A method for identifying the color of first-cured tobacco

Abstract

The present invention provides a method for identifying the color of freshly flue-cured tobacco, comprising the following steps: 1) using the back of freshly flue-cured tobacco leaves as an identification surface; 2) determining the length of the freshly flue-cured tobacco leaves and the main vein of the freshly flue-cured tobacco leaves; 3) evenly selecting three position points on the main vein from the stem end to the tip direction of the main vein; 4) taking the main vein as the symmetry axis, selecting six points at positions with a distance of 2CM to 4CM from both sides of the three position points as measurement points, and measuring the reflection spectrum of the back of the freshly flue-cured tobacco by a spectrophotometer; 5) performing second-order derivative processing on the reflection spectrum of the above-mentioned measurement points, respectively, to obtain the second-order derivative value at 500nm, the second-order derivative value at 670nm, the second-order derivative value at 520nm, and the second-order derivative value at 640nm; 6) obtaining the color scale of the freshly flue-cured tobacco; 7) classifying the freshly flue-cured tobacco according to the value of the color scale. The present invention uses the color of the back of the freshly flue-cured tobacco to accurately identify the overall color of the tobacco leaves.

Description

Technical Field

[0001] The invention belongs to the field of tobacco leaf quality identification, and in particular relates to a method for identifying the color of first-cured tobacco, which is used for quantitatively evaluating the color of first-cured tobacco. Background Art

[0002] The surface color of flue-cured tobacco leaves not only affects the appearance quality of tobacco leaves, but is also closely related to the quality of tobacco leaves. At present, both domestic and foreign countries use artificial sensory methods to grade tobacco leaves in terms of color determination of flue-cured tobacco leaves, that is, graders rely on visual observation and experience to classify tobacco leaves according to the depth of color, such as lemon yellow, orange yellow, red brown, etc. The artificial sensory method cannot meet the requirements of the automated production line for the amount and supply speed of tobacco leaves. Affected by artificial experience, the quality of tobacco leaf grading will also be unstable, affecting the quality stability of finished cigarettes.

[0003] Under the general trend of intelligent tobacco leaf grading, the digital transformation of tobacco leaf characteristics is an urgent problem to be solved. The existing research by Wang Gaili shows that the color of flue-cured tobacco is related to the values ​​of L* (brightness), a* (redness), b* (yellowness), and C* (saturation). Based on this, a variety of methods for judging the color of flue-cured tobacco from the front of the tobacco leaf have been subsequently developed.

[0004] The color of flue-cured tobacco is determined by a variety of color-developing substances. During the growth of tobacco leaves, the front side is usually directly exposed to environmental substances such as rain and sunlight, resulting in uneven color on the front side of the tobacco leaves. The color difference at different measuring points is large, which increases the randomness and uncertainty of the classification of the first flue-cured tobacco grades. The color consistency of the back side of the first flue-cured tobacco is better, which is conducive to digital analysis and is more representative. Therefore, a quantitative model based on the color of the back side of the leaf is needed to determine the color of the first flue-cured tobacco. Summary of the invention

[0005] The purpose of the present invention is to provide a method for identifying the color of freshly-cured tobacco, which improves the recognition accuracy and quality of the color of freshly-cured tobacco by collecting and processing the color information of the back of the freshly-cured tobacco.

[0006] The present invention adopts the following technical solutions to solve the above technical problems:

[0007] A method for identifying the color of first-cured tobacco comprises the following steps:

[0008] 1) The back side of the first-cured tobacco leaf is used as the identification surface;

[0009] 2) determining the length of the first-cured tobacco leaves and the main veins of the first-cured tobacco leaves;

[0010] 3) From the handle end to the tip end of the main vein, select three evenly spaced points on the main vein;

[0011] 4) Taking the main vein as the symmetry axis, select 6 points at a distance of 2CM to 4CM on both sides of the three positions as measurement points, and measure the reflectance spectrum of the back of the primary flue-cured tobacco by a spectrophotometer;

[0012] 5) performing second-order derivative processing on the reflectance spectra of the above-mentioned measuring points respectively to obtain second-order derivative values ​​at 500 nm, 670 nm, 520 nm and 640 nm;

[0013] 6) The color scale of the first-cured tobacco is obtained by the following formula:

[0014] Color scale = (second-order derivative value at 500 nm + second-order derivative value at 670 nm) / (second-order derivative value at 520 nm + second-order derivative value at 640 nm);

[0015] 7) Classify the first-cured tobacco according to the value of the color scale:

[0016] Color scale value ≥ -1.70, tobacco leaves are lemon yellow;

[0017] -1.70>color scale value ≥-3.00, tobacco leaves are orange-yellow;

[0018] Color scale values ​​< -3.00 indicate that tobacco leaves are reddish brown.

[0019] The method for identifying the color of the first-cured tobacco of the present invention finds that the color scale index of the back side of the first-cured tobacco leaf has a strong correlation with the color of the front side of the flue-cured tobacco, which can more accurately reflect the color of the first-cured tobacco, thereby accurately identifying the overall color of the tobacco leaf using the color of the back side of the first-cured tobacco.

[0020] In one embodiment, six points 3CM on both sides of the 1 / 3, 1 / 2, and 2 / 3 positions of the main vein are selected as measurement points. The surrounding colors can accurately represent the color of the leaves, reducing the probability of color collection deviation due to the measurement points being located at the branch veins.

[0021] In one embodiment, the second-order derivative value is measured by a spectrophotometer and then obtained by second-order derivative processing; the detection process has low requirements on the detection environment and samples, and has good application prospects in the intelligent sorting of tobacco leaf color.

[0022] Compared with the prior art, the present invention has the following advantages:

[0023] The present invention proposes a method for identifying the color of flue-cured tobacco, which does not require sample processing before detection, and can reduce the complexity of detection and calculation. By measuring the spectral characteristics of the color of the back of tobacco leaves with a spectrophotometer and performing second-order derivative processing, the three colors of lemon yellow, orange yellow, and reddish brown of tobacco leaves can be judged, which can reduce the misjudgment of the color of flue-cured tobacco due to the experience and sensory differences of graders.

[0024] The present invention provides that the color consistency of the back of the flue-cured tobacco is better, which is conducive to digital analysis and has stronger representativeness, thus solving the problem of intelligent sorting application of tobacco leaves. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 Schematic diagram of the first-baked tobacco leaf of the present invention

[0026] Description of the accompanying drawings

[0027] 1 First-cured tobacco leaves

[0028] L Length of first-cured tobacco leaves

[0029] M Main vein of first-cured tobacco leaves

[0030] M0 Main vein starting point

[0031] M1, M2, M3 location points

[0032] M11, M12, M21, M22, M31, M32 measuring points DETAILED DESCRIPTION

[0033] The technical solution of the present invention is further described in detail below in conjunction with the embodiments, but the embodiments are not intended to limit the protection scope of the present invention.

[0034] like Figure 1 The method for identifying the color of the first-cured tobacco leaves shown comprises the following steps:

[0035] 1) using the back side of the first-cured tobacco leaf 1 as the identification side;

[0036] 2) determining the length L of the first-cured tobacco leaves and the main vein M of the first-cured tobacco leaves;

[0037] 3) The starting point M0 of the main vein is located at the stem end of the freshly cured tobacco leaf 1, and from the starting point M0 of the main vein to the tip of the freshly cured tobacco leaf, the 1 / 3 position point M1, the 1 / 2 position point M2, and the 2 / 3 position point M3 of the main vein are selected;

[0038] 4) Taking the main vein M as the symmetry axis, select 6 points (M11, M12, M21, M22, M31, M32) on both sides of the 1 / 3, 1 / 2, and 2 / 3 positions of the main vein and a distance of 3CM as measurement points, and measure the reflectance spectrum of the back of the sample leaf with a spectrophotometer;

[0039] 5) performing second-order derivative processing on the reflectance spectra of the above-mentioned measuring points respectively to obtain second-order derivative values ​​at 500 nm, 670 nm, 520 nm and 640 nm;

[0040] 6) The color scale of the first-cured tobacco is obtained by the following formula:

[0041] Color scale = (second-order derivative value at 500 nm + second-order derivative value at 670 nm) / (second-order derivative value at 520 nm + second-order derivative value at 640 nm);

[0042] 7) Classify the first-cured tobacco according to the value of the color scale:

[0043] Color scale value ≥ -1.70, tobacco leaves are lemon yellow;

[0044] -1.70>color scale value ≥-3.00, tobacco leaves are orange-yellow;

[0045] Color scale values ​​< -3.00 indicate that tobacco leaves are reddish brown.

[0046] The main vein M is usually located on the back of the first-cured tobacco leaf. Therefore, this method makes it easier to obtain the main vein and related measurement points.

[0047] In order to verify the accuracy of the method described in the present invention, a number of samples were collected. The production areas of the samples included Yuxi, Yunnan, Dali, Yunnan, Chuxiong, Yunnan and Kunming, Yunnan. The varieties were the main flue-cured tobacco varieties grown locally. The sampled colors included: lemon yellow, orange yellow, and red brown, with 5 kg of each color.

[0048] First, an evaluation team composed of tobacco leaf grading experts selected the samples several times through sensory analysis technology, and finally determined the exact color of the collected samples to improve the accuracy of the results. 12 leaves of each color from each part were randomly selected and numbered. The back of the 36 samples was used as the identification surface; after determining the length L of the first-cured tobacco leaves and the main vein M of the first-cured tobacco leaves, from the starting point M0 of the main vein to the tip of the first-cured tobacco leaves, the main veins on the back of the tobacco leaves were selected at about 1 / 3, 1 / 2, and 2 / 3 positions, with the main vein M as the symmetry axis, and 6 relatively flat positions about 3 cm away from the main vein were selected as measurement points. The reflectance spectrum of the back of the first-cured tobacco leaves was measured with a spectrophotometer, and second-order derivative processing was performed. The average value of the second-order derivative at 500nm, the average value of the second-order derivative at 670nm, the average value of the second-order derivative at 520nm, and the average value of the second-order derivative at 640nm were taken as the color index parameters of the first-cured tobacco to be determined.

[0049] Substitute the color index parameters into the color scale judgment model: color scale = (second-order derivative value at 500nm + second-order derivative value at 670nm) / (second-order derivative value at 520nm + second-order derivative value at 640nm), and calculate the color scale value. If the color scale value is ≥-1.70, the tobacco leaf is lemon yellow; -1.70>the color scale value of the tobacco leaf to be tested is ≥-3.00, the tobacco leaf is orange-yellow; the color scale value of the tobacco leaf to be tested is <-3.00, the tobacco leaf is reddish brown, and finally the identification data as shown in Table 1 are obtained:

[0050] Table 1

[0051]

[0052]

[0053]

[0054] The above verification results show that the method for identifying the color of freshly-cured tobacco described in the present invention can achieve a 100% color recognition accuracy for upper tobacco and a 91.66% color recognition accuracy for middle and lower tobacco; the average color recognition accuracy for freshly-cured tobacco in different parts can reach 94.44%.

[0055] The color recognition method of the first flue-cured tobacco of the present invention does not need to process the sample before detection, which can reduce the complexity of detection and calculation. The spectral characteristics of the color of the back of the tobacco leaf are measured by a spectrophotometer and second-order derivative processing is performed to realize the judgment of the three colors of lemon yellow, orange yellow and red brown of the tobacco leaf, which can reduce the misjudgment of the color of the first flue-cured tobacco due to the experience and sensory differences of the graders, is conducive to digital analysis, has stronger representativeness, and solves the problem of intelligent sorting application of tobacco leaves.

Claims

1. A method for identifying the color of first-cured tobacco, characterized in that: The following steps are involved: 1) The back side of the first-cured tobacco leaf is used as the identification side; 2) determining the length of the first-cured tobacco leaves and the main veins of the first-cured tobacco leaves; 3) From the handle end to the tip end of the main vein, select three evenly spaced points on the main vein; 4) Taking the main vein as the symmetry axis, select 6 points at a distance of 2CM to 4CM on both sides of the three positions as measurement points, and measure the reflectance spectrum of the back of the primary flue-cured tobacco by a spectrophotometer; 5) performing second-order derivative processing on the reflectance spectra of the above-mentioned measuring points respectively to obtain second-order derivative values ​​at 500 nm, 670 nm, 520 nm and 640 nm; 6) The color scale of the first-cured tobacco is obtained by the following formula: Color scale = (second-order derivative value at 500 nm + second-order derivative value at 670 nm) / (second-order derivative value at 520 nm + second-order derivative value at 640 nm); 7) Classify the first-cured tobacco according to the value of the color scale: Color scale value ≥ -1.70, tobacco leaves are lemon yellow; -1.70>color scale value ≥-3.00, tobacco leaves are orange-yellow; Color scale values ​​< -3.00 indicate that tobacco leaves are reddish brown.

2. The method for identifying the color of first-cured tobacco according to claim 1, characterized in that: Step 3) is to select 6 points on both sides of the 1 / 3, 1 / 2, and 2 / 3 positions of the main vein as measurement points.

3. The method for identifying the color of first-cured tobacco according to claim 1, characterized in that: Step 4) is to select 6 points 3CM away from the three positions of the main vein on both sides as measurement points.

4. The method for identifying the color of first-cured tobacco according to claim 1, characterized in that: The second-order derivative value is a second-order derivative value obtained by measuring with a spectrophotometer and then performing a second-order derivative process.

Images