EP0512448B1 - Device for reading pale colour marks in printing machines - Google Patents

Abstract

The device for reading a mark (15) printed on an element (10) moving across in front of a light source (20) comprises at least two parallel mark reading channels (32,34,40,50,60) generating as output an electrical pulse when the mark passes, each channel being sensitive to a particular colour. The device moreover comprises electronic means (70) selecting, among the electrical pulses generated by the channels, the pulse which is most representative of the mark. <IMAGE>

Description

The present invention relates, in a polychrome printing machine, to a device for reading marks of different colors printed by each printing station, these marks subsequently making it possible to determine the identification error of each color by relative to the color of the first station taken as reference.

In one of the known devices, such as that described in document DE-A-33 11 352, a diode emitting red light and a diode emitting green light are used in combination to identify a registration mark. The light emitted is sent to a prism and lens assembly through a semi-transparent mirror. A second semi-transparent mirror directs the red and green lights onto a focusing lens which produces two scanning rays on the surface of the band moving in the machine. When the registration marks pass under the scanning rays, the alternately red and green light is reflected through the set of lenses and is received by a photo-sensitive unit composed of a pair of photodetectors bridged with a pair of resistances. A differential amplifier connected diagonally collects a series of electrical signals more or less contrasted depending on the color of the light. A processing stage generates a pulse by accumulating the series in a capacitor.

Document JP-A-61 164 835 describes a device for reading color marks comprising four photoelectric units in parallel. Three of them are composed of a photo-diode behind a color filter while the fourth is not equipped with a color filter. The units are connected to a pre-amplifier, as well as to a single processing channel where the electrical signals from the first three units and the inverted electrical signal representative of the background from the fourth are added in a summing circuit, the pulse result being compared to a threshold value.
Other devices are known, such as for example that described in the document EP 0 094 027. These devices function to satisfaction as long as the yellow, blue or red marks are sufficiently contrasted to be recognized without ambiguity by the device. Some of these known devices may include a bundle of optical fibers carrying light on the printed element and conducting the reflected light to a reading photodiode which produces an electrical signal. In order to increase the contrast between the basic electrical signal corresponding to an area of the unprinted element and an electrical pulse caused by the passage of a mark, a filter, usually blue, is interposed between the optical fibers and the photodiode. .

However, as soon as the printing colors are pale, in particular in printing for packaging involving prepared inks of pastel yellow, cream or sky blue color, conventional devices can no longer safely detect each mark printed, which may prevent a tracking control system from working properly. Indeed, it would then be advisable to try a first filter, to pass a light colored mark to note the quality of the signal obtained and to repeat the test with one or more other filters in order to select the most suitable for reading all brands. However, the most important phase of starting up the printing machine is precisely the search for the initially unknown position of a mark which cannot therefore be validly carried out without a reading device immediately operating. These many necessary tests quickly become prohibitive if you want to perform with the same printing machine many different jobs.

The object of the present invention is a device detecting a printed mark whatever its color, its intensity and its contrast with respect to the background color of the plate element on which it is printed.

These goals are achieved by means of a device for reading a color mark printed on a printed element, in a plate or in a strip, passing in front of a light source, this device comprising at least two photo-sensitive units, sensitive in a strip of separate color frequencies, each generating an electrical signal when the mark passes in front of the light source, and an electrical signal processing channel to generate an electrical pulse as each light-sensitive unit is connected to its own parallel processing channel identical to the others, the device further comprising a comparator / selector circuit selecting, from the electrical pulses generated by the channels, the pulse most representative of the color mark.

• une unité photosensible générant un signal électrique, suivie
• d'un circuit d'amplification à gain automatique fixant à une valeur prédéterminée la tension de base correspondant à une zone de l'élément non imprimée,
• suivi d'un circuit redresseur,
• suivi d'un circuit convertisseur convertissant l'impulsion électrique à flancs obliques provoquée par le passage de la marque de couleur devant l'unité photosensible en une impulsion électrique à flancs raides, chaque flanc raide correspondant au départ de la montée ou de la descente du flanc oblique associé; et
• un circuit comparateur/sélecteur, comparant et sélectionnant, parmi les impulsions électriques issues des canaux de traitement, à une période donnée, celle apparaissant et disparaissant la première.

Advantageously, each processing channel comprises:

• a photosensitive unit generating an electrical signal, followed
• an automatic gain amplification circuit setting the base voltage corresponding to a zone of the unprinted element at a predetermined value,
• followed by a rectifier circuit,
• followed by a converter circuit converting the electrical pulse with oblique flanks caused by the passage of the color mark in front of the photosensitive unit into an electrical pulse with steep flanks, each steep flank corresponding to the start of the ascent or descent of the associated oblique flank; and
• a comparator / selector circuit, comparing and selecting, from the electrical pulses coming from the processing channels, at a given period, that appearing and disappearing first.

Thus, thanks to this device, the most contrasted electrical pulse is systematically retained regardless of the quality of the other pulses considered.

A subsidiary problem can however complicate to a certain extent the design of the comparator / selector circuit since a color mark printed on a white plate element causes a negative pulse compared to the basic signal, whereas a mark of highly reflective color, such as gold or silver, causes an inverted pulse, that is to say positive with respect to the basic signal. This problem is obviated in that each processing channel comprises, before the converter circuit, a rectifier circuit imposing on all the electrical pulses a variation in the same direction with respect to the basic voltage.

Usefully, the photosensitive unit includes a photodiode placed behind a color filter and connected to the input of a current / voltage converter converting the electrical signal from the photo-diode into voltage.

According to a preferred embodiment, the rectifier circuit comprises a first stage for measuring the background of the base voltage, followed by a stage for subtracting the background from the base voltage, followed by an effective rectification stage for the sole positive pulses into negative pulses, followed by a stage of addition of the pulses of all the pulses and followed by a stage of readjustment of the background of the basic voltage.

According to a preferred embodiment, the converter circuit comprises a first peak detection stage followed by a second stage of subtraction of the peak from the input signal of the threshold detected by the first peak detection stage, the difference being applied to a comparison stage toggling as soon as the difference exceeds a predetermined threshold, as well as first electronic means resetting and reversing the direction of detection of the peak detection stage and second electronic means reversing the polarity of the comparison threshold applied to the 'comparison stage after a first tilting thereof.

According to a preferred embodiment, the comparator / pulse selector circuit comprises a first "OR" gate receiving on each of its inputs one of the pulses and the output of which is connected to the clock input of a first seesaw; as well as as many secondary flip-flops as pulses to be analyzed received inversely on their clock input, all the inverted outputs being connected to the input of a second "AND" gate, the output of which is connected to the reset input of the first flip-flop, the reset input of each secondary flip-flop being connected to the output of the first flip-flop, a last control line from a microprocessor being connected to one of the inputs of the second "AND" gate.

According to an advantageous embodiment, the device comprises an analog / digital and digital / analog converter connected to a microprocessor to receive from the rectifier circuit the measurement of the base voltage and to apply on the one hand to the amplifier circuit with automatic gain, when present, an electrical signal representative of the gain to be applied, and on the other hand to the converter circuit an electrical signal representative of the optimum threshold for the comparison stage. Thanks to this latter device, the voltage is permanently maintained at a value of approximately 8 volts and the comparator detection threshold is fixed at a value between 200 and 400 millivolts above the average noise present on the base voltage. .

• la figure 1 est un diagramme schématique du dispositif selon l'invention,
• la figure 1a est une vue partielle d'une exécution particulière du dispositif selon l'invention,
• la figure 2 est un plan du circuit redresseur compris dans le dispositif de la figure 1,
• la figure 3 est un plan du circuit convertisseur lnclus dans le dispositif de la figure 1,
• la figure 4 est un diagramme de l'opération effectuée par le circuit convertisseur de la figure 3, et
• la figure 5 est un plan du circuit comparateur/sélecteur compris dans le dispositif selon la figure 1.

The invention will be better understood from the study of an embodiment taken by way of nonlimiting example and described by the following figures:

• FIG. 1 is a schematic diagram of the device according to the invention,
• FIG. 1a is a partial view of a particular embodiment of the device according to the invention,
• FIG. 2 is a plan of the rectifier circuit included in the device of FIG. 1,
• FIG. 3 is a plan of the converter circuit lnclus in the device of FIG. 1,
• FIG. 4 is a diagram of the operation carried out by the converter circuit of FIG. 3, and
• FIG. 5 is a plan of the comparator / selector circuit included in the device according to FIG. 1.

As illustrated in FIG. 1, the device according to the invention comprises a bundle of optical fibers 25 first conveying the light coming from a light source 20 above the printed element 10 bearing colored marks 15 printed on the upper side. These printed elements 10 can be strips of paper or sheets of cardboard during manufacture. These color marks 15 are printed by each printing station at a non-binding location on the printed element 10, and this in the color of the station. The passage of these colored marks 15 in front of the optical fiber bundle 25 temporarily modifies more or less strongly the reflected light which is routed, after splitting the optical fibers of this beam 25, towards two separate photodiodes 32,33.

According to the invention, the separate photodiodes 32,33 are respectively made sensitive to distinct colors by means of filters 30,31 interposed between the output of the optical fibers and the photodiodes. For example, the filter 30 can be a dark purple filter favoring the yellow marks, while the filter 31 is green promoting the blue marks. The electrical signals from each photodiode 32,33 are first conditioned separately and in parallel by identical processing channels and constituted by members 34,40.50 and 60, respectively 35,41,50a and 60a, before be compared for selection by a comparator / selector circuit 70.

These identical and parallel processing channels each firstly comprise a current / voltage converter 34.35 transforming the variation in intensity within the photodiode caused by the passage of the color mark 15 in front of the fiber. optic in a voltage variation. As symbolized, this current / voltage converter 34.35 is produced in a known manner by means of an operational amplifier and a negative feedback between its negative output and input. Jumpers symbolized at the output allow to implement a first or a second feedback circuit modifying in a ratio of 1 to 10 the gain of the amplification of this stage. This voltage signal is then amplified by an automatic gain amplifier circuit 40,41 so that the basic signal corresponding to an unprinted area of the element 10 is fixed at a value of the order of 8 volts. In fact, depending on the background color of the printed element 10, the length of the bundle of optical fibers 25, any dust that can alter the entry or exit of the fibers as well as the filters, the base voltage collected at the output of the current / voltage converter 34.35 can vary between 150 millivolts and 8 volts.

The electrical signal then passes through a rectifier circuit 50, respectively 50a, which has the function of bringing all the pulses caused by a color mark 15 in the same direction, in this case negative, with respect to the basic voltage. Indeed, in the majority of cases, the color marks 15 are printed with colors darker than the background color and thus cause a reduction in the light reflected in the optical fibers, ie an instantaneous reduction in the current passing through the photodiode 32 , respectively 33 therefore a pulse with a value lower than the base voltage value. Conversely, if the marks 15 appear lighter than the background color, or even are printed with particularly colors reflective such as gold or silver, the reflected light is momentarily greater than the basic light and the same is true for the corresponding electrical pulse. This rectifier circuit 50, 50a makes it possible, by bringing all the pulses to the same side, to greatly simplify the subsequent comparator / selector circuit.

FIG. 1a represents a device similar to that of FIG. 1 in which the bundle of optical fibers 25 has not been split. A light diffusion member 25a has been arranged at the end of the bundle of optical fibers 25 so that the reflected light is routed indifferently on the filters 30 and 31. The construction of the other elements of the device comprising inter alia the photodiodes 32 and 33 as well as the current / voltage converters 34 and 35 remain unchanged compared to the arrangement described in FIG. 1.

With reference to FIG. 2, this rectifier circuit 50, 50a first comprises a bottom measurement stage 51 followed by a bottom subtraction stage 53 followed by the effective rectification stage 55 followed by a stage addition of the pulses 57 which finally finishes a stage of readjustment of the bottom 59.

As illustrated, the bottom measurement stage 51 essentially comprises the combination of a diode 513 and a capacitor 514, the other branch of which is connected to ground. Operational amplifiers 511 and 512 provide stage isolation. The switch 515 by temporarily short-circuiting the diode 513 makes it possible to periodically reset this bottom measurement.

The background subtraction stage 53 comprises in known manner an operational amplifier 533 receiving on the one hand the complete signal through the resistance 531 on its positive input and on the other hand the background to be subtracted through resistance 532 on its negative input.

In the effective rectification stage 55, only the positive pulses are amplified and inverted by the operational amplifier 553 comprising two diodes 551,552 in its feedback circuit. The addition by the operational amplifier 573 of the pulse addition stage 57, receiving on its negative terminal on the one hand the signal directly coming from the background subtraction stage 53 through the resistor 571 and d 'other hand amplified negative pulses counterbalancing the positive pulses, allows to obtain at the output of this effective rectification stage 55 a series of pulses of the same amplitude as initially but now all in the negative direction.

The operational amplifier 593 of the background readjunction stage 59 adds the background received directly from the first bottom measurement stage 51 through the resistor 591 and of the pulses received from the pulse addition stage 57 through resistor 592.

With reference to FIG. 1, the rectifier circuit 50, 50a is followed by a converter circuit 60, 60a of pulses with oblique flanks into pulses with steep flanks, pulses which are more apt to be processed logically thereafter.

Indeed, and as illustrated in FIG. 4, the pulses e1 and e2 generated by the photodiodes 32 or 33 have a first downward oblique slope corresponding to the gradual penetration of the mark in the reading field of the optical fibers of the optical fiber bundle 25, followed by a plateau during the passage of the body of the mark, and that ends a second rising oblique slope corresponding to the mark gradually leaving the reading field.

The detail of this pulse converter circuit 60, 60a will now be described in relation to FIG. 3 from which we can distinguish four important stages: a first peak detection stage 61 followed by a subtraction stage 62 of the peak measured at the instantaneous signal followed by a comparison stage 63 of the difference with a predetermined threshold coming from a stage 64. The result of the comparison is shaped by the operational amplifier 632 of which an inverted signal is generated by the inverter 633. The output of the shaping amplifier 632 is also used in feedback to on the one hand reverse the direction of detection of maximum of the stage of detection of the peak 61 and on the other hand modify the value threshold leaving stage 64.

The peak detection stage 61 essentially comprises a diode 614 (then 615) in relation to a capacitor 613, this device being isolated at the input by the conventional amplifier 611 and at the output by the operational amplifier 612. The direction of detection maximum, either up or down, is initially determined by the state of the relay 65 selecting the diode 614 or 615. This peak detection stage 61 is reset, after a short delay introduced by the inverter 633 by means diodes 616 or 617 as appropriate, using relay 644.

The peak subtraction stage 62 receives on the one hand the signal from the peak detection stage 61 through the resistor 621 and on the other hand, through the resistor 622, the instantaneous signal previously amplified d '' a gain of 1 per the operational amplifier 619. The comparison is carried out by the amplifier 631 receiving the threshold signal on its positive input and the difference signal on its negative input.

As can easily be understood with reference to FIGS. 3 and 4, the detection stage of peak 61 first acquires the value of the base voltage, and the output of the subtraction stage of peak 62 first presents a null signal which increases only with the appearance of the downward oblique slope of a pulse. When this oblique slope of this pulse has exceeded a predetermined threshold v1 with respect to the base voltage, the operational amplifier 631 switches over and there appears a first steep rise in voltage s11 at the output of the inverter 633. This rise in voltage s11 first causes the selection of the diode 615 allowing the capacitor 613 to be discharged through the diode 617 then the connection of the diode 616, after a delay determined by the inverter 633. The detection stage of the peak 61 is then ready to detect a new maximum but in the downward direction. This first voltage increase also caused in stage 64 a modification of the threshold voltage v2 by grounding the positive input gate of an operational amplifier.

The peak detection stage 61 then detects the value of the lower plateau of the input pulse e1 and the output of the subtraction stage of the peak 62 remains at zero during the entire duration of this lower plateau. Again, as soon as the beginning of the oblique flank of the ascent of the input pulse appears, the difference at the output of the subtraction stage of the peak 62 increases until it exceeds the new threshold v2 of the comparator 631 which s reverse thus causing a sudden descent s12 of the shaping amplifier 632.

Thus, the steep flank rising from the output pulse s1 substantially corresponds to the start of the downward oblique flank of the input pulse e1, and the steep flank falling from the output pulse s1 substantially corresponds to the start of the backward oblique flank of the input signal e1.

With reference to FIGS. 1 and 4, the pulses s1 and s2 now in slots, originating respectively from the channel corresponding to the yellow color and from the channel corresponding to the blue color, are applied to a comparator / selector circuit 70 for selection retaining the pulse s1 rising which will descend first and which will correspond to the most oblique initial oblique impulse e1.

" de ces bascules secondaires sont branchées à l'entrée d'une seconde porte "ET" 75 dont la sortie est reliée à l'entrée de réinitialisation "CL" de la première bascule 72. Par ailleurs, la sortie "Q" de cette première bascule 72 est reliée également à l'entrée de réinitialisation "CL" de chacune des bascules secondaires 73,74. Enfin, une dernière ligne d'autorisation ou de blocage 85 du circuit comparateur/sélecteur 70 est branchée à l'une des entrées de la porte "ET" 75.The embodiment of the comparator / selector circuit 70, as illustrated in FIG. 5, firstly comprises a first "OR" gate 71 receiving on each of its inputs one of the pulses in slots, and the output of which is connected to the clock input "CLK" of a first flip-flop 72. The comparator / selector circuit 70 also includes as many secondary flip-flops 73,74 as pulses to be analyzed, these pulses being received in reverse manner respectively on their clock input "CLK". All outputs reversed " $\overline{\text{Q}}$

"of these secondary flip-flops are connected to the input of a second gate" AND "75 whose output is connected to the reset input" CL "of the first flip-flop 72. Furthermore, the output" Q "of this first flip-flop 72 is also connected to the reset input "CL" of each of the secondary flip-flops 73, 74. Finally, a last authorization or blocking line 85 of the comparator / selector circuit 70 is connected to one of the inputs from door "AND" 75.

" correspondante à l'état bas, créant ainsi le front descendant de l'impulsion de sortie. Cet état bas sur la sortie "Q" de la bascule 72 a également comme conséquence de réinitialiser toutes les bascules secondaires 73,74 remettant toutes les sorties inversées " $\overline{\text{Q}}$

" à l'état haut et bloquant par là-même ces bascules, rendant inopérante la montée du signal inversé suivant. La porte "ET" 75 retourne à l'état haut ce qui libère à nouveau la bascule 72 prête pour une sélection suivante tant qu'une autorisation sur la ligne 85 est maintenue.In the initial state of the device, all the inputs of the "AND" gate 75 are in the high state which released the first flip-flop 72 whose output "Q" is initially low implying the blocking of flip-flops 73 and 74 . On the arrival of a pulse on one of the inputs of the door "OR" 71, the output of this door is found high which results in the appearance of a high state on the exit door " Q "of flip-flop 72 constituting the rising edge of the output pulse and, also, unlocking flip-flops 73 and 74. The arrival of the rising edge of the second pulse then has no effect on the comparator / selector circuit 70. On the other hand, the arrival of the first rising edge of an inverted signal, corresponding in fact to the falling edge of this first pulse, changes the state of the corresponding flip-flop 73,74 which has the effect of immediately lowering the corresponding "Q" door. The "AND" door 75 sees at least one of its entry doors set low, and its output also lowers, which resets the first flip-flop 72, therefore resets the door " $\overline{\text{Q}}$

"corresponding to the low state, thus creating the falling edge of the output pulse. This low state on the output" Q "of the flip-flop 72 also has the consequence of resetting all the secondary flip-flops 73,74 resetting all the outputs reversed " $\overline{\text{Q}}$

"in the high state and thereby blocking these flip-flops, making the rise of the next inverted signal inoperative. The" AND "gate 75 returns to the high state which again releases flip-flop 72 ready for a next selection as long that an authorization on line 85 is maintained.

Referring to Figure 1, the device according to the invention further comprises an analog / digital and digital / analog converter 90 in relationship with a microprocessor 80, this device being able to receive on line 81 a measurement of the value of the base voltage in order to return on lines 82 an electrical signal corresponding to the gains to be applied to the automatic gain amplification circuits 40 and 41 , and on the other hand on lines 83 a threshold value for the comparison stage 63 of stages 61 and 64, threshold fixed between 100 and 400 millivolts above the background noise measured on the basic signal. The microprocessor 80 also sends on line 85 a control signal blocking the comparator / selector circuit when no mark is expected.

As we have seen on reading this presentation, this device according to the invention inevitably detects a mark passing in front of the light flux coming from the source 20 by making an instant selection of the best reading channel, for yellow or for blue, taking into account the color, contrast and intensity of the brand sought. For machines that have to carry out delicate work, it is quite possible to add a third or fourth parallel reading channel for other very distinct colors.

Claims (8)

1. Device for scanning a colour mark (15) printed on a printed plate or web-shaped workpiece (10) travelling under a light source (20) within a printing machine, this device including at least two photosensitive units (30, 32, 34; 31, 33, 35), responsive to a colour frequency range and generating each an electric signal when the mark travels through the area of the light source (20), and a processing channel of said electric signal, in order to generate an electric impulse at the output, characterized by the fact that each photosensitive unit is connected to its own processing channel (40, 50, 60; 41, 50a, 60a) parallel to and identical with the others, and by the fact that the device includes a comparating/selecting circuit (70) for selecting the most representative colour mark (15) impulse among the electric impulses generated by the channels.
2. Device according to claim 1, characterized by the fact that each processing channel includes:

   - a photosensitive unit generating an electric signal, followed by

   - an amplifying circuit with automatic gain (40, 41), fixing at a predetermined rate the basic voltage corresponding to a non-printed area of the workpiece (10),

   - followed by a rectifying circuit (50, 50a)

   - followed by a converting circuit (60, 60a) converting the oblique sloped electric impulse caused by the colour mark (15) travelling under the photosensitive unit into a steep sloped electric impulse, every steep slope corresponding to the beginning of the ascent or descent of the associated oblique slope; and

   - a comparating/selecting circuit (70) comparating and selecting among the electric impulses originating from the processing channels at a given moment the impulse appearing or disappearing first.
3. Device according to claim 2, characterized by the fact that every processing channel includes before the converting circuit (60, 60a), a rectifying circuit (50, 50a) imparting to all electric impulses a variation in one direction with regard to the basic voltage.
4. Device according to one of the aforementioned claims, characterized by the fact that the photosensitive unit includes a photodiode (32, 33) situated behind a colour filter (30, 31) and connected to the input of a current/voltage converter (34, 35) converting the electric signal originating from the photodiode under voltage.
5. Device according to claim 3, characterized by the fact that the rectifying circuit (50, 50a) includes a first stage (51) for mesuring the deep value of the basic voltage, the said stage being followed by a stage (53) for subtracting the deep value of the basic voltage, followed by an effective stage (55) for rectifying solely the positive impulses into negative ones, followed by a stage (57) for adding all the impulses and followed by a stage (59) ensuring the re-addition of the deep value of the basic voltage.
6. Device according to one of the aforementioned claims, characterized by the fact that the converting circuit (60, 60a) includes a first peak detecting stage (61), followed by a second stage (62) for substracting the peak from the input signal of the threshold detected by the first peak detecting stage (61), the difference being applied to a comparating stage (63) tilting off as soon as the difference exceeds a predetermined threshold, as well as primary electronic means (65, 644) re-initializing and inverting the detection direction of the peak detecting stage (61), and secondary electronic means (633) inverting the polarity of the comparative threshold applied to the comparating stage (63) after a first tilting of the latter.
7. Device according to one of the aforementioned claims, characterized by the fact that the comparating/selecting circuit (70) for impulse selection includes a first gate "OR" (71) receiving one of the impulses at each of its inputs, and with its output connected to the clock input (CLK) of a first tilting device (72); and as many secondary tilting devices (73, 74) as there are impulses to be analysed, the said impulses being received inverted at their clock input (CLK), all inverted outputs (Q) being connected to the input of a second gate "AND" (75) the output of which is connected to the re-initialization input (CL) of the first tilting device (72), the re-initialization input (CL) of every secondary tilting device (73, 74) being connected to the output (Q) of the first tilting device (72), and a final monitoring line originating from a monitoring microprocessor being connected to the one of the inputs of the second gate "AND" (75).
8. Device according to claims 2, 5, 6 and 7, characterized by the fact that it includes an analogous/digital and digital/analogous converter (90) connected to a microprocessor (80) destined to receive from the rectifying circuit (50, 50a) the basic voltage rate and to feed, on the one hand, the amplifying circuit (40, 41) with automatic gain, if present, with an electric signal representative for the gain to be applied and, on the other hand, the converting circuit (60, 60a) with an electric signal representative for the threshold optimal for the comparating stage (63).

Images