs and kwi,j Decision of target weight of j index i for class k and/>The optimal fuzzy decision recognition under different scenes corresponding to the Newhall navel orange harvesting classification decision scheme integrating appearance quality and internal quality detection can be established to comprise Lagrangian constant/>, according to a multi-level system multi-dimensional fuzzy decision theoryAnd a minimization objective function of xi/>
For a pair ofThe first and second derivatives are obtained with respect to kuh,j,kwi,j:
When (when) When in use, it is obvious thatIt can be seen that kuh,j,kwi,j is present to makeIf there is a minimum, the objective function/>, is minimizedThere is a solution.
From the following componentsFirst derivative expression and/>, with respect to kuh,j The theoretical value expression of kuh,j can be obtained:
From the following components First derivative expression and/>, with respect to kwi,j The theoretical value expression of kuh,j can be obtained:
Theoretical value expressions kuh,j and kwi,j can be substituted into an iterative process as shown in fig. 1 to give actual values ku'h,j and kw'i,j corresponding to kuh,j and kwi,j, and a priority, recommended and marginal use newel navel orange harvesting classification decision scheme can be given according to the maximum membership rule, so that the evaluation result of newel navel orange harvesting classification is given:
Step 1: in the initialization process, the iteration times t are set to be 0, and kxi,j、kri,j、ksi,h and kwi,j (t=0) can be given out through training set data corresponding to different indexes and historical data of a harvesting classification decision scheme in the Newhol navel orange harvesting classification process;
step 2: the iteration times t are replaced by t+1, kxi,j、kri,j、ksi,h and kwi,j (t-1) are substituted into a theoretical value expression of kuh,j so as to output kuh,j (t);
Step 3: substituting kxi,j、kri,j、ksi,h and kuh,j (t) into the theoretical value expression of kwi,j to output kwi,j (t);
Step 4: if both conditions [ kuh,j(t)-kuh,j(t-1)]/kuh,j(t)|≤δu and [ kwi,j(t)-kwi,j(t-1)]/kwi,j(t)|≤δw ] are satisfied, wherein both iteration process change thresholds δ u and δ w of kuh,j and kwi,j are set to 1%, kuh,j (t) and kwi,j (t) are output as actual values ku'h,j and kw'i,j, thereby constructing (ku'h,j)H×J and kw'i,j)I×J, otherwise, step 2 is entered;
Step 5: according to the decision evaluation set D= { D 1,D2,D3,D4 }, a multi-dimensional fuzzy decision matrix of the category k (ku'h,j)H×J can obtain a fruit harvesting classification decision set E h={eh=j|ku'h,j≥εh corresponding to the evaluation level h, wherein E h is an element in the fruit harvesting classification decision set E h, the decision threshold epsilon h of the evaluation level h is set to be 0.9, and then the corresponding fruit harvesting classification decision can be selected to be used according to the evaluation level h from priority to marginal use;
Step 6: if the fruit harvest classification decision given in step 5 is j ', a weight vector of appearance quality is constructed from the target weights kwi,j′ of the index i corresponding to the category k decision j ' (1wi,j′)1×4 and the weight vector of internal quality (2wi,j′)1×7, and the normalized values of the eigenvalues kxi,j′ of the index i of the category k decision j ' are counted to establish (1ri,j′)1×4 and (2ri,j′)1×7, the evaluation result of the fruit harvest classification is determined after setting the quality weight ρ=0.36 can be expressed as:
Wherein, The M (+,) computation mode was used.
In fig. 2 and 3, normalized values of appearance quality and internal quality of 30 new york navel orange fruits after being tested are shown, from which evaluation results of the new york navel orange harvest classification are shown in fig. 4, wherein the evaluation results of the new york navel orange harvest classification are comprehensively considered, such as appearance quality and internal quality, and the weight vectors of the appearance quality and internal quality are effectively constructed (1wi,j′)1×4 and the weight vector of the internal quality (2wi,j′)1×7 normalized values are constructed) according to the specific embodiment, such as C 1 (size), C 2 (shape), C 3 (color), C 4 (smell), C 4 (hardness), C 5 (soluble solid), C 6 (sugar degree), C 7 (acidity), C 8 (internal browning), C 9 (freeze injury), C 10 (insect attack), C 11 (hollow) and the corresponding target weight kwi,j′.
Claims (2)
1. A fruit harvesting and sorting method integrating appearance quality and internal quality detection, which is characterized by comprising the following steps:
a) Let the fruit harvest classification decision set be Γ= { η j |1.ltoreq.j.ltoreq.j }, where η j is the J-th fruit harvest classification decision scheme integrating appearance quality and internal quality detection, J is the number of decision schemes, and the index size C 1, Shape C 2, color C 3, odor C 4, hardness C 4, soluble solids C 5, Sugar degree C 6, acidity C 7, internal browning C 8, freezing injury C 9, insect pest C 10, The hollow C 11 constructs an index set C= { C 1,...,C11 } of the fruit harvest classification decision set Γ, a category set B= { B 1,B2 } is constructed by the appearance quality B 1 and the internal quality B 2, let the different decision schemes of fruit harvest classification be divided into h=4 different evaluation grades during use as decision evaluation set D, expressed as:
D={D1,D2,D3,D4},
Wherein D 1 represents priority use, D 2 represents recommended use, D 3 represents marginal use, and D 4 represents no recommendation;
Let the data characteristic value matrix of the fruit harvest classifying index corresponding to the category k decision j index I be expressed as (kxi,j)I×J, wherein, 1 is less than or equal to k is less than or equal to 2,1 is less than or equal to I is less than or equal to 11, kxi,j is the characteristic value of the category k decision j index I, kri,j is the normalized value of the characteristic value kxi,j of the category k decision j index I, ksi,h is the target value of the standard decision mode of the category k evaluation grade h, wherein, 1 is less than or equal to h is less than or equal to 4, kuh,j is the relative membership degree of the characteristic value of the decision j belonging to the evaluation grade h and kwi,j Decision of target weight of j index i for class k and/>Establishing a system containing Lagrangian constant/>, according to a multi-level system multi-dimensional fuzzy decision theory, for optimal fuzzy decision recognition under different scenes corresponding to fruit harvesting classification decision schemes integrating appearance quality and internal quality detectionAnd a minimization objective function of xi/>
B) When (when)When the first derivative with respect to kuh,j,kwi,j is 0,/>With respect to the second derivative of kuh,j, kwi,j being greater than 0, kuh,j,kwi,j exists such that/>If there is a minimum, the objective function/>, is minimizedA solution;
c) From the following components The first derivative with respect to kuh,j is kuh,j with respect to/>Expression of/>To yield the theoretical value expression of kuh,j:
Wherein t and l are intermediate variables;
d) From the following components The first derivative with respect to kwi,j yields kwi,j expression with respect to ζ and/>To yield the theoretical value expression of kwi,j:
Wherein m and d are intermediate variables;
e) Substituting the theoretical value expressions kuh,j and kwi,j into an iterative process to give actual values ku'h,j and kw′i,j corresponding to kuh,j and kwi,j, and giving fruit harvest classification decision schemes which are preferentially used, recommended to be used and barely used according to the maximum membership principle so as to give evaluation results of fruit harvest classifications;
b) In, The first derivatives with respect to kuh,j,kwi,j are expressed as:
b) In, The second derivatives with respect to kuh,j,kwi,j are expressed as:
c) In kuh,j about Expression of/>The expressions of (2) are expressed as:
wherein m and c are intermediate variables;
d) In kwi,j, expression and expression for ζ The expressions of (2) are expressed as:
2. the fruit harvesting classification method integrating appearance quality and internal quality detection as claimed in claim 1, wherein: e) In the above, the iterative process is described as:
step 1: kxi,j、kri,j and ksi,h are given out through training set data corresponding to different indexes in the fruit harvesting and classifying process and historical data of a harvesting and classifying decision scheme in the initializing process, and if no historical data is used for setting the harvesting and classifying decision scheme to comprise all indexes of appearance quality B 1 and internal quality B 2, and the target weight kwi,j (t) of the index I of the gate k decision j is 1/I, the iteration times t is set to be 0;
step 2: the iteration times t are replaced by t+1, kxi,j、kri,j、ksi,h and kwi,j (t-1) are substituted into a theoretical value expression of kuh,j so as to output kuh,j (t);
Step 3: substituting kxi,j、kri,j、ksi,h and kuh,j (t) into the theoretical value expression of kwi,j to output kwi,j (t);
step 4: if both conditions | [ kuh,j(t)-kuh,j(t-1)]/kuh,j(t)|≤δu and | [ kwi,j(t)-kwi,j(t-1)]/kwi,j(t)|≤δw ] are satisfied, wherein δ u and δ w are iterative process change thresholds of kuh,j and kwi,j, respectively, kuh,j (t) and kwi,j (t) are output as actual values ku'h,j and kw′i,j, thereby constructing (ku'h,j)H×J and kw′i,j)I×J, otherwise step 2 is entered;
Step 5: according to the decision evaluation set D= { D 1,D2,D3,D4 }, obtaining a fruit harvesting classification decision set E h={eh=j|ku'h,j≥εh corresponding to the evaluation level h by a multidimensional fuzzy decision matrix (ku'h,j)H×J), wherein E h is an element in the fruit harvesting classification decision set E h corresponding to the evaluation level h, epsilon h is a decision threshold of the evaluation level h, and then selecting to use the corresponding fruit harvesting classification decision according to the evaluation level h by a sequence from priority to marginal use;
Step 6: if the fruit harvest classification decision given in step 5 is j ', a weight vector of appearance quality is constructed from the target weights kwi,j′ of the corresponding category k decision j ' index i (1wi,j′)1×4 and the weight vector of internal quality (2wi,j′)1×7, and the normalized values of the eigenvalues kxi,j′ of the category k decision j ' index i are counted to establish (1ri,j′)1×4 and (2ri,j′)1×7, the evaluation result of the fruit harvest classification is expressed as:
A=ρ·(1wi,j′)1×4*(1ri,j′)1×4+(1-ρ)·(2wi,j′)1×7*(2ri,j′)1×7;
wherein, M (+, ·) computation mode is adopted, ρ is the quality weight.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211172555.5A CN115615987B (en) | 2022-09-26 | 2022-09-26 | Fruit harvesting and classifying method integrating appearance quality and internal quality detection |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211172555.5A CN115615987B (en) | 2022-09-26 | 2022-09-26 | Fruit harvesting and classifying method integrating appearance quality and internal quality detection |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115615987A CN115615987A (en) | 2023-01-17 |
CN115615987B true CN115615987B (en) | 2024-06-25 |
Family
ID=84858233
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211172555.5A Active CN115615987B (en) | 2022-09-26 | 2022-09-26 | Fruit harvesting and classifying method integrating appearance quality and internal quality detection |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115615987B (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112163779A (en) * | 2020-10-14 | 2021-01-01 | 南开大学 | Method for evaluating quality of pear fruits |
CN115033756A (en) * | 2022-06-17 | 2022-09-09 | 西南大学 | Screening and planting suitability area evaluation method of citrus fruit comprehensive quality evaluation model |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020092648A1 (en) * | 2018-10-30 | 2020-05-07 | Teakorigin Inc. | Techniques for scoring food specimens, and related methods and apparatus |
WO2022060983A1 (en) * | 2020-09-16 | 2022-03-24 | Colorado State University Research Foundation | Fruit maturity and quality scanning |
-
2022
- 2022-09-26 CN CN202211172555.5A patent/CN115615987B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112163779A (en) * | 2020-10-14 | 2021-01-01 | 南开大学 | Method for evaluating quality of pear fruits |
CN115033756A (en) * | 2022-06-17 | 2022-09-09 | 西南大学 | Screening and planting suitability area evaluation method of citrus fruit comprehensive quality evaluation model |
Also Published As
Publication number | Publication date |
---|---|
CN115615987A (en) | 2023-01-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109308697A (en) | A kind of leaf disease recognition method based on machine learning algorithm | |
Iraji | Comparison between soft computing methods for tomato quality grading using machine vision | |
Sharma et al. | Rapid ripening stage classification and dry matter prediction of durian pulp using a pushbroom near infrared hyperspectral imaging system | |
Kumar et al. | Apple Sweetness Measurement and Fruit Disease Prediction Using Image Processing Techniques Based on Human‐Computer Interaction for Industry 4.0 | |
CN115615987B (en) | Fruit harvesting and classifying method integrating appearance quality and internal quality detection | |
James et al. | High‐throughput phenotyping for breeding targets—Current status and future directions of strawberry trait automation | |
Goyal et al. | Tomato ripeness and shelf-life prediction system using machine learning | |
Liu et al. | Rapid discrimination of high-quality watermelon seeds by multispectral imaging combined with chemometric methods | |
Subbarao et al. | Investigations on Feature Ranking Algorithms for Date Fruit Genotype Detection | |
Wang et al. | Qualitative and quantitative detection of aflatoxins B1 in maize kernels with fluorescence hyperspectral imaging based on the combination method of boosting and stacking | |
Anushya | Quality recognition of banana images using classifiers | |
Aviso et al. | Age factor identification of tomato using labview via image processing | |
Polinar et al. | Near-infrared spectroscopy for non-destructive prediction of maturity and eating quality of ‘Carabao’mango (Mangifera indica L.) fruit | |
Taghadomi‐Saberi et al. | Integration of wavelet network and image processing for determination of total pigments in bitter orange (Citrus aurantium L.) peel during ripening | |
Handayanto | Nondestructive Banana Ripeness Classification using Neural Network | |
Ali et al. | Quality inspection of food and agricultural products using artificial intelligence | |
CN116739403A (en) | Fruit comprehensive quality evaluation method based on quality condition distribution | |
Lavindi et al. | Neural network based on GLCM, and CIE L* a* b* color space to classify tomatoes maturity | |
Shruthi et al. | Trait Based Modelling Approach for Selection of Elite Germplasm Accessions in Soybean [Glycine max (L). Merrill] | |
Al-Mashhadany et al. | Using Machine Learning Algorithms to Predict the Sweetness of Bananas at Different Drying Times. | |
Germšek et al. | Forecasting apple fruit color intensity with machine learning methods | |
Pantazi et al. | Automatic detection of different harvesting stages in lettuce plants by using chlorophyll fluorescence kinetics and supervised self organizing maps (SOMs) | |
Kumar et al. | A novel and effective ensemble machine learning model for identifying healthy and rotten fruits | |
CN111079672B (en) | Grape classification extraction method based on maximum entropy characteristics | |
Pornprasit et al. | Mango fruit quality prediction using associative classification rules |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
CB02 | Change of applicant information |
Address after: 341600 No. 9, Shuanglong Avenue, High tech Zone, Xinfeng County, Ganzhou, Jiangxi Province Applicant after: Lvmeng Technology Co.,Ltd. Address before: 341,600 Jiangxi Lumeng Technology Holding Co., Ltd., Chengxin Avenue, Xinfeng Industrial Park, Ganzhou City, Jiangxi Province Applicant before: JIANGXI REEMOON TECHNOLOGY HOLDINGS Co.,Ltd. |
|
CB02 | Change of applicant information | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant |