A fruit shape quality detection method based on semi-transverse diameter and longitudinal diameter ratio

Abstract

This invention discloses a fruit shape quality detection method based on the ratio of the semi-horizontal diameter to the longitudinal diameter. It can be applied to the quality detection of fruit types with a circular cross-section based on the ratio of their horizontal and vertical diameters, using the F-set probability measure theory. The method sets a range or similar range for the ratio of the left and right semi-horizontal diameters to the longitudinal diameter based on a dataset of sampling points at different longitudinal diameter intervals for a certain standard fruit shape. If the relevant data for a fruit falls within the above-mentioned range or similar range, and is within the tolerable maximum deviation ratio between the left and right sides of the fruit, then the probability that the fruit belongs to that standard fruit shape can be given. This provides a reference for grading and detecting fruit shape based on its appearance quality.

Description

Technical Field

[0001] This invention belongs to the field of postharvest processing technology of fruit, specifically involving a method for detecting the fruit shape and appearance quality of fruit based on the observation angle of the ratio of semi-horizontal diameter to longitudinal diameter and combined with the probability measure theory of F-set. Background Technology

[0002] Each type of fruit has its unique shape, reflecting its overall form and serving as a key indicator for distinguishing fruit grades. Fruits with deformed shapes are often unpopular with consumers, while those with excellent shapes are more readily sought after in the market. In current fruit production, factors such as branch nutrition, flower bud quality, and the application of growth regulators can all influence fruit shape. Furthermore, the number and distribution of seeds during fruit development also affect shape and quality. Therefore, irregular fruit shapes, especially deformed or distorted appearances, significantly impact the fruit's quality. Thus, ensuring that certain grades of fruit conform to or closely approximate their unique shape characteristics during sorting and testing can guarantee the overall quality and grade standards of the fruit at the sales stage, making it better suited to market demands. Summary of the Invention

[0003] The purpose of this invention is to propose the ratio of semi-horizontal diameter to longitudinal diameter as a detection benchmark for several standard fruit shapes of fruit types with round transverse slices. If the ratio of semi-horizontal diameter to longitudinal diameter of a fruit belongs to or is similar to a certain set of ratios of semi-horizontal diameter to longitudinal diameter of the standard fruit shape, and is within the range of the maximum tolerance ratio of the left and right sides of the fruit, then the fruit is considered to conform to the standard fruit shape if the probability of the fruit conforming to the standard fruit shape exceeds the probability threshold of the standard fruit shape. This provides a reference for the grading and detection of fruit shape based on appearance quality.

[0004] To achieve the above objectives, the fruit shape quality detection method based on the ratio of semi-horizontal diameter to longitudinal diameter provided by the present invention includes the following three steps:

[0005] Step 1: Establish a dataset of the ratio of the left and right semi-horizontal diameters to the longitudinal diameter at sampling points at different longitudinal diameter intervals for a certain standard fruit shape. Combine the ratios of the semi-horizontal diameters to the longitudinal diameters of several fruits in this dataset into several sets of similar ratios of the semi-horizontal diameters to the longitudinal diameters of the standard fruit shape within the fluctuation threshold of the ratio of the semi-horizontal diameters to the longitudinal diameters.

[0006] Step 2: Set the range of values ​​for the half-diameter to longitudinal diameter ratio of a certain sampling point in several sets of similar half-diameter to longitudinal diameter ratios of the standard fruit shape, and the tolerable difference in the half-diameter to longitudinal diameter ratio. If the measurement data of a certain type of fruit is within the similar range of the half-diameter to longitudinal diameter ratios in this set, the possibility of its similarity to the half-diameter to longitudinal diameter ratios in this set can be analyzed.

[0007] Step 3: If the ratio of the semi-horizontal diameter to the longitudinal diameter of a fruit belongs to or is similar to a certain set of ratios of the semi-horizontal diameter to the longitudinal diameter of the standard fruit shape, and is within the range of the maximum deviation ratio between the left and right sides of the fruit that can be tolerated, then it can be known that the probability of the fruit being the standard fruit shape exceeds the probability threshold of the standard fruit shape, and the fruit can be identified as conforming to the standard fruit shape. If there are several standard fruit shapes, the one with the highest probability and exceeding the probability threshold of the standard fruit shape is taken as the final standard fruit shape.

[0008] In step 1, the specific steps for establishing a dataset of the ratios of the left and right semi-horizontal diameters to the longitudinal diameter at sampling points at different longitudinal diameter intervals for this standard fruit shape are as follows: Considering that a certain fruit variety may include I types of standard fruit shapes, fruits K that conform to the i-th standard fruit shape are manually selected. i To establish a sample set Φ for this standard fruit shape i And measure the sample set Φ from the side by rotating it by an angle ω to achieve full coverage. i The kth i Sampling points were taken at different locations along the longitudinal diameter of the fruit, from the stem to the navel. The size of the upper left semi-horizontal diameter B(k) i The magnitude of the right half-diameter E(k,j,ω) and the relationship between the two values. i ,j,ω), to establish a dataset Γ of the ratios of the left and right semi-horizontal diameters to the longitudinal diameter at the longitudinal diameter interval positions for fruit shape i. i ={(j,ω,B(k)} i ,j) / A(k i ),E(k i ,j) / A(k i ))}, where 1≤i≤I, 1≤j≤J, 1≤k i ≤K i , W represents the number of times the fruit is rotated sideways for measurement, 1 ≤ w ≤ W, A(k i ) for fruit k i Let J be the length of the fruit from the pedicel to the navel, and J be the number of sampling positions for the transverse diameter at the longitudinal position. Then the sample set Φ i Chinese Fruit K i The ratios of the left and right semi-transverse diameters to the longitudinal diameter in the direction from the fruit stalk to the fruit navel can be expressed as (B(k)). i ,j,ω) / A(k i )) 1×J and (E(k) i ,j,ω) / A(k i )) 1×J ;

[0009] In step 1, the specific steps for merging the semi-horizontal diameter to longitudinal diameter ratios of several fruits in the dataset into several sets of similar semi-horizontal diameter to longitudinal diameter ratios of the standard fruit shape within the fluctuation threshold of the semi-horizontal diameter to longitudinal diameter ratio are as follows: Set the fruit k under different conditions of angle ω. i Fruits i The ratio of the left half of the transverse diameter to the longitudinal diameter The maximum fluctuation range of the ratio of the right half of the transverse diameter to the longitudinal diameter All are less than or equal to the fluctuation threshold ρ of the ratio of the semi-transverse diameter to the longitudinal diameter of the standard fruit shape of the i-th type of fruit. i It can be identified as fruit product k i Fruits i Both are the ratios of similar semi-horizontal diameters to vertical diameters, where max() is a function that takes the maximum value within the set, 1≤k. i ,l i ≤K i Therefore, the sample set Φ of the standard fruit shape of the i-th fruit can be obtained. i Zhong K i The dataset Γ corresponding to the ratio of the left and right semi-horizontal diameters to the vertical diameter. i By comparing the two values, we can obtain H within the fluctuation threshold of the ratio of the semi-horizontal diameter to the longitudinal diameter. i The ratio of the semi-transverse diameter to the longitudinal diameter of similar fruits of standard fruit shape i in group 1, where 1≤H i ≤K i ;

[0010] In step 2, the calculation process for the average ratio of the semi-horizontal diameter to the longitudinal diameter is as follows: In H... i Among the similar transverse and longitudinal diameter ratios of standard fruit shape i in group h, for group γ i From the data of similar semi-horizontal diameter to longitudinal diameter ratios, the average semi-horizontal diameter to longitudinal diameter ratio at the j-th sampling point in the longitudinal diameter position of this group can be calculated as follows: Where r is an intermediate variable, 1≤h≤H i ;

[0011] In step 2, the tolerable difference in the ratio of semi-horizontal diameter to longitudinal diameter is calculated as follows: the tolerable difference in the ratio of semi-horizontal diameter to longitudinal diameter at the j-th sampling point of the h-th group of standard fruit shape i is...

[0012]

[0013] In step 2, the range of values ​​for the ratio of the semi-horizontal diameter to the longitudinal diameter is set as follows: the range of values ​​for the ratio of the h-th group of the standard fruit shape i to the longitudinal diameter is u. i (h) is set to ([C) i,j (h)-σ i,j (h),C i,j (h)+σ i,j (h)])1×J ;

[0014] In step 2, the setting of the similar range of the ratio of the semi-horizontal diameter to the longitudinal diameter is as follows: h-th i The similar range of the ratio of the semi-horizontal diameter to the longitudinal diameter of the group v i (h) can be set to ([C) i,j (h)-2σ i,j (h),C i,j (h)-σ i,j (h)]) 1×J With ([C) i,j (h)+σ i,j (h),C i,j (h)+2σ i,j (h)]) 1×J A collection;

[0015] In step 2, the calculation process for the probability of similar semi-horizontal diameter to longitudinal diameter ratios is as follows: If the measured data of the semi-horizontal diameter to longitudinal diameter ratio of a certain fruit q observed from the angle ω is not within the range of the semi-horizontal diameter to longitudinal diameter ratio of the h-th group of the standard fruit shape i of that fruit, u i (h) but in v i Within (h), the probability that the ratio of the semi-horizontal diameter to the longitudinal diameter of the j-th sampling point in the h-th group of the standard fruit shape i is similar to that of the fruit can be expressed by equation (1):

[0016]

[0017] Where, q j,ω Let q be the ratio of the semi-horizontal diameter to the vertical diameter of the fruit observed at the j-th sampling point from angle ω, and q j,ω >C i,j (h)+σ i,j (h) or q j,ω <C i,j (h)-σ i,j (h);

[0018] The probability that fruit q is similar to the h-th group of standard fruit shape i can be:

[0019] In step 3, the setting of the maximum tolerable deviation ratio between the left and right sides of the fruit is as follows: (Based on dataset Γ) i The maximum deviation ratio of the ratio of the left half of the transverse diameter to the longitudinal diameter and the ratio of the right half of the transverse diameter to the longitudinal diameter at the same position observed from different ω angles can be obtained as follows: It can be seen that for the standard fruit shape of the i-th type of fruit, the tolerable deviation threshold between the left and right sides of the fruit can be set as δ. i ;

[0020] In step 3, the specific steps for determining that a fruit conforms to the standard fruit shape because the probability of the fruit being of that shape exceeds the standard fruit shape probability threshold are as follows: Let the domain of discussion be...

[0021] U i ={u i (1),...,u i (h),...u i (H i ),v i (1),...,v i (h),...v i (H i )}, and “standard fruit shape i” F i For U i Let F be the set on the set F. i It can be expressed using the Zadeh notation with equation (2):

[0022]

[0023] Therefore, the proposition "fruit q is the standard fruit shape i" can link the variable q with the probability distribution in equation (3):

[0024]

[0025] If the ratio of the semi-horizontal diameter to the longitudinal diameter of fruit q belongs to u i (h) or v i (h), then the probability corresponding to the proposition "fruit q is the standard fruit shape i" can be given by equation (3) as 1 or p. i,h (q); Let Ξ be the probability threshold of the standard fruit shape i for the standard fruit shape i. i And 0 < Ξ i <1, when 1 or p i,h (q) exceeds Ξ i The fruit q can be deemed to meet the standard fruit shape i. Attached Figure Description

[0026] Figure 1 This is a flowchart illustrating the fruit shape quality detection steps for the ratio of semi-horizontal diameter to longitudinal diameter according to the present invention.

[0027] Figure 2 This is a flowchart illustrating the standard fruit shape determination process in step 3 of this invention. Detailed Implementation

[0028] Figure 1 This is a flowchart illustrating the fruit shape quality detection steps for the ratio of semi-horizontal diameter to longitudinal diameter according to the present invention. The following detailed description of the technical solution provided by the present invention will be based on a specific embodiment applied to Korla fragrant pears, specifically including the following steps:

[0029] Step 1: Considering that the fruit shape of Korla fragrant pears varies between nearly round, oval, and spindle shapes, and that the axial diameter and minor axis diameter of the pears are similar, we can assume that this fragrant pear variety includes I = 3 standard fruit shapes. We will manually select 200 pears each conforming to the three standard fruit shapes (nearly round, oval, and spindle) to establish a sample set Φ for different standard fruit shapes i. i And rotated 3 times from the side to satisfy the full coverage of the measurement sample set Φ at angle ω. i The kth i Five sampling points for each fruit. The left half of the transverse diameter B(k) i ,j,ω) and the right half of the transverse diameter E(k i ,j,ω), to establish a dataset Γ of the ratios of the left and right semi-horizontal diameters to the longitudinal diameter at the longitudinal diameter interval positions for fruit shape i. i ={(j,ω,B(k)} i ,j) / A(k i ),E(k i ,j) / A(k i ))}, where 1≤i≤3, 1≤j≤5, 1≤k i ≤200, A(k i ) for fruit k i The length of the fruit from the pedicel to the navel is the sample set Φ. i Chinese Fruit K i The ratios of the left and right semi-transverse diameters to the longitudinal diameter in the direction from the fruit stalk to the fruit navel can be expressed as (B(k)). i ,j,ω) / A(k i )) 1×5 and (E(k) i ,j,ω) / A(k i )) 1×5 ;The sample set Φ of the standard fruit shape of the i-th type of fruit i Data set Γ of the ratio of the left and right semi-horizontal diameters to the longitudinal diameter i By comparing and combining the results, we can obtain H within the fluctuation threshold of the ratio of semi-horizontal diameter to longitudinal diameter. i The ratio of the semi-transverse diameter to the longitudinal diameter of similar fruit standard shape i is given. Based on the sample set Φ3 of spindle-shaped pears, it can be merged into 50 groups of similar semi-transverse diameter to longitudinal diameter ratios.

[0030] Step 2, among the similar transverse and longitudinal diameter ratios of the standard pear fruit shape i in group H3 = 50, for group h γ i From similar transverse and longitudinal diameter ratio data, the average value C of the transverse and longitudinal diameter ratio at the j-th sampling point at the longitudinal diameter position can be calculated. i,j (h) and the tolerable difference σ between the ratio of semi-horizontal diameter to longitudinal diameter i,j(h) allows setting the range of values ​​for the ratio of the semi-horizontal diameter to the longitudinal diameter of the h-th group of standard fruit shape i. i (h) and the hth i The similar range of the ratio of the semi-horizontal diameter to the longitudinal diameter of the group v i (h), and give the calculation method for the probability that a certain fragrant pear is similar to the h-th group of standard fruit shape i;

[0031] Step 3, the specific process for judging standard fruit shape is as follows: Figure 2 As shown, if the ratio of the semi-transverse diameter to the longitudinal diameter of a certain pear belongs to a certain range of values ​​u for the ratio of the semi-transverse diameter to the longitudinal diameter of the standard fruit shape i, i (h), according to equation (3), the probability that the pear is of standard fruit shape i is 1, so the pear q can be considered to conform to the standard fruit shape i; if the ratio of the semi-transverse diameter to the longitudinal diameter of a pear q belongs to the corresponding similar range v i (h) Then it is necessary to determine whether the maximum deviation ratio of the left half of the transverse diameter to the longitudinal diameter and the right half of the transverse diameter to the longitudinal diameter at the same position observed from different ω angles is lower than the tolerable deviation ratio threshold δ between the left and right sides of the fruit. i If this value is lower than δ i And the probability p that the pear q is of standard fruit shape i i,h (q) The probability threshold of exceeding the standard fruit shape i Ξ i If the pear q meets the standard fruit shape i, then the pear q can be identified as meeting the standard fruit shape i. If there are several standard fruit shapes, the one with the highest probability and exceeding the standard fruit shape probability threshold is taken as the final standard fruit shape, thus completing the detection of the fruit shape appearance quality of the pear.

[0032] Any parts not explicitly stated in this embodiment can be implemented using existing technologies.

Claims

1. A fruit shape quality detection method based on semi-transverse diameter and longitudinal diameter ratio, characterized in that, Includes the following steps: Step 1: Establish a dataset of the ratio of the left and right semi-horizontal diameters to the longitudinal diameter at sampling points at different longitudinal diameter intervals for a certain standard fruit shape. Combine the ratios of the semi-horizontal diameters to the longitudinal diameters of several fruits in this dataset into several sets of similar ratios of the semi-horizontal diameters to the longitudinal diameters of the standard fruit shape within the fluctuation threshold of the ratio of the semi-horizontal diameters to the longitudinal diameters. Step 2: Set the range of values ​​for the half-diameter to longitudinal diameter ratio of a certain sampling point in several sets of similar half-diameter to longitudinal diameter ratios of the standard fruit shape, and the tolerable difference in the half-diameter to longitudinal diameter ratio. If the measurement data of a certain type of fruit is within the similar range of the half-diameter to longitudinal diameter ratio of the set, analyze the possibility that it is similar to the half-diameter to longitudinal diameter ratio of the set. Step 3: If the ratio of the semi-horizontal diameter to the longitudinal diameter of a fruit falls within or is similar to a certain range of the ratio of the semi-horizontal diameter to the longitudinal diameter of the standard fruit shape, and is within the range of the maximum tolerance ratio of the left and right sides of the fruit, then the fruit is deemed to conform to the standard fruit shape if the probability of the fruit being the standard fruit shape exceeds the standard fruit shape probability threshold. If there are several standard fruit shapes, the one with the highest probability that exceeds the standard fruit shape probability threshold is taken as the final standard fruit shape.

2. The fruit shape quality detection method based on the ratio of the semi-transverse diameter to the longitudinal diameter according to claim 1, characterized in that: In step 1, the specific steps for establishing a dataset of the ratios of the left and right semi-horizontal diameters to the longitudinal diameter at sampling points at different longitudinal diameter intervals for this standard fruit shape are as follows: Considering that a certain fruit variety includes Regarding the standard fruit shape of various types of fruit, manual selection is required to meet the following criteria. Fruits with standard fruit shape This will establish a sample set for this type of standard fruit shape. and rotate from the side Angle full coverage measurement sample set The Middle Sampling points were taken at different locations along the longitudinal diameter of the fruit, from the stem to the navel. Size of the upper left semi-horizontal diameter Size of the right half of the transverse diameter Establish a basis for fruit shape Data set of ratios of the left and right semi-horizontal diameters to the longitudinal diameter at longitudinal diameter intervals. ,in , , , , The number of times the fruit is rotated to the side for measurement. , Fruit The length of the fruit from the pedicel to the navel. The sample set represents the number of sampling locations for the transverse diameter at the longitudinal diameter position. Chinese Fruits The ratios of the left and right semi-transverse diameters to the longitudinal diameter in the direction from the fruit stalk to the fruit navel are respectively expressed as: and .

3. The fruit shape quality detection method based on the ratio of semi-transverse diameter to longitudinal diameter according to claim 1, characterized in that: The specific steps for merging the semi-horizontal diameter to longitudinal diameter ratios of several fruits in this dataset into several sets of similar semi-horizontal diameter to longitudinal diameter ratios within the fluctuation threshold of the semi-horizontal diameter to longitudinal diameter ratio for the standard fruit shape are as follows: Set at... Fruits under different angles Fruits The ratio of the left half of the transverse diameter to the longitudinal diameter The maximum fluctuation range of the ratio of the right half of the transverse diameter to the longitudinal diameter All less than or equal to the first The fluctuation threshold of the ratio of semi-transverse diameter to longitudinal diameter of standard fruit shape Identified as fruit Fruits The two are similar ratios of semi-horizontal diameter to longitudinal diameter, among which A function that takes the maximum value within a set. Therefore, the first Sample set of standard fruit shapes middle Datasets of the ratios of the left and right semi-horizontal diameters to the vertical diameters corresponding to each group. By comparing the two values, the fluctuation threshold of the ratio of the semi-horizontal diameter to the longitudinal diameter can be obtained. Standard fruit shape The ratio of similar semi-horizontal diameter to longitudinal diameter, among which .

4. The fruit shape quality detection method based on the ratio of semi-transverse diameter to longitudinal diameter according to claim 3, characterized in that: In step 2, the calculation processes for the average ratio of the semi-horizontal diameter to the longitudinal diameter and the tolerable difference in the ratio of the semi-horizontal diameter to the longitudinal diameter are as follows: exist Standard fruit shape Among the similar ratios of semi-horizontal diameter to longitudinal diameter, for the first... Group By calculating the ratios of similar transverse and longitudinal diameters, the ratio of the semi-transverse diameter to the longitudinal diameter in this group can be obtained at the longitudinal diameter position. The average ratio of the semi-horizontal diameter to the longitudinal diameter of each sampling point is: ,in As an intermediate variable, ; Standard fruit shape No. Group 1 The tolerable difference in the ratio of the semi-horizontal diameter to the longitudinal diameter of each sampling point is: ; wherein C i (j) represents the jth set of similar semi-transverse diameter to longitudinal diameter ratios for the ith fruit standard shape.

5. The fruit shape quality detection method based on the ratio of the semi-transverse diameter to the longitudinal diameter according to claim 4, characterized in that: In step 2, the settings for the range and similar range of the ratio of the semi-horizontal diameter to the longitudinal diameter are as follows: Fruit standard shape The first Group of half transverse diameter to longitudinal diameter ratio value range Set to ; The first Similar ranges of group semi-transverse to longitudinal diameter ratios Set as With The collection.

6. The fruit shape quality detection method based on the ratio of semi-transverse diameter to longitudinal diameter according to claim 5, characterized in that: In step 2, the calculation process for the probability that the ratios of the semi-horizontal diameter and the longitudinal diameter are similar is as follows: If a certain fruit Depend on The measurement data of the ratio of semi-horizontal diameter to longitudinal diameter observed from the angle is not within the standard fruit shape range for this fruit. The The range of values ​​for the ratio of the group's semi-horizontal diameter to its longitudinal diameter. But inside The inside conforms to the standard fruit shape. The first Group 1 The probability that the aspect ratios of the sampling points are similar is expressed by equation (1): (1)； in, For this fruit Depend on Angle observation in the first The ratio of the semi-horizontal diameter to the vertical diameter of the sampling point, and or ; then fruit With fruit standard fruit shape The Group similar to the possibility of .

7. The fruit shape quality detection method based on the ratio of the semi-transverse diameter to the longitudinal diameter according to claim 6, characterized in that: In step 3, the setting of the maximum tolerable deviation ratio between the left and right sides of the fruit is as follows: (Based on the dataset) It can be obtained from different The maximum deviation ratio of the ratio of the left half of the transverse diameter to the longitudinal diameter and the ratio of the right half of the transverse diameter to the longitudinal diameter at the same position, as observed from an angle, is: It can be seen that for the first The standard fruit shape for a given type of fruit, with a tolerable threshold for the percentage deviation between the left and right sides of the fruit set as follows: .

8. The method for detecting fruit shape quality based on the ratio of semi-transverse diameter to longitudinal diameter according to claim 7, characterized in that: In step 3, the specific steps for determining that a fruit conforms to the standard fruit shape because the probability of the fruit being of the standard fruit shape exceeds the standard fruit shape probability threshold are as follows: Let the domain of discussion be... And "standard fruit shape" " for Set F on the above, let It can be expressed using equation (2) through Zadeh notation: (2) Thus the proposition "fruit is a standard fruit shape " makes the variable associated with the likelihood distribution in equation (3): (3) If fruit The ratio of the semi-horizontal diameter to the longitudinal diameter belongs to or Then, the proposition "Fruit" is given by equation (3). Standard fruit shape The probability of "" corresponds to a value of 1 or Assuming a standard fruit shape The standard fruit shape probability threshold is and When 1 or Exceed The fruit was identified at that time. Fruit shape that meets this standard .

Images