DE3503204C2 - Device for identifying colors using color component signals - Google Patents

Description

The invention relates to a device for identifying Colors using color component signals according to the preamble of the patent claim 1.

A device of the type specified in the preamble of claim 1 is described in the main patent 33 27 954. With the device described in the main patent, the values of the color fraction signals can be determined independently of lighting fluctuations. Changes in the brightness of the color therefore do not cause errors in the measurement results. The brightness of the color changes many times due to external influences, e.g. B. by aging of the light source or fluctuations in the supply voltage supply of the light source.

A method for identifying colors whose color is known is known parts of red, green and blue using filters and photoelectric Receivers for the acquisition of color fraction signals are measured. The Measured values are fed to amplifiers with an adjustable degree of amplification and then processed into a sum signal. There will be each Output signals for the three pure colors red, green and blue together generated with basic proportions. From the output signals and the basic divide (resting potentials) absolute color fractions are obtained that be added up. The color components and the sum are a Kom paratorbank fed in which a weighted comparison for determination the respective color is carried out (DE 28 32 382 A1).

A color measuring device with several photosensitive devices is also known Receivers, some of which are different spectral Possess sensitivities. The spectral becomes sensitive through calibration speed of the respective photoelectric receiver determined. The recipients are using a multiplexer and an analog-digital converter with one Computer connected, which consists of the received signals and the stored spectral sensitivities of the receivers and based on stored Standard spectral value curves, color value components calculated (DE 25 46 253 A1).  

For some bodies there is a connection between the color of the Surface, the degree of saturation of the color and other physical Characteristics. When making the body can therefore be desired Properties can be achieved if the surface color or the Degree of saturation to the value corresponding to the respective property is set.

The invention has for its object to develop a device with which the color type on the surface of a body by influencing a parameter influencing the color type to a constant one Value can be regulated.

The task is in a device in the preamble of the patent Proof 1 described type according to the invention by the features of characterizing part of claim 1 solved. With that in the patent device described, a color component signal is selected, that changes with the physical property that relates to you certain value should be set. The regulation of the degree of saturation the color in the manufacture of the body lends itself when the desired property by measurement otherwise only cumbersome or cannot be recorded at all.

It can happen that a certain physical property clearly related only to the sum of two color component signals. Then it is expedient to achieve the physical property from which Surface radiation, the z. B. based on reflection, by filtering two To detect color components that are converted into two electrical color component signals be multiplied by a coefficient, then added and then compared with the setpoint.

The parameter is preferably the concentration of a dye which a filler of a given color is added. With that at Production of a paint the dosage of a filler can be regulated the color of the mixture to a constant degree of saturation adjust.

Advantageous refinements of those in claims 1 and 2 Measures specified are described in claims 3 to 6.

The invention is illustrated below with reference to a drawing presented embodiments explained in more detail.  

Show it:

Figure 1 shows a device for generating a control signal with which a manipulated variable for controlling the type of color on the surface of a body can be influenced.

Fig. 2 shows an arrangement for the signed multiplication of the Farban partial signals with selected coefficients;

Fig. 3 shows an arrangement for the illumination fluctuation-independent generation of color component signals and

Fig. 4 color fraction signals of a body as a function of the concentration of mixed substances of the body.

A surface area 1 of an object 2 is imaged by optics 3 on three photoelectric receivers 4 , 5 , 6 arranged closely next to one another. Three color separation filters 7 , 8 , 9 are arranged in front of the photoelectric receivers 4 , 5 , 6 in the beam path of the light beam that images the area. The color separation filter 7 is e.g. B. permeable to the primary color red. The color separation filters 8 , 9 are each permeable to the primary colors green and blue. There are therefore three signals at the outputs of the photoelectric receivers 4 , 5 , 6 , each corresponding to the color type and the brightness prevailing during the scanning for the primary colors red, green and blue.

The receivers 4 , 5 , 6 are each followed by an arrangement 10 , 11 , 12 with which the output signal of the assigned receiver 4 , 5 , 6 and adjustable values can be amplified and optionally inverted. Each arrangement 10 , 11 , 12 contains the same structure. An adjustable resistor 13 is provided on the input side in each arrangement 10 , 11 , 12 , which is followed by an amplifier 14 with a feedback resistor 15 . The amplifier 14 feeds a differential amplifier 17 , which has a feedback resistor 18 , via a further resistor 16 . The second input of the differential amplifier 17 is connected to a contactless changeover switch 19 , for example, the switching element of which can optionally be connected to the output of the amplifier 14 or to reference potential. The arrangements 10 , 11 , 12 can optionally be connected to the receivers 4 , 5 , 6 via switching elements 20 , 21 , 22 . The switching elements 20 , 21 , 22 , which are shown in Fig. 1 following the receiver 4 , 5 , 6 , can also be arranged within the Anord openings 10 , 11 , 12 or after these. With the switching elements 20 , 21 , 22 one or two of the three color component signals can be selected for a control process. At the arrangements 10 , 11 , 12 resistors 23 , 24 , 25 of a summing circuit 25 a are connected, which contains a further adjustable resistor 26 , which can be applied to a positive or negative reference voltage either via a switching element 27 . With the summing circuit 25 a, an amplifier 28 is connected to an adjustable feedback resistor 29 .

With the resistors 13 , the color component signal selected in each case via the associated switches 20 , 21 , 22 can be multiplied by a coefficient, the size of which depends on the resistance value. The polarity of the product on the color component signal and coefficient is predetermined by the position of the switch 19 . The products with the corresponding sign are superimposed on one another in the summing circuit 25 a. Via the resistor 26, the setpoint is in the summing circuit 25 a fed. Resistor 26 can also be used to shift the actual value to zero. Depending on the position of the switch 27 and the set value of the resistor 26 , the actual value or the control deviation occurs at the output of the summing circuit 25 a. If necessary, both can be multiplied by a further coefficient via the setting of the feedback resistor 29 .

The arrangements 10 , 11 , 12 can also have the structure shown in FIG. 2. A provided with a feedback resistor Difference amplifier 31 is connected to an input via a resistor 32 to the respective photoelectric receiver, for. B. 4 connected. The receiver 4 can be connected to a preamplifier 33 . The second input of the differential amplifier 31 is connected to the tap of a potentiometer 34 , which is arranged between the input-side connection of the resistor 32 and the reference potential. With the potentiometer 34 , the coefficient can be continuously adjusted between +1 and -1.

The adjustable resistors 13 , 26 and 29 ( Fig. 1) can be replaced by coding switches and suitably stepped resistors in order to digitally specify the coefficients, the shift or the setpoint or the amplification of the control deviation or the setpoint or the amplification of the To allow control deviation or the actual value.

The output signals of the photoelectric receivers 4 , 5 , 6 can also be fed to amplifiers 35 , 36 and 37 ( FIG. 3), the output signals of which are combined with resistors (not specified) to form a sum signal.

This sum signal is compared as the actual value of a controlled variable with a pre-given constant setpoint, the z. B. is generated by a reference voltage source 38 . The common connection point of the resistors is connected to an input of a differential amplifier 39 , to the second input of which the reference voltage source 38 is connected. The differential amplifier 39 generates a control deviation signal corresponding to the difference between the input signals, which is fed together to the amplifiers 35 , 36 , 37 designed as amplifiers with a controllable degree of amplification. With the control deviation signal, the gain factors are set so that the control deviation signal becomes zero or has a very small value. At the outputs of the three amplifiers 35 , 36 , 37 color component signals are therefore available which are always related to the same brightness corresponding to the setpoint value and which are therefore independent of the fluctuations in the illuminance on the surface in the area.

The amplifier 35 , 36 , 37 have outputs 35 a, 36 a, 37 a, to which the arrangements 10 , 11 , 12 are connected if the color of an upper surface area is to be detected independently of lighting fluctuations.

An infrared filter 67 and an interference edge filter 68 are each arranged in front of the color separation filters 7 , 8 , 9 in the beam path.

In the arrangement shown in Fig. 1, the photoelectric receivers 4 , 5 , 6 are close together. This means that different parts of area 1 are imaged on the receivers 4 , 5 , 6 . Such an arrangement works precisely when these parts have the same color, ie there are no or only negligible color changes between the parts. This arrangement is sufficient for a large number of objects, since the conditions explained above exist. If necessary, the three color components can also be filtered out from the radiation of the same surface element by using dichroic mirrors in conjunction with filters. The use of notch filters also enables filtering with reference to the same surface element.

The parameters for changing the color of the respective object can be influenced.  

The arrangements described above are intended, for example, to regulate the Color of a mixture of substances can be used by adding a Dye to a filler of certain color by influencing the Concentration is established.

An example will be treated explicitly for clarification. A fill Fabric of a certain color becomes another dye in selectable concentrations tration added. At a certain concentration, that gives Mixed product the desired color, which is constant via a control should be kept.

For this purpose, the mixed product is viewed with a color sensor that supplies the three color component signals. In order to be able to use them as a control criterion, their behavior when the concentration changes is required, ie their characteristic curve must be recorded in advance. For this purpose, the concentrations of the dye are changed in a defined manner and the resulting color fraction signals are measured. It results, for. B. the course shown in Fig. 4 (which was somewhat idealized here). The Farban partial signals have a kink in the concentration k₀; their slopes change there:

tgα _R = dU _RM / dk; tgα _G = dU _GM / dk; tgα _B = dU _BM / dk.

Due to the regulation of the three color component signals described above, theirs remains Sum constant, e.g. B. 10 V.

It can be seen from FIG. 4 that none of the three partial voltages can be used directly as a controlled variable:

• - U _RM and U _BM do not change when the concentration is smaller or larger than k₀;
• - U _GM has a maximum at k₀.

Therefore U _RM and U _{BM are combined} to a new size by z. B. U _BM negated and then potential shifted and added to U _RM after appropriate amplification.

The following applies to the resulting tension:

U _is = c₁ U _RM + c₂ U _BM + c₃ U _ref ,

where U _{ref is} a (constant) reference voltage. The result should be a variable proportional to the concentration k:

U _is = tg α _i k with tgα _i = 0.2 V /%.

The numerical value results from the area of interest for k - 0 to 50% - and the available voltage range - 10 V.  

The object is achieved with the circuit according to FIG. 1, the switch 21 being opened. For simplification, all fixed resistors in FIG. 1 should have the value 10 kOhm. Furthermore, switch 20 is at the top (non-inverting) and switch 22 at the bottom (inverting). The following relationships then apply to the coefficients:

c₁ = R₂₉ / R _{13 ′} ; c₂ = -R₂₉ / R _{13 ′ ′} ; c₃ = -R₂₉ / R₂₆,

with R'₁₃, R''₁₃, R₂₆ and R₂₉ the values of the resistors 13 of the arrangements 10 and 12 and the resistors 26 , 29 are designated.

For k = 0 should also be U = _is 0;

c₁ U _RM (0) + c₂ U _BM (0) + c₃ U _ref = 0.

Furthermore, the slopes to the left and right of k₀ should be the same:
c₁ · dU _RM (k <k₀) = c₂ · dU _BM (K <K₀); from it:

c₁ / c₂ = tgαB / tgα _R.

With the numerical values

U _RM (0) = 4 V; U _BM (0) = 6 V;
tgα _R (k <k₀) = 0.1 V /%; tgα _B (k <k₀) = -0.2 V /%;
R₂₉ = 20 kOhm (selectable)

you get the results

R'₁₃ = 10 kOhm; R''₁₃ = 20 kOhm; R₂₆ = 100 kOhm.

R₂₆ can perform two functions. With the dimensioning above, an actual value is obtained which is proportional to k; in particular for k = 0 and U _is = 0. But you can also specify a setpoint via R₂₆, z. B. such that for k = 30% the output is 0 (R₂₉ = 25 kOhm). The actual value is no longer obtained, but the control deviation, which can influence the manipulated variable k and thus also the color as the controlled variable in a known manner via an actuator. So you can calibrate the scale of R₂₆ in target values.

The characteristic, i.e. H. the dependence of the color component signals on the respective physical size, is recorded, the physical Size - in the example the concentration - must be numerically defined. This measurement specifies the range in which the setpoint may lie. So cannot be regulated to a concentration k <50% in the example, since the characteristics of the color component signals used are not available.

Claims (6)

1.Device for identifying colors by means of color component signals, with which the color type of a surface area of an object can be detected independently of lighting fluctuations, with an optical system which images the surface area on photoelectric receivers, with filters for obtaining the color component signals for red, green and blue interposed and the Photoelectric receivers are each followed by amplifiers ( 35 , 36 , 37 ) with an adjustable degree of amplification, the outputs of the amplifiers being connected on the one hand to evaluation circuits and on the other hand each with summing resistors, the output signal of the summing resistors being input to a device ( 39 ) for comparison with a a signal corresponding to a target value is placed and the output signal of this device ( 39 ) corresponding to the deviation from the target value for comparison to the control inputs for setting the degree of amplification in the sense the reduction of the deviation from the nominal value is set to zero or to a very small value, characterized in that arrangements ( 10 , 11 , 12 ) which can be switched on by means of switching elements ( 20 , 21 , 22 ) are each switched on to the photoelectric receivers ( 4 , 5 , 6 ) for multiplying the respective color component signal by an assigned coefficient and for optionally inverting at least one product of a color component signal and an assigned coefficient, which is the quotient of the color component signal and a parameter on which the color of the surface area depends, and that switchable arrangements ( 10 , 11 , 12 ) is followed by a summing circuit ( 25 a) into which a setpoint value of the parameter for generating a control deviation signal for regulating the parameter can be fed. 2. Device according to claim 1, characterized, that the parameter is the concentration of a dye that Filler of specified color is added.   3. Apparatus according to claim 1 or 2, characterized in that the switchable arrangements ( 10 , 11 , 12 ) each have a re-coupled amplifier ( 14 ) with an adjustable resistor ( 13 ) at the input, the output signal of which is optionally via an inversion of the Input signal switchable amplifier ( 17 ) is connected to the summing circuit ( 25 a). 4. Apparatus according to claim 1 or 2, characterized in that the switchable arrangement ( 10 , 11 , 12 ) has a differential amplifier ( 31 ), the input via a resistor ( 32 ) with the respective photoelectric receiver ( 4 , 5 , 6 ) is connected, the resistor ( 32 ) being connected on the input side to a potentiometer ( 34 ) connected to reference potential, the tap of which is connected to the second input of the differential amplifier ( 31 ). 5. Device according to one of claims 3 or 4, characterized in that the summing circuit ( 25 a) consists of resistors ( 23 , 24 , 25 , 26 ) connected to a common connection, one of which ( 26 ) adjustable and optionally to one positive or negative reference voltage can be switched on. 6. The device according to claim 3, characterized in that the adjustable resistor ( 13 ) from digitally graduated resistors can be assembled, which can be connected in parallel and / or in series with one another by coding switches.