JPS63145928A - Color identifier - Google Patents

Abstract

PURPOSE:To eliminate the need for inputting an allowable discrimination value by an operator, by storing the allowable discrimination value for the formulation of a discrimination reference into a memory beforehand, instead of requiring an operator to input it at each teaching. CONSTITUTION:Lights different in color detected with light receiving elements 11-13 are outputted as color signal components R, G and B from a color sensor 10 via an amplification circuit 14, switched sequentially with a multiplexer 16 and stored into a memory 21 after being taken into a CPU 20 via an A/D converter 24. Then, the total value of the color signal components is computed with the CPU 20 and stored into the memory 21. Subsequently, allowable discrimination values are read out corresponding to the color component signals and the total value at levels of the signals to calculate upper and lower discrimination values. From the results of the calculation, upper and lower limit values are stored into the memory 21 in the form of a table as discrimination reference for color samples. This eliminates the need for inputting required data by an operator at each teaching which repares a reference to discriminate color of an object to be identified thereby reducing working time.

Description

[Detailed description of the invention] Background of the invention Technical field This invention relates to a color identification device.

Conventional technology and its problems One of the conventional color identification devices is to have a color sensor detect in advance a color sample of the object to be identified (hereinafter, this work is referred to as "teaching"), and the color components output from the single color sensor. Some methods create discrimination criteria based on signals. It is determined whether the color is the same as the color sample by comparing the color component signal outputted from the color sensor for the object to be judged with the discrimination standard created in this way. In a conventional color identification device, an operator had to manually input the discrimination tolerance values necessary for creating two discrimination criteria into the device.

That is, when a color sample is presented to the color sensor during teaching, a color component signal, for example R(
The color component signals of red), G (green), and B1 (blue) are output and input to a controller (typically including a microprocessor) of a one-color discrimination device. This control device calculates the total S −R+G +8. continue.

1 1. 1 1 Based on these values, color component data X -R/S,,Y
-G/S and Z-B/5 are calculated. On the other hand, the operator inputs the permissible values xp, yp, zp, and SP (%) for each of the color component data and the total value S of the color component signals. In the control device, based on these data, determinations i and 4I for color identification, that is, determination upper and lower limit values are created as follows.

Upper limit of discrimination X -Xt (1+XP/100)
ax Y -Yt (1+YP/100)l1ax Z -Z (1+ZP/100)IIlax
Lower limit value for determining t S -9 (i+SP/1001 00) X, -X (1-XP/10
0) ffllnt Y -Y (1-YP/100100)
t Z -Z (1-ZP/100) in
t S -8 (1-3P/100) Marin t These values are stored in the memory of the controller.

In color identification processing, when a colored object is detected by a color sensor and a color component signal is output from one color sensor, these color component signals R, G, B and their total value 5 (-R+G+B
) from two color component data X-R/S, Y-G/S, Z-B
/S is calculated, and the following comparison and verification is performed using the above-mentioned discrimination criteria.

X<X<Xff1ax o+In Y＜Y＜YIIla 工 Respectfully In Z<Z<ZIIlax in s < s < s,,.

The in color is determined based on whether these conditions are satisfied. If all the conditions are met, it is determined that the colored object is the same color as the color sample that is the basis for creating the discrimination reference data.

In the conventional color identification device as described above, the permissible discrimination values XP and YP of color component signals are determined during teaching.

An operator must manually operate ZP and SP. Even if the operator can freely decide this value, there are many cases in which the operator cannot determine the value based on the product, and is not sure whether to make a judgment based on fil, and if there are multiple color samples, it may be difficult to determine the input data. Sometimes there are many. For this reason, there was a problem in that the operator had to repeat trial and error, which took two hours and time.

Summary of the Invention Purpose of the Invention The object of the present invention is to provide a color discrimination device that does not require an operator to input data for teaching to create a standard for determining the color of an object to be identified. .

Structure and Effects of the Invention The present invention uses a plurality of colors to represent different colors by combinations thereof based on the output of a color sensor.

a means for creating component data, and a discrimination criterion for identifying the color of the object to be colored based on a plurality of color component data created by the creating means for a predetermined color sample and storing the same. and memory means.

Determining whether or not the color is specified by one discrimination criterion by comparing the plurality of color component data created by the above-mentioned creation means with respect to the color identification object with the discrimination criteria stored in the above-mentioned storage means. The present invention is applied to a color identification device including means for outputting a color identification signal. The color identification device of the present invention includes a storage means for storing a discrimination tolerance value set in advance according to a value related to the output level of one color sensor, and a plurality of discrimination tolerance values created by the above creation means for a predetermined color sample. Create a discrimination standard condition for color discrimination with a certain width from the color component data and the "-2 discrimination tolerance value set for the value related to the output level of the color sensor for creating this color component data." There are means to do so.

In this invention, since the discrimination tolerance value for creating discrimination criteria, which the operator had to input during teaching, is stored in memory in advance, it is not necessary to input this discrimination tolerance value, and the operator can input the color sample. All that is required is a series of steps to set the color within the detection range of the color sensor. Therefore, there is no need to repeat trial and error to find out how much to set the discrimination tolerance as in the past, and the time required for teaching can be shortened. Also,
Since neither complicated key operations nor special knowledge are required, the system can be used even by people who are not experts in the series of operations. Furthermore, in the present invention, the above-mentioned discrimination tolerance value is not set as a series of the same values, but is set according to a value related to the level of the color component signal of the two-color sensor.

The noise component ratio (S/
This is in accordance with the actual situation where the N ratio) is different, and highly accurate color identification can be expected.

DESCRIPTION OF EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to one drawing.

FIG. 1 shows the electrical configuration of the entire two-color identification device.

In FIG. 1, the one-color sensor 10 is one with an integrated light source.
Three light receiving elements 11-13. Amplification circuit 14 that amplifies the output signal (color component signal) of each light receiving element 11 to 13, respectively.
and a light source 15. A filter is provided on the front surface of each of the light receiving elements 11 to 13 to pass light in a wavelength band representing a different specific color, and each of these light receiving elements 11 to 13 detects light of a different color, It outputs an output signal according to the intensity of the light. These output signals are:

For example, they represent the color components of R (red), G (green), and B (blue), respectively. In FIG. 1, color component signals after being amplified are also represented by R, G, and B. light? R
15 is preferably one that generates white light. Although not shown, the light source 15 and the light receiving element 1
A condensing optical system consisting of a lens or the like is provided on the front surface of each of the sensors 1 to 13, and an infrared cutoff filter or the like may be provided if necessary. The light-receiving elements 11 to 13 may be formed separately, or may be monolithically manufactured on one substrate together with a corresponding filter. The light source 15 may not be included in the color sensor 10 but may be provided separately.

Color component signals R and G output from the color sensor IO.

B are sequentially switched by the multiplexer 16.

After being converted into a digital signal by the A/D converter 24,
It is taken into the CPU 20. For convenience, color component signals converted into digital signals are also expressed as R, G, and B. In this embodiment, by adjusting the amplification factor of the amplifier circuit 14,
The color component signals R, G, and B for the white color sample are at the same level, for example, 4V. The CPU 20 adjusts the color component signals for white to the same level.
This can also be done by internal software processing.

The CPU 20 includes a display device 22 including a CRT and an indicator light, a keyboard 23 . It also includes a memory 21 that stores an execution program and a data table storing discrimination tolerance values searched during teaching. The memory 2I also has an area for storing a plurality of discrimination reference data created by teaching as a table, and an area for storing color samples, color component signals detected from the object to be identified, and the like.

The discrimination tolerance data table has a memory configuration shown in FIG. 4, and includes color component signals R and G output from the color sensor 10 during teaching.

Corresponding to each type of B and their total value S, j
t is provided. In Fig. 4, two color component signals R, ,
The data tables 51 to 54 corresponding to C; and B and their total value S are provided with an address table storing the respective table addresses, and each data salary table 51 to 54 is stored in this address table 55.
The search is performed using the address read from . Color sensor 1
The change range (θ~4V) of the color component signal level of 0 is n (
In this example, it is divided into level zones (5 for RlG and B, 4 for S), and the 1 discrimination tolerance value (%) is set for each of these level zones. take different values. Therefore, each data table 51 to 54 stores an upper limit value and a lower limit value that define the range of each level zone, as well as a discrimination tolerance value for that zone.

Each data table 5 is explained using specific numerical values.
The contents stored in 1 to 54 are as shown in the table below.

The discrimination tolerance shown in the above table is set to be smaller as the output level of the two color component signals is higher (that is, the stimulus value is larger). For example, the level of the color component (g −t3 H is the fifth level zone (1000<R<400
0ffly), the discrimination tolerance is 3%.
Then, the first rubel zone (0<R< 100ffl■
) is 40%. This is because the higher the signal level is, the better the S/N ratio becomes, and color identification is possible even if the 2-discrimination tolerance is small; however, when the signal level is low, the S/N ratio deteriorates, so a large tolerance is required. This is because it is necessary. For the total 31゛ value S, the light source 15
Fluctuations in light intensity are also taken into account.

2 and 3 show the processing procedure by the CPU 20. FIG. FIG. 2 shows a process for creating a discrimination reference table for one-color discrimination, and FIG. 3 shows a color discrimination process using this discrimination reference table.

When creating one discrimination standard table with reference to FIG. 2, the operator places the color sample of the object to be color-identified in the detection range of color sensor 0 or causes it to pass through the detection range. Then, color component signals R, Gt, B from one color sensor
is output, this is taken into the CPU 20, and the memory 21
(Step 31). In order to obtain the total value of these color component signals, the CPU 20 uses 5t-Rt+
G, +B are calculated, and this total value St is also stored in the memory 21 (step 32). For each of these color component signals R, G, B, and S, the discrimination tolerance data tables 51 to 54 are searched using the address in the address table 55, and these tS
Discrimination tolerance value corresponding to the level of the issue (XP, YP for convenience)
, denoted by ZP and SP) are read out (step 3
3). Then Xt-R/S, Y-G/S
and Z −B/S are calculated (step 34), and the discrimination tolerance values XP, Y read out by the stirrup 33 are
Using P, ZP and the calculated x, yt, and z, determine the upper limit value and lower limit value, that is, the
, Y corresponding to G and Y , Bman
t max mtn Z and 2 corresponding to t and tIIax corresponding to S
l1in t S and S are determined, respectively (step 11a
)! +1111135). From these calculation results, these lower limit values are stored in the memory 21 in the form of a table as a criterion for the color sample (step 36).

When there are multiple objects to be color-identified.

The above process is repeated for all color samples of these objects, and a discrimination reference table is created for each color sample and stored in the memory 21 (step 37).

Since a table serving as a reference for one-color identification has been created in the above manner, color identification processing can be performed using this table in the primary stage. Referring to FIG. 3, when information regarding the color of the object to be identified is picked up by the color sensor 10, each color component signal R, G is generated.

B is taken into the CPU 20 and the total of these is 5 (-R+G+
Using the calculation result of B), calculate X (-R/≦).

Y (-G/S) and Z (-B/S) are calculated (steps 41 to 43). Then, the upper and lower limit values of the discrimination standard table stored in the memory 21 and the above x, y,
The values of z and S are compared and verified sequentially. x
<x<x, y<y<y, 1lax.

gain wax ll1inz <
z<z, s<S<S (7) Condition ff
1in ff1aX mln I
la! If all of the following are satisfied, the color is the same as the color sample for which the discrimination criterion table was created, and a color identification result to that effect is output to the display device 22 (steps 45 and 46). For example, a corresponding indicator light on the display device 22 lights up. Other predetermined operations may be performed using this color identification signal. If there is no discrimination reference table that satisfies all of the above conditions, a signal indicating that there is no corresponding color is output (step 47).

In this embodiment, the 1 discrimination tolerance value is for the color component signals R9G and B levels and their sum S.

■It is established for each level competition zone divided into numbers, and each zone is set as having a certain value.

Therefore, even if color component signals of different levels are included in the same level zone, the same discrimination tolerance value will be given. Given color component signals R, G, B or color component data X.

An arithmetic program or arithmetic expression for determining the corresponding discrimination tolerance value using Y, Z, and S is stored in the memory 21, and this operation is executed instead of searching the data table. It is also possible to obtain a separate tolerance value.

[Brief explanation of the drawing]

Figure T41 is a block diagram showing the electrical configuration of the entire one-color identification device, Figure 2 is a flowchart showing the procedure for creating a discrimination standard table, and Figure 3 is a flowchart showing the color identification processing means. It shows the memory organization of tables and data tables. IO...Color sensor, 20...CPU. 21...Memory. 51-54... Data table of discrimination reference values. Patent applicant Tateishi Electric Co., Ltd. Representative Patent attorney Kenji Ushiku (1 other person) Figure 2

Claims (1)

[Scope of Claims] Means for creating a plurality of color component data representing the color of an object to be identified by a combination thereof based on the output of a color sensor; a first storage means for setting and storing a discrimination criterion for identifying the color of an object to be colored based on a plurality of color component data; means for determining whether the color is specified by the discrimination criterion by comparing the color component data of the color component data with the discrimination criterion stored in the first storage means, and outputting a color identification signal. In the identification device, a second storage means stores a discrimination tolerance value set in advance according to a value related to the output level of the color sensor, and a plurality of color component data created by the creation means for a predetermined color sample. and means for creating a discrimination standard for color discrimination having a certain width from the discrimination tolerance value set for the value related to the output level of the color sensor for creating the color component data. It is characterized by