JPH0740905B2 - Color detection type leaf tobacco type device - Google Patents

Description

Description: FIELD OF THE INVENTION The present invention relates to a leaf tobacco type device suitable for use in grading the quality of harvested and dried leaf tobacco.

[Conventional technology]

Conventionally, the quality of leaf tobacco is determined by the leaf-positioning position, maturity,
It is evaluated based on quality factors such as color tone, texture, and mesophyll, and it is classified by the naked eye as shown in the table below. Then, pricing is performed corresponding to each purchase category (category marked with "○").

Among the purchase categories in the table above, the color loss type P type is leaf tobacco, which is called quick-drying leaf, and dehydration after yellowing is urgent in the post-harvest drying operation, and the original browning of leaf tobacco is sufficient. There is no yellow color left. In addition, among the mesophylls with poor maturity (synthetic leaves, middle leaves, and lower leaves), white bokeh leaves that are dull and whitish in appearance do not appear glossy, and are also P-type. Is rated as. These P type leaf tobaccos are different from ordinary ones in terms of price because they have problems such as bad taste in tobacco taste and insufficient effect of aroma in the tobacco manufacturing process. For this reason, special attention is paid to P-type ratings.

However, since the P type leaf tobacco as described above is similar to the ordinary type A type leaf tobacco and hue, there is a problem that even a skilled person is difficult to rate by the conventional visual grading method.

An object of the present invention is to make it possible to automatically determine the type of leaf tobacco or the like based on the color of a sample.

[Means for Solving the Problems]

The color detection type leaf tobacco classification device of the present invention made to solve the above-mentioned problems detects a color of a sample and detects a color space of a color system represented by a lightness index of 1 and a perceptual color index of 2. Color measuring means for outputting coordinate information, and memory for storing, for a plurality of leaf tobaccos classified into a plurality of types in advance, information indicating a characteristic boundary in the distribution region of the leaf tobaccos in the color space in association with the type of the sample. Means, input means for inputting and setting the displacement amount of the determination boundary based on the characteristic boundary from the characteristic boundary in the storage means, coordinate information from the color measuring means, and the characteristics stored in the storage means. Based on the information indicating the boundary, the calculating means for obtaining the distance between the measurement point of the leaf tobacco and the characteristic boundary in the color space, the distance obtained by the calculating means and the displacement amount set in the storage means. Based A type of leaf tobacco is determined based on the determination result obtained by the determination means for the leaf tobacco measured by the color measurement means, the determination means determining the inclusion relationship between the measurement point and the area inside the determination boundary. It is characterized by doing so.

[Action]

The information indicating the characteristic boundary of the leaf tobacco distribution area is stored in the storage means, and the determination means identifies the measurement point in the color space of the leaf tobacco as the object to be measured based on the coordinate information from the color measurement means. Then, it is determined whether or not this measurement point is inside the feature boundary and inside the determination boundary where the displacement amount can be determined.

The color of leaf tobacco is distributed with a width on a fixed curve in a color space such as the L ^* a ^* b ^* color system, and the P-type tobacco is distributed on the outer peripheral side in this distribution area.

Therefore, when the measurement point is outside the determination boundary, that is, on the outer peripheral side of the distribution region, it can be determined that the measured leaf tobacco is P type, and when it is inside the central region, it is not P type. it can. In addition, since the amount of displacement between the judgment boundary and the feature boundary can be set and input from outside,
Even if there is an error in the distribution region of leaf tobacco stored in the storage means according to the information indicating the characteristic boundary, the determination condition can be changed appropriately.

〔Example〕

Hereinafter, examples of the present invention will be described by taking leaf tobacco of Burley type as an example.

FIG. 11 is a diagram showing the color distribution of leaf tobacco in the color space of the L ^* a ^* b ^* color system, and FIG. 12 is a diagram showing a cross section taken along the line AA of FIG. 11, and the leaf tobacco distribution is as described above. Corresponding to the landing positions shown in Table 1, a constant color transition path in the color space is shown. Normal-type tobacco is distributed along a transitional path that is curved in the order of lower leaf, middle leaf, synthetic leaf, true leaf, upper leaf and top leaf, and P-type leaf tobacco is biased to an area outside the normal-leaf tobacco. And especially, it is distributed abundantly between synthetic leaves and true leaves.

The present invention measures the data of the lightness index L ^* and the perceptual chromaticity indices a ^* and b ^* in the color space of the L ^* a ^* b ^* color system (hereinafter referred to as the color space value) for the color of leaf tobacco. The P-type leaf tobacco is discriminated based on the distance (color difference) between the measurement point in the color space corresponding to the color space value and the distribution area.

Next, a discrimination method in the embodiment will be described.

(Discrimination Method) First, with respect to the color space of the L ^* a ^* b ^* color system described with reference to FIG. 11, z ₁ , z ₂ , and z ₃ obtained by coordinate conversion shown in the following expression (1) are obtained. While setting a three-dimensional discriminant space,
The distribution state of leaf tobacco in the color space is assumed in the discriminant space by this coordinate transformation, and the distribution state in the discriminant space is set in advance by a determination boundary described later. Also, the color space values from the color sensor 1 are similarly coordinate-converted,
P-type leaf tobacco is discriminated in this discrimination space.

z ₁ = -0.567a ^* + 1.424b ^* + 0.642L ^* -59.052 z ₂ = -0.926a ^* -2.180b ^* + 1.006L ^* +19.048 (1) z ₃ = -1.823a ^* + 0.016b ^* −0.411L ^* +36.144 When the leaf tobacco distribution in the color space is transformed into the discriminant space as described above, the center line of the normal system leaf tobacco distribution area is as shown in FIG. It can be approximated by a circular arc and a straight line. Then, the center line of the distribution region of the middle lobe (lower lobe, middle lobe, and synthetic leaf) is an arc shown by the following equation (2), and the true lobe (true leaf, upper lobe, and top leaf) is the following equation (3). It becomes a straight line indicated by. The figure shows z ₁ , z ₂ , z ₃
Of the three-dimensional discriminant space of z ₁ z ₂ coordinate system (Fig. (A)) and z ₁ z ₃
The coordinate system ((b) in the figure) is shown separately.

On the other hand, the leaf tobacco distribution area is around each center line as shown in FIG. 11, but the width is not constant, and the P-type tobacco area is all the center lines. Since it does not exist in
The center line is divided into 6 sections (~) in the z ₁ z ₂ plane,
A judgment criterion is set for each of the divided sections.

In FIG. 6, when divided into counterclockwise angles centered on the origin with the positive direction of the z ₁ axis as a reference, the distribution state of leaf tobacco is as shown in Table 2 below.

As shown in the above table, since the P type leaf tobacco is present in the categories ~, the determination process first performs coordinate conversion of the measured color space value, and the measured point of the converted coordinate is in the z ₁ z ₂ plane. Whether or not it belongs to the category is determined, and if it belongs to the categories to, the P-type leaf tobacco is determined.

The coordinates (z ₁ , z ₂ , z ₃ ) of the measurement points in the discriminant space are obtained by the coordinate transformation described above, and the angle α indicating the division in the z ₁ z ₂ plane is obtained.
Is Obtained by

As a criterion for each of the categories ~, along the centerline (circle part) of the category ~, a z ₄ z ₃ coordinate system that is perpendicular to this centerline is set, and the characteristics of the boundaries of the distribution area of ordinary leaf tobacco are set. Set the equation of ellipse that shows in the above z ₄ z ₃ coordinate system. The origin of this z ₄ z ₃ coordinate system is on the center line, and z ₄
The coordinate axes are in the z ₁ z ₂ plane of the discriminant space, and the z ₃ coordinate is the same as the z ₃ coordinate of the discriminant space.

The following equation shows the equation of the ellipse of the section ~, and its graph is shown in Figs. 7 (a) to 7 (e). Note that θ in the following equation is the inclination of the major axis of the ellipse, and is represented by a counterclockwise angle centered on the origin with the positive direction of the z ₄ axis as the reference.

Piecewise ellipse equation Piecewise ellipse equation θ = 36 ° where 0 ° ≦ α <45 ° Sectional elliptic equation θ = 169 ° where 335 ° ≦ α <360 ° Sectional elliptic equation θ = 163 ° where 300 ° ≦ α <335 ° Elliptic equation θ = 158 ° However, 270 ° ≤ α <300 ° Ordinary leaf tobacco and P type leaf tobacco are discriminated by using the ellipse for each section set as described above as a characteristic boundary in the discrimination space and measuring points from this characteristic boundary. Based on the distance to (color difference).

The distance x is obtained as follows.

First, the coordinates of the measurement points (z ₁ , z ₂ , z ₃ ) in the discriminant space are converted by the following equation (5) to obtain the coordinates (z ₄ , z ₃ ) of the measurement points on the z ₄ z ₃ plane. .

z ₃ = z ₃ r: radius of the center line, r = 5.942 Then, the distance x is calculated by the following formula (6) from the length of the perpendicular line hung from the point (z ₄ , z ₃ ) to the feature boundary. The value of the distance x is negative inside the feature boundary and positive outside the feature boundary.

Now, the elliptic equation of the feature boundary is Where A _{0 is} the major axis length of the ellipse B _{0 is} the minor axis length of the ellipse The measurement point is (Z _4D , Z _3D ), and the intersection of the perpendicular line hung from the measurement point to the ellipse and the ellipse is (Z _4P , Z _3P ). Then, the distance x can be calculated by the following equation.

However, Z _4D ′ = Z _4D cos θ − Z _3D sin θ Z _3D ′ = Z _4D sin θ + Z _3D cos θ Note that Z _4P is obtained from the following quartic equation using the well-known Ferrari's formula, etc. and a · (Z _4P ) ⁴ + b ・ (Z _4P ) ³ + c ・ (Z _4P ) ² + d ・ (Z _4P ) +
e = 0 ^{_{However, a = (A O 2 -B}} O 2) 2 b = -2 · A 0 2 · Z 4D '· (A O 2 -B O 2) c = -A O 2 · {(A O ^2- B _O ² ) ^2- A _O ²・ Z _4D ′ ^2- B _O ²・ Z _3D ′
_{^{2} d = 2 · A O}} 4 · Z 4D '· (A O 2 -B O 2) e = -A O 6 · Z 4D' 2 addition, Z _3-Way is obtained by the following equation.

As another method, the intersection point (Z _4P , Z _3P ) of the perpendicular line hung from the measurement point to the ellipse and the ellipse is obtained by the following equation, and the distance x is obtained by the equation (6) as described above. Good.

Z _4P = A _O・ cosφ Z _3P = B _O・ sinφ By the way, since the classification of the ordinary type and the P type in the distribution area initially set in the color space is based on the visual rating result, the ordinary type leaves and the P type leaf cigarettes are included in the determination space like the color space. There is an uncertain area where colorimetric values are mixed.

Therefore, as shown in FIG. 3, an elliptic feature boundary as a determination condition for each of the sections 1 to 3 is set so as to be included in the uncertain region, and as shown in FIG.
Probability density relation fn (x) of ordinary system with respect to distance x from the feature boundary and P-type deterministic density function fp
(X) is determined, and the discriminant function D in which the normal production rate of the ordinary type tobacco and the P-type tobacco is added according to the following equation.
(X) is calculated.

FIG. 5 is a diagram showing the discriminant function D (x), where x = 0 corresponds to a feature boundary, a section of x ≦ 0 is inside the feature boundary (normal system area side), and a section of x> 0 is a feature. It corresponds to the outside of the boundary (P-type area side). The range of 0 <D (x) <100 corresponds to the uncertain area.

Therefore, a judgment value Ds is set in advance as a standard for measuring ordinary type tobacco and P type leaf tobacco, and when this determination value Ds is used as a threshold value, it is judged that the ordinary type tobacco is Ds ≦ D. In the case of (x), it is determined that it is a P-type leaf tobacco.

The following equation shows an example of the discriminant function of each section.

The judgment value Ds for comparing the above discriminant functions D (x) is determined from the relation with the current appraisal result, and is set for each category.

By the way, there is a boundary on the Z ₄ z ₃ plane corresponding to the set judgment value Ds. Then, as described above, the value of the discriminant function D (x) is compared with the discriminant value Ds, and P
The distinction from the type corresponds to the value of the judgment value Ds.
This means that the boundary on the Z ₄ z ₃ plane is used as the judgment boundary.
It should be noted that this judgment boundary can be arbitrarily set by the judgment value Ds.

Next, the configuration of the apparatus of the embodiment will be described.

(Arrangement of Apparatus) FIG. 1 is a block diagram of an embodiment of the present invention, in which ordinary type tobacco and P-type tobacco are determined based on the above determination method.

In the figure, 1 is a color sensor that illuminates the object to be measured (leaf tobacco) with illumination light and outputs the colorimetric value as digital data based on the reflected light, and is the lightness index L of the L ^* a ^* b ^* color system. ^{* And} perceived chromaticity index a ^* b ^* data (color space value) is output.
As this color sensor, for example, a "color / color difference meter CR-110" manufactured by Minolta can be used.

Reference numeral 2 denotes a judgment control means composed of a microcomputer or the like, and the functions obtained based on a control program stored in a memory (not shown) are shown in a functional block diagram. The determination control means 2 takes in the color space value from the color sensor 1 via the interface 3, performs various calculations, and displays the determination result on the display means 4. Further, the data for setting the judgment standard is input from the input means 5 such as a keyboard.

In the determination control means 2, the storage means 21 is configured in the memory of the microcomputer, and calculates the data of the angle range indicating each section to, the above-mentioned coordinate conversion formula (1), and the angle in the Z ₁ z ₂ plane. Equation (4), the coordinate value of the Z ₄ z ₃ plane is calculated from the coordinates (z ₁ , z ₂ , z ₃ ) of the measurement point in the discriminant space (5),
Equation (6) for obtaining the distance x from the point (z ₄ , z ₃ ) to the feature boundary, and the discriminant function D (x) of each section to are programmed and stored. Further, the storage means 21 stores the judgment values Ds of the respective categories, which are set and input via the input means 5.

The color space value reading means 22, the coordinate conversion means 23, the calculation means 24, the classification determination means 25, the determination value calculation means 26, the comparison means 27, the output means 28 and the control means 29 are stored in advance in a memory such as a microcomputer. The functions of the respective control means are associated with each other by the control means 29 based on the determination flow described later.

When the color space value reading means 22 detects that the measurement value has been determined by the color sensor 1 based on the interrupt signal from the color sensor 1, it outputs the L ^* , a ^* , b ^* signals output by the color sensor 1. Captured via the interface 3, this L ^* , a ^* , b ^*
The signal of is output as numerical data.

The coordinate conversion means 23 generates coordinate data z ₁ , z ₂ , z ₃ in the discriminant space by the coordinate conversion based on the above-mentioned expression (1), and the calculation means 24 calculates Z ₁ based on the above-mentioned expression (4). In addition to finding the angle in the z ₂ plane, Z ₄ ．. Obtain the coordinate data z ₄ , z ₃ on the z ₃ plane.

The section discriminating unit 25 discriminates the section of the measurement points (z ₁ , z ₂ , z ₃ ) in the discriminating space based on the angle range of the sections stored in the storage unit 21 and the obtained angle, Discriminant value calculation means
26 sets a feature boundary and a discriminant function for the discriminated section from the storage means 21, and sets the set feature boundary and z ₄ ,
The distance d is calculated from z ₃ and the discriminant value D (d) is calculated.

The comparison means 27 sets the determination value Ds for the determined classification from the storage means 21, compares the determination value D (d) with the determination value Ds, and determines that it is a normal system when D (d) ≦ Ds. Then D
When (d)> Ds, it is determined to be P type. Then, the data indicating the determination result is output to the output means 28, and the output means 28
The display data of the determination result is output to the display unit 4 based on the data indicating the determination result.

FIG. 2 is a flow chart showing the determination processing performed by the determination control means 2 under the control of the control means 29. The color space value reading means 22 causes the color sensor 1 to operate on the basis of the interrupt signal from the color sensor 1. When it is detected that the measured value is fixed, the L ^* , a ^* , b ^* signals output by the color sensor 1 are taken in through the interface 3, and the L ^* , a ^* , b ^* signals are output as numerical data. Yes (step S ₁ ).

By the coordinate conversion means 23, obtained by the color space value reading means 22
The coordinate data based on the equation (1) described above is applied to the numerical data of the L ^* , a ^* , and b ^* signals to generate coordinate data z ₁ , z ₂ , z ₃ in the discrimination space (step S ₂ ).

The calculation means 24 further includes z ₁ , z ₂ , obtained by the coordinate conversion means 22.
The angle in the z ₁ z ₂ plane is calculated based on z ₃ and the above equation (4), and further the coordinate data z ₄ , z ₃ on the z ₃ z ₄ plane is calculated based on the above equation (5). (step S _3).

The division determining means 25 determines the division of the points (z ₁ , z ₂ , z ₃ ) based on the angle range of the divisions stored in the storage means 21 and the angle obtained by the calculating means 24, and makes a determination. Value calculation means
By 25, the feature boundary and the discriminant function for the section discriminated by the section discriminating unit 25 are set from the storage unit 21 (step S ₄ ).

The discriminant value calculation means 25 determines the length of the perpendicular line hung from the point (z ₄ , z ₃ ) to the feature boundary based on the set feature boundaries and z ₄ and z ₃ obtained by the calculation means 24. Then, the distance d is generated (step S ₅ ), and the discriminant value D (d) is obtained from the discriminant function (step S ₆ ).

By the comparison means 26, the judgment value Ds for the classification judged by the classification judgment means 25 is set from the storage means 21 and compared with the judgment value D (d) obtained by the judgment value calculation means 25 (step
S ₇ ), when D (d) ≦ Ds, it is judged as a normal system (step
S _8), when the D (d)> Ds is determined that P-type (Step S _9).

The determination result obtained as described above is displayed on the display means 4 by the output means 28.

In the above embodiment, the z ₁ z ₂ plane is divided into 6 sections (sections ~), but one of them is
As shown in FIG. 9, it may be divided into five sections.
In this case, although there is a drawback that the final P type discrimination accuracy is slightly lowered, there is an advantage that the arithmetic logic for discrimination is simplified. An example of the ellipse of the feature boundary in the division in this case is shown in the following equation and FIG.

θ = 160 ° However, 270 ° ≦ α <335 ° [Effect of the invention] As described above, according to the color detection type leaf tobacco type device of the present invention, in the color space such as L ^* a ^* b ^* color system. The information indicating the distribution area of leaf tobacco is stored, the measurement point in the color space is identified by the color space value of leaf tobacco measured by a color sensor, etc., and the distance (color difference) between the determination boundary and the measurement point of the stored distribution area is determined. ), It is possible to determine the inclusion relation between the measurement points and the distribution area and to set the determination boundary from the outside, so it is possible to distinguish the type of leaf tobacco by color, and it is possible to distinguish between ordinary-type tobacco with similar colors. And P type leaf tobacco can be easily discriminated. Further, the criterion for discrimination can be set arbitrarily.

[Brief description of drawings]

FIG. 1 is a block diagram of an embodiment of the present invention, FIG. 2 is a flow chart relating to the embodiment, FIG. 3 is a diagram for explaining characteristic boundaries of a normal region and a P type region in the embodiment, and FIG. FIG. 5 is a diagram showing a probability density function in an example, FIG. 5 is a diagram showing a discriminant function in the example, FIG. 6 is a diagram illustrating divisions in the example, and FIG. 7 is an example of an ellipse of a feature boundary in the example. FIG. 8 is a diagram showing a center line of a distribution area in a discriminant space in the embodiment, FIG. 9 is a diagram for explaining divisions in another embodiment, and FIG. 10 is a characteristic boundary in the embodiment of FIG. Fig. 11 is a diagram showing an example of an ellipse of Fig. 11, Fig. 11 is a diagram showing the distribution state of leaf tobacco in the color space according to the present invention, and Fig. 12 is a diagram showing a cross section of the distribution region of Fig. 11. 1 ... Color sensor, 2 ... Judgment control means, 3 ... Input means, 4 ... Display means.

Claims (1)

[Claims] 1. A color measuring means for detecting color of a sample and outputting coordinate information of a color space of a color system represented by a lightness index of 1 and a perceptual color index of 2, and a plurality of types are classified in advance. Storage means for storing the information indicating the characteristic boundary in the distribution area of the leaf tobacco in the color space in association with the type of the specimen for the plurality of leaf tobaccos, and the characteristic boundary of the determination boundary based on the characteristic boundary. Based on the input means for inputting and setting the displacement amount from the storage means, and the coordinate information from the color measuring means and the information indicating the characteristic boundary stored in the storage means, the measurement of leaf tobacco in the color space is performed. Computation means for obtaining the distance between the point and the characteristic boundary, and inclusion of the measurement point and the area inside the judgment boundary based on the distance obtained by the calculation means and the displacement amount set in the storage means. Determine relationship A color detection type leaf tobacco type device, characterized in that the type of leaf tobacco is determined based on the determination result obtained by the determination means for the leaf tobacco measured by the color measurement means.