A method for judging the maturity of fresh tobacco

CN114577739BActive Publication Date: 2024-11-19TOBACCO RESEARCH INSTITUTE OF CHINESE ACADEMY OF AGRICULTURAL SCIENCES (QINGZHOU TOBACCO RESEARCH INSTITUTE OF CHINA NATIONAL TOBACCO COMPANY)
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210257974.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-03-16
Publication Date
2024-11-19
Estimated Expiration
2042-03-16

AI Technical Summary

Technical Problem

但是,该方法仍需将烟叶采摘后,通过多种装备辅助且在特定条件下对烟叶成熟度进行判定

Benefits of technology

[0019] Beneficial effects: Compared with the existing technology, the advantage of the present invention is that: the present invention collects the color value matrix of fresh tobacco leaves through the color sensor matrix under the light source matrix. According to the color value matrix of the fresh tobacco leaves and the set maturity value, The relationship between the two is established, and a regression prediction model is established to obtain a maturity calculation model; the maturity value is calculated based on the maturity calculation model to achieve rapid and accurate maturity judgment. It is highly efficient, does not need to collect tobacco leaf images, requires small computing resources, and has accurate output results. .

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114577739B_ABST
    Figure CN114577739B_ABST
Patent Text Reader

Abstract

The method for determining the maturity of fresh tobacco leaves disclosed in the present invention comprises: selecting fresh tobacco leaves of different maturity, classifying the fresh tobacco leaves according to their appearance by artificial classification, and assigning maturity values; collecting color values ​​of the classified fresh tobacco leaves to obtain the R value, G value, and B value of each fresh tobacco leaf and the R value, G value, and B value of each type of fresh tobacco leaves; establishing a regression prediction model, and outputting a maturity calculation model; setting a maturity value interval corresponding to a maturity grade; collecting color values ​​and calculating maturity values ​​of unharvested fresh tobacco leaves; determining the maturity value interval in which the maturity value is located, and outputting the corresponding maturity grade. The present invention does not need to collect tobacco leaf images, is not affected by human subjectivity and environmental conditions, and the maturity judgment result is easy to be read by tobacco farmers, and is easy to promote and apply.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of tobacco leaf maturity detection, and in particular to a method for determining the maturity of fresh tobacco. Background Art

[0002] Baking is the last step in the production of cigarette raw materials and the first step in industrial processing. The maturity of fresh tobacco harvest is the basis and necessary condition for determining the quality of baking. The maturity characteristics of field tobacco leaves are essentially manifested by yellow pigments such as lutein and carotenoids during the degradation of chlorophyll, accompanied by the transformation of chemical substances such as starch, reducing sugar, and nicotine. Scholars have done a lot of research on the appropriate maturity determination index for fresh tobacco. For example, Zhu Zunquan suggested judging by the leaf age of different parts of different growing tobacco plants; Zhang Shutang (CN201110069396.1) believed that the maturity can be judged by SPAD value (relative content of plastid pigment) as a determination index; Wu Shengjiang (CN201310195869.1) disclosed a method of judging the maturity of tobacco leaves by multi-point determination of the L*a*b / L*C*h color value of tobacco leaves by a colorimeter; Gao Xianhui and others used the yellow-green area ratio of tobacco leaves and the degree of whitening of the main veins as quantitative indicators of the maturity of fresh tobacco; others proposed to use chemical components such as tobacco leaf sucrose and α amino acids (CN201720959920.5) as determination indicators. Similar methods are widely used in the United States. In production practice and research, tobacco leaf villi shedding, stem-leaf angle, conductivity, reflectance spectrum characteristics (CN201410791732.7) and other indicators are also used as determination indicators. The judgment methods based on SPAD meter and colorimeter both have the problem of only measuring one point at a time. Although multi-point measurements can be performed by increasing the workload, in actual measurements, we found that the SPAD values ​​and color values ​​at different points vary greatly, especially when thicker fresh tobacco leaves mature, green and yellow spots often appear alternately distributed. Even the different parts of normally mature tobacco leaves (leaf tip, leaf middle, leaf base, leaf edge) vary greatly, and the judgment accuracy of this type of method is still unreliable.

[0003] In terms of judging the maturity of fresh tobacco, there are many similar patents in China that imitate the colorimetric cards used in Japan, such as gloves with tobacco leaf colorimetric cards (CN201821741815.5), colorimetric cards for the maturity of the upper leaves of flue-cured tobacco (CN201220189553.2), a device that can improve the consistency of tobacco leaf maturity when picking flue-cured tobacco (CN201721627389.8), a hand-cranked fan for distinguishing the maturity of tobacco leaves (CN201620857896.X), an instrument for distinguishing the maturity of tobacco leaves and a method for distinguishing the maturity of tobacco leaves (CN201610648510.9), a portable tobacco leaf mature harvest colorimetric bracelet (CN201920288156.2), a flue-cured tobacco red large variety mature harvest color identification method (CN201910658971.8), a fresh tobacco leaf maturity grading colorimetric ring (CN201822003998.7), the biggest disadvantage of this method is that tobacco farmers still need to use the naked eye to compare colors to judge the maturity of tobacco leaves, which is highly subjective and greatly affected by environmental conditions, especially light conditions. In addition, the color of normal mature and yellowing tobacco leaves in the field is manifested as the leaves gradually turning yellow from the tip to the base of the leaves and from the edge to the middle of the leaves, and the main veins gradually turning white and shiny from the tip to the base of the leaves. In the current tobacco production, the maturity of tobacco leaves is still judged by the yellow-green ratio of the leaf area and the white-green ratio of the main vein length. Therefore, the feasibility of judging the maturity of fresh tobacco by a method similar to the color card remains to be discussed.

[0004] In order to reduce or avoid the problem of strong subjectivity in the above-mentioned methods in manually determining the maturity of tobacco leaves, some scholars have established a method for determining the maturity of fresh tobacco leaves by collecting tobacco leaf images and extracting and assigning values ​​to image color features. For example, a tobacco leaf maturity detection method and detection device (CN201110144539.0; CN201120180257.1) converts the collected tobacco leaf images into HSV (hue / saturation / brightness) color mode images, and calculates the maturity grade through H, S values ​​and corresponding mathematical models; a non-destructive rapid detection method for the maturity of flue-cured tobacco leaves (CN201811541440.2) collects tobacco leaf color images, extracts the R value in the image, and calculates the maturity value through the formula "maturity value = 18*R / 255"; a tobacco leaf classification method based on spectral and machine vision coupling (CN201911071652.3) also collects tobacco leaf color images, performs image processing, and extracts a series of image feature values ​​to calculate the maturity. This type of method requires first collecting images of tobacco leaves, which is cumbersome and costly. The light environment in which the tobacco leaves are located when collecting images has a great influence on the color characteristics of the image, which in turn affects the extraction of image feature values ​​and the establishment of mathematical models. In addition, the scope of application of the model is still a problem that needs to be solved.

[0005] In order to avoid the influence of environmental factors on the color characteristics of the image during the collection of tobacco leaf images, Shen et al. (Determination of the maturity of fresh tobacco leaves based on the skewed distribution pattern of leaf color Tobacco Science and Technology 54.08 (2021): 26-35.) collected tobacco leaf images after the tobacco leaves were picked, extracted multiple color parameters, and modeled them through back-transfer neural networks to determine the maturity of the tobacco leaves. However, this method still requires the tobacco leaves to be picked, assisted by a variety of equipment and under specific conditions to determine the maturity of the tobacco leaves. However, the significance of determining the maturity of tobacco leaves is to guide tobacco producers to determine in advance whether the leaves growing on the tobacco plants have reached the maturity standard, so as to achieve appropriate harvesting. Summary of the invention

[0006] Purpose of the invention: The purpose of the present invention is to address the deficiencies in the prior art and provide a method for determining the maturity of fresh tobacco, which does not require the collection of tobacco leaf images and is not affected by human subjectivity and environmental conditions. The maturity judgment result is easy for tobacco farmers to read, and the cost is low and easy to promote and apply.

[0007] Technical solution: The method for determining the maturity of fresh tobacco leaves of the present invention comprises establishing a maturity calculation model and determining the maturity:

[0008] The establishment of the maturity calculation model comprises: S101: selecting fresh tobacco leaves of different maturity, classifying the fresh tobacco leaves according to their appearance by manual classification, obtaining N types of fresh tobacco leaves of different maturity levels, and assigning maturity values ​​to the N types of fresh tobacco leaves from low to high maturity levels; S102: collecting color values ​​of the classified fresh tobacco leaves, placing the fresh tobacco leaves in a sealed dark box containing a light source matrix and a color sensor matrix, obtaining each fresh tobacco leaf exposed to the light source matrix and collected by the color sensor matrix The obtained RGB value matrix is ​​averaged to obtain the R value, G value, and B value of each fresh tobacco leaf; S103: for each type of fresh tobacco leaves with the same maturity level, the R value, G value, and B value of all fresh tobacco leaves of this type are averaged to obtain the R value, G value, and B value of this type of fresh tobacco leaves; S104: the R value, G value, B value of N types of fresh tobacco leaves and the combined characteristic value of the R value, G value, and B value are used as model input, and the fresh tobacco leaf maturity assignment is used as the model output to establish a regression prediction model, and the maturity calculation model is output;

[0009] The maturity determination includes: S201: setting maturity value intervals corresponding to N types of maturity levels; S202: using the method of S102 to collect color values ​​of unharvested fresh tobacco leaves to obtain R values, G values, and B values ​​of the fresh tobacco leaves; S203: substituting the collected R values, G values, and B values ​​of the fresh tobacco leaves into a maturity calculation model to calculate a maturity value Y; S204: determining the maturity value interval in which the maturity value Y is located, and outputting the maturity level corresponding to the fresh tobacco leaves.

[0010] To further improve the above technical solution, the fresh tobacco leaves are divided into four categories according to their appearance: under-ripe, moderately ripe, moderately ripe, and over-ripe by manual classification. There are no less than 100 fresh tobacco leaves of each maturity level, and the maturity levels are assigned Y1, Y2, Y3, and Y4 in sequence, and the values ​​of Y1 to Y4 show an increasing trend.

[0011] Furthermore, the values ​​of Y1, Y2, Y3 and Y4 are assigned to 80, 90, 100 and 110 respectively; S104 includes: taking the R value, G value, B value and the values ​​of the combined features 2G-RB, R / G and GR of the four types of fresh tobacco leaves as model inputs, assigning the maturity values ​​of the four types of fresh tobacco leaves to Y1, Y2, Y3 and Y4 as model outputs, establishing a regression prediction model, and the output maturity calculation model is Y=0.738*R-0.759(GR)-0.653B.

[0012] Furthermore, the S201 includes: the maturity classification includes underripe, moderately ripe, moderately ripe, and overripe, and the maturity value interval corresponding to moderately ripe is determined. A maturity value 10% or more higher than the upper limit of the moderately ripe maturity value interval is overripe, a maturity value 10% to 20% lower than the lower limit of the moderately ripe maturity value interval is moderately ripe, and a maturity value 20% or less lower than the lower limit of the moderately ripe maturity value interval is underripe.

[0013] Furthermore, the method for determining the maturity value interval corresponding to the appropriate maturity includes: manually selecting 10 to 20 pieces of fresh tobacco leaves of the same tobacco variety in the same tobacco area of ​​the year that are appropriate to mature, collecting color values ​​of the selected fresh tobacco leaves of the appropriate maturity by the method of S102, substituting the collected color values ​​into the maturity calculation model to calculate the maturity values, and determining the maturity value interval corresponding to the appropriate maturity according to the calculated maturity values.

[0014] Furthermore, S102 is completed by a color value acquisition device and an intelligent terminal, the color value acquisition device includes an upper panel and a lower panel, a color sensor matrix and a light source matrix are provided on the inner surface of the upper panel, a controller is provided inside the upper panel, and the inner side of the lower panel is a black frosted flat plate structure, and the upper panel and the lower panel form a sealed dark box after clamping the fresh tobacco leaves; the controller interacts with the intelligent terminal after being connected, and the intelligent terminal outputs a control instruction to the controller, and the controller starts the color sensor matrix and the light source matrix according to the control instruction to collect the color values ​​of the fresh tobacco leaves, and the color values ​​collected by the color sensor matrix are sent to the intelligent terminal via the controller.

[0015] Furthermore, the color sensor matrix is ​​composed of a plurality of color sensors arranged in a matrix, the light source matrix is ​​composed of a plurality of LED lights arranged in a matrix, and a single light source is arranged at the center of four color sensors in a 2×2 array.

[0016] Furthermore, the color sensor matrix includes 15 to 25 color sensors, and the light source matrix includes 8 to 16 LED lamps; the color sensor is provided with a light-blocking sleeve for shielding direct light emitted by the LED lamp.

[0017] Furthermore, the color value acquisition device is 35-45 cm long, 15-20 cm wide, and 2-4 cm thick; the light-blocking sleeve is 0.4-0.8 cm high, and the distance between the bottom end of the light-blocking sleeve and the inner surface of the lower panel or the fresh tobacco leaf to be tested is 0.4-1 cm.

[0018] Furthermore, the intelligent terminal is provided with a color value transmission unit, a maturity value calculation unit, a maturity determination unit and an input-output unit. The color value transmission unit is used to receive the color value sent by the controller and transmit it to the maturity value calculation unit. The maturity value calculation unit calculates the color value and outputs the maturity value to the maturity determination unit. The maturity determination unit is used to receive the setting of the maturity determination interval and output the maturity determination result according to the determination interval in which the maturity value is located. The input-output unit is used to receive the setting input of the maturity determination interval and the display output of the maturity value and the maturity determination result.

[0019] Beneficial effect: Compared with the prior art, the advantages of the present invention are: the present invention obtains a color value matrix of fresh tobacco leaves under a light source matrix through a color sensor matrix, and establishes a regression prediction model to obtain a maturity calculation model based on the relationship between the color value matrix of fresh tobacco leaves and the set maturity value; the maturity value is calculated based on the maturity calculation model to achieve fast and accurate maturity judgment, with high efficiency, no need to collect tobacco leaf images, small computing resources required, and accurate output results.

[0020] The present invention collects the color value of fresh tobacco leaves through a color value collection device. The upper panel and the lower panel of the color value collection device clamp the tobacco leaves to form a sealed dark box. The light source matrix of the upper panel provides light to ensure that the light remains consistent during each tobacco leaf color collection process, thereby avoiding the influence of light on color collection in different weather conditions. A color sensor matrix is ​​used to collect the color of leaves on one side of the main vein of the fresh tobacco. The color collection area covers half of the leaf, and the matrix design can ensure the uniformity of the collection points. The RGB collected by the color sensor matrix is ​​averaged to avoid deviation caused by collecting only part of the tobacco leaf area. The maturity value is quickly calculated through a maturity calculation model, and the maturity judgment interval can be set according to different tobacco planting areas and different maturity requirements, and the adjustment is convenient and fast.

[0021] The color value acquisition device provided by the present invention has a simple structure, a small size and a low cost. When in use, it is only necessary to clamp the leaf on one side of the main vein of the fresh tobacco leaf to obtain the determination result of the maturity value of the leaf. The operation is convenient and accurate determination of the maturity of the leaf can be achieved without the need for professional personnel. The device is easy to promote and fills the gap in portable fresh tobacco leaf maturity determination equipment on the market. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a flowchart for establishing a maturity calculation model;

[0023] Figure 2 It is a flowchart of maturity determination;

[0024] Figure 3 It is a structural schematic diagram of the color value acquisition device in the present invention;

[0025] Figure 4 It is a structural schematic diagram of the upper panel of the color value acquisition device in the present invention;

[0026] Figure 5 It is a structural schematic diagram of the lower panel of the color value acquisition device in the present invention;

[0027] Figure 6 It is a schematic diagram of the exploded structure of the upper panel of the color value acquisition device in the present invention.

[0028] In the figure: 1. upper panel, 2. lower panel, 3. light source, 4. color sensor, 5. light blocking sleeve, 6. upper sealing ring, 7. host control board, 8. power module, 9. lower sealing ring, 10. handle. DETAILED DESCRIPTION

[0029] The technical solution of the present invention is described in detail below with reference to the accompanying drawings, but the protection scope of the present invention is not limited to the embodiments.

[0030] The method for judging the maturity of fresh tobacco provided by the present invention includes establishing a maturity calculation model and determining the maturity. Figure 1 As shown, establishing a maturity calculation model includes:

[0031] S101: Select fresh tobacco leaves of different maturity levels, and have three certified technicians in the tobacco industry classify the fresh tobacco leaves into four categories according to their appearance: under-mature, moderately mature, moderately mature, and over-mature. No less than 100 leaves of each maturity level are selected, and the maturity levels are assigned 80, 90, 100, and 110 in order from low to high maturity levels;

[0032] S102: The color value of the classified fresh tobacco leaves is collected by a color value collection device to obtain an RGB value matrix of the fresh tobacco leaves, and the RGB value matrix is ​​averaged to obtain an R value, a G value, and a B value of each fresh tobacco leaf;

[0033] S103: for each type of fresh tobacco leaves of the same maturity grade, average the R value, G value, and B value of all the fresh tobacco leaves of the type to obtain the R value, G value, and B value of the fresh tobacco leaves of the type;

[0034] S104: The R value, G value, B value and the combined characteristic values ​​of 2G-RB, R / G and GR of the four types of fresh tobacco leaves are used as model inputs, and the maturity values ​​of the four types of fresh tobacco leaves are assigned as 80, 90, 100 and 110 as model outputs. A regression prediction model is established, and the output maturity calculation model is Y=0.738*R-0.759(GR)-0.653B.

[0035] In the process of maturity determination, it is necessary to first set the maturity value range corresponding to the maturity level. Because different tobacco producing areas, different tobacco varieties, tobacco leaves from different parts, and tobacco leaves of the same variety and the same leaf position under different climatic conditions in the same producing area in different years may require different harvesting maturity, the tobacco production technical guidance department in the same tobacco area can select 10-20 suitable mature tobacco leaves in the field for measurement based on the quality of fresh tobacco of the year, and then determine the Y value range for tobacco leaves suitable for harvesting, and issue the Y value range to each grassroots technical guidance unit (such as tobacco station), so as to ensure that the guidance work on tobacco leaf maturity harvesting can be quantified and promote accurate mature harvesting of tobacco leaves.

[0036] like Figure 2 As shown in Figure 1, the maturity determination process includes:

[0037] S201: manually selecting 10 to 20 pieces of fresh tobacco leaves of the same tobacco variety from the same tobacco area of ​​the year, collecting color values ​​of the selected fresh tobacco leaves through a color value collection device, substituting the collected color values ​​into a maturity calculation model to calculate a maturity value, and determining a maturity value interval corresponding to maturity according to the calculated maturity value, over-mature if it is 10% or more higher than the upper limit of the maturity value interval, moderately mature if it is 10% to 20% lower than the lower limit of the maturity value interval, and under-mature if it is 20% or less lower than the lower limit of the maturity value interval;

[0038] S202: Using a color value collection device to collect color values ​​of unharvested fresh tobacco leaves, to obtain R values, G values, and B values ​​of the fresh tobacco leaves;

[0039] S203: Substituting the R value, G value, and B value of the collected fresh tobacco leaves into a maturity calculation model to calculate a maturity value Y;

[0040] S204: Determine the maturity value interval in which the maturity value Y is located, and output the maturity level corresponding to the fresh tobacco leaves.

[0041] The color value collection of the fresh tobacco leaves in the present invention is completed by a color value collection device, such as Figure 3As shown, it is a handheld clamping device, which is 35~45cm long, 15~20cm wide, and 2~4cm thick, and includes an upper panel 1 and a lower panel 2. Figure 4 As shown, a light source matrix (color temperature 6000K) and a color sensor matrix are evenly arranged on the inner surface of the upper panel 1. The color sensor matrix includes 20 color sensors 4 arranged in 4 rows and 5 columns. The light source 3 is set at the center of the color sensor 4. The light source 3 is an LED lamp. The color sensor 4 is surrounded by a square or circular light-blocking sleeve 5 with a height of about 0.5 cm to block the direct light emitted by the light source 3 to interfere with the color sensor 4 in capturing the reflected light of fresh tobacco leaves and thus affect the judgment of the color value. At the same time, the distance between the end of the light-blocking sleeve 5 and the inner side of the lower panel 2 or the surface of the tobacco leaves to be tested (the clamped tobacco leaves) is ensured to be 0.5 cm.

[0042] Figure 4 In the embodiment, a 2 mm thick and 5 mm wide foam strip is glued to the inner edge of the upper panel 1 to form an upper sealing ring 6; Figure 5 As shown, the lower panel 2 is a black frosted (non-reflective) flat plate, and a 2mm thick and 5mm wide foam strip is glued to the inner edge to form a lower sealing ring 9. When clamping tobacco leaves, the lower sealing ring 9 is completely corresponding to the upper sealing ring 6, which plays a role in shielding from light and preventing tobacco leaves from being damaged.

[0043] like Figure 6 As shown, the color value acquisition device is equipped with a controller 7, a 10000 mAh power module 8 and a charging socket, and the power module 8 is used by the color sensor 4 and the light source 3. The color value acquisition device is equipped with a retractable multifunctional data cable, which can be connected to an intelligent terminal, and the intelligent terminal can be a mobile phone or other device. The intelligent terminal can output control instructions to the color value acquisition device. At the same time, the intelligent terminal is provided with functional modules such as a color value transmission unit, a maturity value calculation unit, a maturity determination unit and an input and output unit. The color value transmission unit is used to receive the color value sent by the controller and transmit it to the maturity value calculation unit. The maturity value calculation unit calculates the color value and outputs the maturity value to the maturity determination unit. The maturity determination unit is used to receive the setting of the maturity determination interval and output the determination result according to the determination interval where the maturity value is located. The input and output unit is used to receive the setting of the maturity determination interval and display and output the maturity value and the maturity determination result.

[0044] During the measurement, after turning on the switch of the color value acquisition device, connect the multifunctional data connection line or Bluetooth on the color value acquisition device to the mobile phone, open the upper panel 1 and the lower panel 2 through the handle 10, clamp the middle section of one side of the main vein of the fresh tobacco leaf (the 15 cm of the leaf tip and the leaf base are not included in the measurement range), and form a relatively closed and light-proof dark box. Open the mobile phone APP and click on the "Fresh Tobacco Maturity Determination" function. The color value acquisition device will turn on the light source matrix and the color sensor matrix to collect the color values ​​of fresh tobacco leaves (the basic three colors of red R, green G and blue B). The color value of fresh tobacco leaves is collected through the color sensor matrix composed of 20 color sensors 4, and 20 groups of R value, G value and B value can be obtained. The 20 groups of R value, G value and B value are transmitted to the mobile phone APP via a data cable or Bluetooth, and the average value of the 20 groups of R value, G value and B value is input into the tobacco maturity calculation model. The tobacco maturity calculation model Y=0.738*R-0.759(GR)-0.653B set in the APP can immediately display the maturity information on the mobile phone, expressed as Y value, and output the corresponding maturity according to the interval of the Y value.

[0045] The present invention acquires a color value matrix of fresh tobacco leaves under a light source matrix through a color sensor matrix, and establishes a regression prediction model to obtain a maturity calculation model according to the relationship between the color value matrix of the fresh tobacco leaves and a set maturity value; the maturity value is calculated based on the maturity calculation model to achieve fast and accurate maturity judgment, with high efficiency, no need to acquire tobacco leaf images, small required computing resources, and accurate output results.

[0046] As described above, although the present invention has been shown and described with reference to specific preferred embodiments, it should not be construed as limiting the present invention itself. Various changes in form and details may be made without departing from the spirit and scope of the present invention as defined in the appended claims.

Claims

1. A method for determining the maturity of fresh tobacco leaves, characterized in that: Including the establishment of maturity calculation model and maturity determination: The establishment of the maturity calculation model comprises: S101: selecting fresh tobacco leaves of different maturity levels, and classifying the fresh tobacco leaves by maturity according to their appearance by manual classification, so as to obtain N types of fresh tobacco leaves of different maturity levels, and assigning maturity values ​​to the N types of fresh tobacco leaves from low to high maturity levels; S102: color values ​​of the classified fresh tobacco leaves are collected, the fresh tobacco leaves are placed in a sealed dark box containing a light source matrix and a color sensor matrix, an RGB value matrix of each fresh tobacco leaf exposed to the light source matrix and collected by the color sensor matrix is ​​obtained, and the R value, G value, and B value of each fresh tobacco leaf are obtained by average calculation of the RGB value matrix; S103: for each type of fresh tobacco leaves of the same maturity grade, average the R value, G value, and B value of all the fresh tobacco leaves of the type to obtain the R value, G value, and B value of the fresh tobacco leaves of the type; S104: using the R value, G value, B value and the combined characteristic value of the R value, G value and B value of the N types of fresh tobacco leaves as model inputs and the maturity value of the fresh tobacco leaves as model outputs to establish a regression prediction model and output a maturity calculation model; The maturity determination includes: S201: Setting the maturity value interval corresponding to N maturity levels; S202: using the method of S102 to collect color values ​​of unharvested fresh tobacco leaves, to obtain R values, G values, and B values ​​of the fresh tobacco leaves; S203: Substituting the R value, G value, and B value of the collected fresh tobacco leaves into a maturity calculation model to calculate a maturity value Y; S204: Determine the maturity value interval in which the maturity value Y is located, and output the maturity level corresponding to the fresh tobacco leaves.

2. The method for determining the maturity of fresh tobacco leaves according to claim 1, wherein: The fresh tobacco leaves are classified into four categories according to their appearance: under-ripe, moderately ripe, moderately ripe and over-ripe by manual classification. There are no less than 100 fresh tobacco leaves of each maturity level, and the maturity levels are assigned Y1, Y2, Y3 and Y4 in sequence, with the values ​​from Y1 to Y4 showing an increasing trend.

3. The method for determining the maturity of fresh tobacco leaves according to claim 2, wherein: The values ​​of Y1, Y2, Y3 and Y4 are assigned as 80, 90, 100 and 110 respectively; S104 includes: taking the R value, G value, B value and the values ​​of the combined features 2G-RB, R / G and GR of the four types of fresh tobacco leaves as model inputs, assigning the maturity values ​​of the four types of fresh tobacco leaves as Y1, Y2, Y3 and Y4 as model outputs, establishing a regression prediction model, and the output maturity calculation model is Y=0.738*R-0.759(GR)-0.653B.

4. The method for determining the maturity of fresh tobacco leaves according to any one of claims 1 to 3, characterized in that: The S201 includes: the maturity classification includes under-mature, moderately mature, moderately mature, and over-mature, and the maturity value interval corresponding to moderately mature is determined. If it is 10% or more higher than the upper limit of the moderately mature maturity value interval, it is over-mature, if it is 10% to 20% lower than the lower limit of the moderately mature maturity value interval, it is moderately mature, and if it is 20% or less lower than the lower limit of the moderately mature maturity value interval, it is under-mature.

5. The method for determining the maturity of fresh tobacco leaves according to claim 4, characterized in that: The method for determining the maturity value interval corresponding to the appropriate maturity includes: manually selecting 10 to 20 pieces of fresh tobacco leaves of the same tobacco variety in the same tobacco area of ​​the year that are appropriate to mature, collecting color values ​​of the selected fresh tobacco leaves of the appropriate maturity by the method of S102, substituting the collected color values ​​into the maturity calculation model to calculate the maturity values, and determining the maturity value interval corresponding to the appropriate maturity according to the calculated maturity values.

6. The method for determining the maturity of fresh tobacco leaves according to claim 3, characterized in that: The S102 is completed by a color value acquisition device and an intelligent terminal. The color value acquisition device includes an upper panel and a lower panel. A color sensor matrix and a light source matrix are provided on the inner surface of the upper panel. A controller is provided inside the upper panel. The inner side of the lower panel is a black frosted flat plate structure. The upper panel and the lower panel form a sealed dark box after clamping the fresh tobacco leaves. The controller interacts with the intelligent terminal after being connected, and the intelligent terminal outputs a control instruction to the controller. The controller starts the color sensor matrix and the light source matrix according to the control instruction to collect the color values ​​of the fresh tobacco leaves. The color values ​​collected by the color sensor matrix are sent to the intelligent terminal via the controller.

7. The method for determining the maturity of fresh tobacco leaves according to claim 6, characterized in that: The color sensor matrix is ​​composed of a plurality of color sensors arranged in a matrix, the light source matrix is ​​composed of a plurality of LED lights arranged in a matrix, and a single light source is arranged at the center of four color sensors in a 2×2 array.

8. The method for determining the maturity of fresh tobacco leaves according to claim 7, characterized in that: The color sensor matrix includes 15 to 25 color sensors, and the light source matrix includes 8 to 16 LED lamps; the color sensor is provided with a light-blocking sleeve for shielding direct light emitted by the LED lamp.

9. The method for determining the maturity of fresh tobacco leaves according to claim 8, characterized in that: The color value acquisition device is 35-45 cm long, 15-20 cm wide, and 2-4 cm thick; the light-blocking sleeve is 0.4-0.8 cm high, and the distance between the bottom end of the light-blocking sleeve and the inner surface of the lower panel or the fresh tobacco leaf to be tested is 0.4-1 cm.

10. The method for determining the maturity of fresh tobacco leaves according to claim 6, characterized in that: The intelligent terminal is provided with a color value transmission unit, a maturity value calculation unit, a maturity determination unit and an input-output unit. The color value transmission unit is used to receive the color value sent by the controller and transmit it to the maturity value calculation unit. The maturity value calculation unit calculates the color value and outputs the maturity value to the maturity determination unit. The maturity determination unit is used to receive the setting of the maturity determination interval and output the maturity determination result according to the determination interval where the maturity value is located. The input-output unit is used to receive the setting input of the maturity determination interval and the display output of the maturity value and the maturity determination result.